ML20072M138
| ML20072M138 | |
| Person / Time | |
|---|---|
| Site: | Nine Mile Point  |
| Issue date: | 07/08/1983 |
| From: | Delwiche D, Gordon G NIAGARA MOHAWK POWER CORP. |
| To: | |
| Shared Package | |
| ML18041A108 | List: |
| References | |
| NUDOCS 8307140350 | |
| Download: ML20072M138 (363) | |
Text
{{#Wiki_filter:.. . _-_ _ .__ NINE MILE POINT 1 ; PIPE CRACK TASK FORCE REPORT SPONSORED BY NIAGARA MOHAWK POWER CORPORATION SYRACUSE, N.Y. I "8R72n a=L P PDR A
This report is made available by Niagara
'lobawk without consideration in tne inte>'est of promoting technical knowledge. Neither
, ':iagara Mchawk nor the authors nahe any warranties, express or implied, with regard to the use or accuracy of the information disclosed in this report, or that the use of such information may not infringe privately coned rights. Niagara Mohawk assumes no responsioility for liability or damages which ray result f rom the use of any information contained in this report. t e e _ - - - - - - - - - - _ - _ _-
l
}
NINE MILE POINT I PIPE CRACK TASK FORCE REPORT Compiled by D. E. Delwiche i Approved: h2 , G. M. Gordon, Chairman ' Pipe Crack Task Force E \ .. . ..
CONTRIBUTORS TO NINE MILE POINT I PIPE CRACK TASK FORCE REPORT TASK FORCE MEMBERS l G.M. Gordon, Chairman l (General Electric Company) R. Oleck (Niagara Mohawk Power Corporation) J. Danko (Electric Power Research Institute) K. Schmidt (Nuclear Energy Services) R. Hookway (Teledyne Engineering Services) SUPPORTI!iG CONTRIBUTORS Niagara Mohawk Power Co. F. Hawksley Electric Power Research Institute M. Behravesch (NDE Center) L. Becker (flDE Center) S. Kazuoka G. Dau R. Stone (NDE Center) W. Childs Nutech R. Cargill Battelle Columbus Laboratories M. Landow V. Pasupathi i
SUPPORTING CONTRIBUTORS (Cont'd) Nuclear Energy Services P. Barry
- 11. Stamm Teledyne Engineering Services R. Pace General Electric Co. - Nuclear Energy Business Operation
- 11. Bensch R. Horn J. Sundberg B. Gordon B. Rajala J. Cutt D. Delwiche R. Carnahan S. Ranganath K. Ramp Acknowledgements Sincere appreciation Is expressed to the following individuals and companies for tne use of their data in the preparation of Section X: 1. Hamada, Y. Mori, F.
Hataya and S. Hattori of Hitachi Research Laboratory, M. Akashi of Ishikawajima - Harima Heavy Industries Co., Ltd. and M. Hishida of loshiba Corporation.
l t TABLE OF CONTENTS Page MAJOR CONCLUSIONS......................................................... 1 E X E C UT I V E S UMMAR Y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . S - 1 A. Extent of Cracking.......................................... S-2 B. Metallographic Examination.................................. S-3 C. Stresses.................................................... S-4 D. Materials................................................... S-5 E. E n v i r o nm e n t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . S - 7 F. Ul tras oni c Exami nation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . S-7 Furnace Sensi ti zed Safe-End Procedures. . . . . . . . . . . . . . . . . . . 5-8 P i p i n g P ro ce d u re s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . S -8 Crack Growth /U.T. Detectabili ty. . . . . . . . . . . . . . . . . . . . . . . . . . . S-10 G. C a u se of Crac ki n g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . S-12 SECTION I. INTRODUCTION................................ .............. 1-1 SEC1IGN II. EXTENT AND DISTRIBUTION OF CRACKING. . . . . . . . . . . . . . . . . . . . . . . . 2-1 E xw:.i n a t i on R e s ul t s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2- 1 Correlation of U.T. and PT Resul ts. . . . . . . . . . . . . . . . . . . . . . . . . 2-2 SECTION III. LA50RATORY CHARACTERIZATION OF CRAC KS. . . . . . . . . . . . . . . . . . . . . . 3-1 Introduction............................................... 3-1 Observations and Concl usions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1 Field Examination............. ............................ 3-2 Metall ographic Exami nations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-6 A. B o a t S am p l e " A" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-6 B. B oa t S am pl e " B" . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3- 8 C. Sample 3 - Piping Sections........................... 3-10 D. Boa t Sampl e s "C" , "D" , " E" , and " F" . . . . . . . . . . . . . . . . . . 3-12 E. Section of Elbow-to Pipe Weld. . . . . . . . . . . . . . . . . . . . . . . . 3-14 Composi tion - Chemistry and Structure. . . . . . . . . . . . . . . . . . . . . . 3-16 Sensitization Measurements................................. 3-16 A. ASTM A- 26 2 P ra c ti ce A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-16 B. Electrochemical Potentiokinetic Reacti va ti on (EPR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-16 Hardness................................................... 3-17 Ferrite Survey............................................. 3-17
TABLE OF CONTENTS (Continued) Page SECTION IV. STRESSES AND STRESS RULE INDEX............................. 4-1 Background................................................. 4-1 Weight During Normal 0peration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2 The rmal Exp an si on . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4- 3 Opera ti onal Concerns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-4 A. Snubbers............................................. 4-4 B. P ump V i b ra ti on . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-5 C. Sway Braces.......................................... 4-5 Stress Summary............................................. 4-6 Residual Stresses.......................................... 4-6 Stress Ruel Index.......................................... 4-7 References................................................. 4-12 SECTION V. MATERI ALS AND FABRICATION HIST 0 RIES. . . . . . . . . . . . . . . . . . . . . . . . 5-1 Reci rcul ation System Confi guration. . . . . . . . . . . . . . . . . . . . . . . . . 5-1 Material and Fabricatien................................... 5-2 A. Pipe................................................. 5-2 B. Elbow and Tees....................................... 5-3 C. Safe-Ends............................................ 5-4 Welding.................................................... 5-4 SECTION VI. WATER CHEMISTRY AS P0TENTI AL CONTRIBUTING FACTOR. . . . . . . . . . . 6-1 Evaluation of General Electric Water Chemistry Data Base of NMP-1......................................... 6-1 A. Reactor Water Conductivity........................... 6-3 B. Reactor Water Chloride............................... 6-4 C. Hydrogen Ion Concentration........................... 6-4 D. Reactor Water Sili ca. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-5 E. Feedwater Conductivities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-5 F. Feedwater Dissolved 0xygen........................... 6-6 G. Metall i c Impuri ti e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-6 Evaluation of Water Chemi stry Transients. . . . . . . . . . . . . . . . . . . 6-6 S u nm a ry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 - 8 I
TABLE OF CONTENTS (Continued) Page SECTION VII. EFFECTS OF DECONTAMINATION................................. 7-1 Introduction............................................... 7-1 Decontamination and Stress Corrosion Cracking. . . . . . . . . . . . . . 7-2 A. Laboratory Studies................................... 7-2 B. Operating Experiences................................ 7-6 Discussion................................................. 7-7 Conclusions................................................ 7-8 References................................................. 7-9 SECTION VIII. ULTRASONIC EVALUATION...................................... 8-1 Introduction and Background................................ 8-1 A. Seq uen ce o f Eve nts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1 B. Principle Aspects of the 1982 U.T. Examination at Xine Mile Point 1..................... 8-3 Recirculation System ISi E> amination Resul ts. . . . . . . . . . . . . . . 8-5 ISI Inspection Comparisons 1979-1982....................... 8-6 A. Safe-Ends............................................ 8-7 B. B11ance of Recirculation Pipe System. . . . . . . . . . . . . . . . 8-8 Evaluation of Ultrasonic Examination Performance' Following Pipa Renoval...................................... E-10 A. Correlation of U.T. and PT Data...................... 8-11 B. Search Uni t and Procedure Correlation. . . . . . . . . . . . . . . . 8-11 C. Effects of Chemical Decontamination on Detection..... 8-12 D. In-Situ vs. Laboratory U.T. Examination.............. 8-15 l E. Crack Depth I nforma ti on. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-16 , APPENDIX A - Procedure Excerpts............................ 8-31 l APPENDIX B - Transducer Usage for Safe End Examination..... 8-36 SECTION IX. APPARENT CRACK GROWTH RATES - GENERIC IMPLICATIONS......... 9-1 A. Ma te ri al s Da ta Ba s e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1 B. Crack Growth Rates of Nine Mile Point Material....... 9-2 C. Model Qualifications and Piedictions................. 9-4
- 1. Ci rcumferential Fl aw. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-5
- 2. Axial Crack...................................... 9-8
l TABLE OF CONTENTS (Continued) Page D. U.T. Detectability vs. Geometry vs. Crack Growth Rate......................................... 9-10 E. References.......................................... 9-14 SECTION X. STRESS CORROSION PERFORMANCE OF TYPE 316 NUCLEAR GRADE S/S.................................................. 10-1 Introduction............................................... 10-1 T he oreti cal An aly s i s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-2 A. T h e rmo dy nami c s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 -3 10 - 5 B. Ki n e t i c s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensitizati on and Corrosion Test Results. . . . . . . . . . . . . . . . . . . 10-7 A. Crack I ni ti ati on . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-7 B. Crack Propa gation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-13 Discussion................................................. 10-14 Conclusions........................... .................... 10-15 References................................................. 10-18 SECTION XI. EVALUATION OF RECIRCULATION SYSTEM REPAIRS AND REPLACEMENT............................................ 11-1 Safe End Replaccment....................................... 11-1 P i pe Repl aceme nt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-2 Expected Performance of Repairs / Replacement................ 11-2 APPENDIX A. Mill Test Certificates and Test Reports. . .... . A-1 6
MAJOR CONCLUSIONS
- 1. Cracking at Nine Mile Point-1 (NMP-1) was due to weld Heat Affected Zone (HAZ) Intergranular Stress Corrosion Cracking (IGSCC) with initia-tion on inner surface.
- 2. The extent of IGSCC cracking at NMP-1 was somewhat greater than that observed at most other plants. 85% of recirculation piping welds have indications of cracking.
- 3. Shop and field welds were equally affected, indicating welding fabrica-tion was not a significant variable.
- 4. The base material was normal (e.g., not sensitized) but weld heat affected zones (HAZ) were lightly sensitized Type 316 S/S.
- 5. The leaking safe-end cracks were all axially oriented, potentially making ultrasonic examination (U.T.) diffic. ult.
- 6. Cracking in piping welds was not detected prior to 1982 due to limited sampling and the U.T. technique employed for the assumed non-service sensitive system.
- 7. The U.T. techniques employed in 1982 were highly effective.
- 8. A high percentage of welds have Stress Rule Indices (SRI) >1.2. This was perhaps due to the larger percentage of welds to fittings which generally experience higher stresses. Also the pressure stress contribu-tion to the SRI may have been slightly higher.
- 9. Plant steady state water chemistry was equivalent to the average U.S.
BWR plant.
- 10. Higher than average Cl- and conductivity transients occurred in '71 and
'79 but it is unlikely those ontributed to crack initiation.
1
~ --- - _ _ - _ _ _ _ _ _ _ _
- 11. Aside from a-high incidence of SRI >1.2, none of the contributing factors (stress, environment, sensitized material) were excessive enough to expect the degree of IGSCC which occurred. This preponderance of cracking, even for this old a plant, may be an upper bound in the IGSCC data base.
- 12. Pipe decontamination did not aggravate cracking but appears to have made cracking more visible to ultrasonic examination.
- 13. Even for those welds with the maximum extent of cracking, full structural margin existed, and would have been maintained for at least an additional 18 months.
' - NMP-1 CRACKING FULLY CONSISTENT WITH OUR UNDERSTANDING
- THEREFORE, TYPE 316 NUCLEAR GRADE REPLACEMENT MATERIAL IS FULi.Y ADEQUATE FOR PLANT DESIGN LIFE.
l ii
EXECUTIVE
SUMMARY
On March 23, 1982, during a hydrotest at Nine Mile Point-1 (NMP-1) prior to startup, inspection revealed through wall leaking cracks in two of the ten furnace sensitized recirculation safe ends. On March 26, 1982, ultrasonic examinations on these two safe ends and one other confirmed crack-like indi-cations. In mid-April,1982, two boat samples were obtained from one of the safe ends in the vicinity of the through wall cracks. One each of these samples was sent to General Electric and Battelle Laboratories for evaluation. The results of those evaluations in mid-May confirmed the presence of inter-granular stress corrosion cracking (IGSCC). Subsequent ultrasonic examination of the 28" recirculation piping revealed crack-like indications in many of the piping weld heat affected zones examined at both shop and field welds. Following the discovery of extensive IGSCC at NMP-1, Niagara fiobawk Pcwer Corporation (HMPC) decided to replace the entire 28-inch recirculation system pipir;g censisting of five loops including safe ends with Type 316 Nuclear Grade material . To provice added assurance that the 316 Nuclear Grada replacement material will not suffer similar cracking due to some unique l plant specific causa, NMPC appointed an interdisciplinary Task Force on September 29, 1982. The stated objectives of the Task Force were:
- 1. Investigate the cause of cracking in the recirculation system piping and safe ends.
- 2. Determine the reasons why cracking was not seen before 1982.
- 3. Evaluate the adequacy of ultrasonic testing techniques.
- 4. Evaluate system pipe stresses, and stress rule index values.
- 5. Examine other relevant issues which may bear on generic concerns.
- 6. Evaluate the adequacy of replacement material.
The Task Force has held four meetings to fully review the NMP-1 IGSCC situation and to establish the necessary inputs needed to allow accomplishment of the stated objectives. This report documents results of relevant investi-gations, findings, and assessments made by the Task Force. i S-1
A. Extent of Cracking To date, 62 of ti ? 76 welds in the five recirculation loops have been examined using either t.ltrasonic (U.T.) and/or dye penetrant (PT) examination procedures. Of these, o3 (85%) were found to have indications of cracking associated with a heat affected zone (HAZ). Shop welds and field welds were approximately equally affected. Because each weld has two heat affected zones (HAZ) adjacent to it, it is perhaps more appropriate to consider the frequency of HAZ cracking. Con-sidering a total of 104 HAZ's examined, 75 (72%) had crack indications, with approximately equal percentages of shop and field HAZ's cracked. Figure S-1 is a layout of tne cracked and uncracked weld locations in each loop. In a number of the welds with U.T. and/or PT indi-cations, the extent of cracking was found to be quite extensive with the cracking-covering a signifi-cdnt fraction of the circumference. An example of such an extensively cracked weld is shown in Figure S-2. It can be seen from the map of PT indications, the cracking on the elbow side cf the weld covers about 130' of circumference. The measured depth of 0.2 to 0.45" based on sensitive U.T. examination from the cut edge of the pipe are also shown. The region of maximum crack depth was further evaluated by progressive grinding and crack depth was found to be about 0.5 inches, about 50% of wall. A structural margin evaluation was per-formed for the most extensively cracked weld characterized and it was verified that adequate margin existed for all cases. Further, the plant could have continued to operate for another 12-18 months without dropping below code allowable structural margins. Significant additional in-service-ultrasonic inspection on several older BWR plants has been performed subsequent to the NMP-1 examination in response to the IE Bulletin 82-03. The results to date are summarized in Figure S-3. As may be seen, for most of the plants evaluated, less than 10% of the welds inspected had reportable indications. However, in the cases of NMP-1, Hatch 2, Browns Ferry-1, and Vermont Yankee, a significant fraction of the welds did have reportable indications. I S-2
B. Metallographic Examination Boat samples and piping segments containing cracks were removed from the recirculation piping system of NMP-1 and sent to laboratories for metallo-graphic evaluation and failure analysis. The boat samples and piping segments contained representative examples of axially oriented and circumferentially oriented cracks removed from furnace sensitized RPV safe ends as well as from shop and field welds of the recirculation loops. Optical metallographic and scanning electron fractographic examinations have resulted in the following conclusions: o The through wall safe end cracking was due to IGSCC of a furnace
- ensitized high carbon Type 316 stainless steel.
o Although trace elements were found on the fracture faces of one of the boat samples, the origin of the containments could not be deter- , mined from the examinations performed, nor can the presence of the containments be associated with the presence of cracking. o The througn wall leakage was, in each case examined, due to axially oriented cracks, o All shop and field weld heat affected zone (HAZ) pipe cracking was intergranular stress corrosion cracking, with inner surface initiation. o Electrochemical potentiokinetic reactivation (EPR) measurements and other sensitization studies showed that pipe and elbow base metal was nonsensitized (annealed) material, while the weld heat affected zones (HAZ's) were found to be lightly sensitized in a typically narrow zone. Cracking was associated with the weld HAZ's. o Some crack propagation into weld metal had occurred; however, ) arrest would occur as the crack propagated towards weld regions of higher ferrite content. i S-3
- s. . ..- .
i ) C. Stresses A detailed investigation of the NMP-1 recirculation system stresses was made to determine whether the extensive cracking that took place was attribu-table to unduly high stress levels. As part of this evaluation, investigations were made to assure that there were no outside influences on the stress in the system that would contribute to an increased frequency of cracks. The details are presented in Section IV of this report. The conclusions are:
- 1. A'.1 weight, thermal and vibratory stresses are low for both normal and abnormal operating modes and in no way contribute to cracking here more than in any otner plant.
- 2. The pressure stress contribution to the SRI is somewhat higher than that for more recent BWR recirculation loop designs. The pipe wall thickness in more recent BWR is higher, mainly to provide greater seismic margins.
Nevertheless, the pressure stresses are still within the appropriate design code allowables. The Stress Rule Index (SRI) was computed by Teledyne Engineering Services for a typical NMP-1 recirculation loop (loop 15). The values are presented in Table S-1 for reference. These values were based on stresses developed by Te . dyne in the re-analysis for this task forced in which the minimum thickness for stress computation was used instead of nominal values. The SRI values calculat.ed here by TES compare closely with those values calculated by General Electric and presented to the NRC in October 1982. The differ-ences which do exist are due primarily to the treatment of a branched (Sweep-o-let) type of connection. The number used by TES for this type of connection was higher (2.1 vs.1.7) than that used by GE, reflecting a greater level of conservatism in the TES results. For the NMP-1 case, the main contributors to the SRI are pressure stress and weld residual stress. As can be seen, the SRI values exceed the field threshold cracking value of 1.1 for all welds evaluated. This is unusual and may be S-4
Table S-1 SRI Values for NMP Loop 15 (Developed by Teledyne Engineering Services) Node Location As Desianed SRI 101 Elbow 1.2 105 Tee 2.2 115 Tee 2.1 135 Valve 1.2 155 Elbow 1.2 156 Pump 1.3 200 Pump 1.2 205 Elbow 1.2 225 Valve 1.2 270 Elbow 1.4 due to somewhat higher pressure stress plus the absence of a significant num-bar of pipe / pipe butt welds in the NMP recirculation system. Pipe / pipe butt welds have lower code stress intensification factors and are generally below the threshold SRI value in o;her BWR plants. D. Materials The five 28-inch recirculation loops were constructed of Type 316 stain-less steel pipes, elbows, and fittings. The straight sections of pipe were fabricated by the National Annealing Box Company of rolled and welded plate. The elbows and reducing tees were fabricated by the Crane Company of wrought plate. All wrought pipes, elbows, and reducing tees were solution annealed and water quenched. Shop welded subassemblies were fabricated by the Grinnell Company. Shop welds were not given a subsequent heat treatment. l The National Annealing Box spool pieces were formed from four heats of material with carbon contents ranging from 0.042% to 0.055%. The elbows and tees were fabricated from 19 heats of material with carbon contents ranging from 0.050 to 0.075%. In most cases, the heats from which the specific piping segments were fabricated is not documented. S-5
r _ g The ten 28-inch recirculation safe ends were fabricated from the same } heat of 0.054% carbon Type 316 stainless steel forgings per ASTM A336. , g Although the safe end forgings were solution heat treated followin.g fabrica- 7-tion, the safe ends were subsequently post-weld heat treated with the reactor - pressure vessel resulting in a furnace sensitized condition. g h i ' The GE purchase specification for the recirculation system required all - k welded joints to be made by the inert-gas tungsten-arc process with internal 5 gas purging for at least the root pass and the second layer. Shielded metal ; y arc was allowed for the remainder of the weld. The weld filler metal was _ required to meet ASTM A298 or A371. The acceptance criteria also called for i
'E a 5% minimum ferrite, as measured on undiluted weld deposit, and a Cr to Ni ratio minimum of 1.9. j g 9 e
T k Estimates for typical heat input values during welding are 26 to 88 $ f h K. joules / inch for the shop welds, and 25 to 44 K. joules / inch for the field f h welds. j , g a [ The replacement piping for the NMP-1 recirculation system piping is $ fabricated from a Type 316 Nuclear Grade material. This material choice, with I F l controlled fabrication procedures, is anticipated to be fully resistant to IGSCC, and will preclude this BWR from any additional IGSCC concerns in the recircu-lation piping system. i Prior to the bulk of the in-service U.T. inspection performed in March 1982 at NMP-1, the five recirculation loops were decontaminated to reduce . personnel Man-Rem exposure. A review of the relevant laboratory data, and ] the operating experience of reactor systems decontaminated by the same pro-prietary CAN-DECON process indicates that decontamination, per se, had no impact on the observed occurrence of intergranular stress corrosion cracking (IGSCC) of the large diameter Type 316 stainless steel piping at Nine Mile Point Unit 1. The only contribution that the CAN-DECON cleaning process appears to have had on the piping IGSCC is that it increased the detectability , of the cracks during ultrasonic examination. A similar effect of decontamina-tion on U.T. crack detectability may have occurred at Vermont Yankee recently, ; where U.T. examination subsequent to decontamination resulted in 58% of the S-6
v: E f welds exhibiting reportable indications (up to 360* around the HAZ in several
- cases).
f
- - E. Environment
= { With respect to the NMP-1 coolant environment, the steady state long term conductivity values have been equivalent to that of a typical domestic 1 U.S. BWR, while the weekly maximum chloride levels have oeen somewhat higher than average although still well within specification. In addition to these = steady state values, twc significant off-chemistry transients occurred and these were evaluated. As a result it is concluded that there is no obvious y basis for believing that water chemistry played a significant role in accel-1 erating IGSCC at NMP-1. E 4 F. Evaluation of Ultrasonic Effectiveness { s Following the discovery of the visually detected leakage of two of the [ ten furnace-sensitized recirculation system safe ends, U.T. exeminations of j the two affected safe ends, and one other safe end, as well as a significant i portion of the balance of the recirculation piping system performed. As
- described previously, these added examinations revealed indications in the
, heat affected zones of recirculation system welds at the inner surface. A [ large number of indications, in the five recirculation loops were identified, and by subsequent penetrant examination and metallographic laboratory examina .
- tions were confirmed to be cracks caused by a stress corrosion mechanism.
Since the plant contained extensive IGSCC, including leaking axial safe end cracks in 1982, the question arises as to why no reportable indications were found by U.T. in earlier examinations including the one nine-months earlier in 1981. To evaluate this question, a detailed assessment was per-formed for the U.T. procedures employed over the 1979-82 period for the furnace sensitized safe ends and also for the balance of the recirculation piping welds which were examined by a less sensitive procedure prior to 1982.' S-7
FURNACE SENSITIZED SAFE END PROCEDURES A review of the ISI records of the 1979-82 period compares the inspection parameters of the safe ends in loops 11-15. In reviewing these and other supporting records, the following conclusions have been reached relative to the comparison of the 1981 and 1982 NMP-1 safe end inspections.
- 1. A higher sensitivity was used for the scanning in 1982.
In 1982, more gain was added to the calibration sensitivity for scanning than in 1981 (typically 10dB versus 6dB).
- 2. The time devoted to some of the safe end examinations in 1981 appears to be too short for optimum detection of IGSCC.
- 3. The 1981 inspection took place after the presence of IGSCC had been confirmed by leakage, cre6 ting a psychology of inspection contributing to more careful examination and more willingness to call cracks.
- 4. The through wall leaking safe end cracks are axial and thus were probably somewhat shielded from effective U.T. examination by the unground weld crowns.
PIPING PROCEDURES A similar review of the piping inspections derived from the ISI records of 1981 and 1982 for the common joints inspected in both years was also performed. The following conclusions are considered to relate to the difference between the 1981 and 1982 inspection results on the balance of the recircu-lation system, and may explain why the cracking condition was not detected in earlier inspections: 1 i S-8
- 1. Only two joints were inspected during the 1981 ISI, namely 32-FW-10-W and 32-FW-36-W. Comparison of ISI results in 1981 and 1982 is hence limited to these two welds.
- 2. The procedure used in 1981 (a procedure acceptable to Section XI, Appendix III) is ineffective for detection of IGSCC because of the 50%
DAC reporting level. Indications were found in both joints in 1982 with amplitudes less than 50% DAC (10% notch) using 1/2" diameter 1.5 MHz transducers.
- 3. The transducers used in 1982 resulted in effectively a more sensitive examination compared to 1981.
The 0.5"x1.0" 2.25 MHz transducer used in 1981 will have a lower sen-sitivity to small defects (due to its large size) than the transducers used in 1982. It is expected that thc.se indications would have been on the order of 20% DAC or less in 1981. The 1981 procedure required a 50% DAC reporting level.
- 4. Unground crowns may interfere (often do) with detection of cracks, in particular axially oriented ones.
PT of 32-FW-36-W revealed a substantial axial crack. This crack is not easily detectable ultrasonically due to interference from unground crown.
- 5. In 1982, more gain was added to the calibration sensitivity for scanning than'1981 (both 10 dB and 20 dB versus 6 dB).
- 6. The time spent on scanning and recording is considerably lower for 1981 than 1982, and may be too short for optimum inspection for IGSCC.
f
- 7. The same psychology of inspection after confimation of IGSCC was present in 1982 as in the case of the safe end inspection.
S-9 A.
- 8. IGSCC experience of inspection personnel was higher in 1982 than 1981 (availability of IGSCC samples and participation in EPRI NDE Center workshops).
- 9. Chemical decontamination appears to have increased detectability of cracks. A high frequency of 360* intermittent cracking was found in pipe segments which had been examined subsequent to chemical decontamination.
CRACK GROWTH /U.T. DETECTABILITY The above discussion leads to the conclusion that the 1982 examination was more sensitive than the 1981 examination, such that axially oriented cracks of some depth, on the order of 20% wall thickness, might not have been detected in 1981. However, if assuming a 20% wall crack was present in 1981, a higher than expected residual stress and an abnormally hign crack growth rate would be required to drive the crack through-wall in the time between the 1981 and 1982 exam dates. To help resolve this disparity, analytically anc experimentally determined stress profiles were developed to establish a residual stress estimate for axial cracks in the furnace sensitized NMP safe-end welds. In addition, a section of a cracked NMP safe-end was characterized by U.T. examination, and select cracks were removed for a three dimensional profiling. The analytical residual stress analysis, with supportive experimentally determined through-wall residual stresses from similar large diameter piping configurations, predicts that a flaw of 10% wall will grow to a length of 10.50" in 9 months of operation. And it would require an additional 12 months to propagate the crack through-wall. With these levels of residual stress, the crack would have to have been at least 45 to 50% through-wall 10 months prior to leakage. A correlation between the crack size prediction and a U.T. characteriza-tion was then accomplished by duplicating the 1981, and improved 1982 ISI examination techniques on a segment of a cracked NMP pipe to safe end weld. Following U.T. evaluation, the pipe segment was sectioned, and four cracks were selected for a three dimensional profiling. The first crack would have S-10
been called a reportable indication using the 1981 techniques. The second crack would be marginally detectable with the 1981 technique, and the third and fourth would not have been called reportable indications. Through the use of the more sensitive 1982 ISI methods all except one are reportable , indications, with estimated crack depths by U.T. from 0.12 inches to 0.20 inches. The three dimensional crack profiling has shown the axial cracks to be on the order of 40 to 60% wall, rather than the 12 to 20% as U.T. examination would indicate. The results are tabulated below: [' s Detectable
- by Detectable
- by d' Crack 1981 ISI "ethods 1982 Improved Techniques Crack 93_ptn .
6 Yes Yes 0.52 in. (SC% wall , and 0.20 in. de?p secondary crack) 4 Marginally Yes 0.610 in. (48% wall) 1 No Yes 0.620 in. , (59% wall) 7 No Marginally 0.400 in. (38% wall)
- With unground weld crown. .
The table suggests the limit of detectability for the 1981 examination without ground weld crowns was approximately 45% to 60% thru-wall. (Visibility of crack No. 6 was enhanced by the secondary crack.) With the improved tech-nique, the detectability is significantly improved. However, further study would be required to establish the actual lower limit of detectability of an axial crack in the configurations observed at NMP-1. It can be concluded from this study that because of the U.T. methods employed and the presence of unground extended weld crowns at the safe end to pipe / elbow welds, axial - cracks up to 45 percent thru-wall, were probably present at the time of the 1981 ISI examination, but not detected. With the improved methods used in S-11
the 1982 ISI examination, the axial cracking of this depth would have been detected. G. Cause of Cracking The pattern of intergranular cracking observed at NMP-1 was somewhat greater than that observed at most other BWR plants. The cracking covers a sig-nificant fraction of the circumference of a large percentage of the weld heat affected zones (HAZ), in the 28-inch diameter recirculation system piping. Based on the observed pattern of extensive U.T. and P.T. indications, it is important to establish any plant unique causative factors that might influence the long term performance of the Type 316 Nuclear Grade replacement material. In the present NMP-1 Task Force investigation of causative and poten-tially aggravating factors, each of the three concurrent necessary conditions associated with IGSCC of welded stainless steel in BWR environment was examined in great detail. These factors are (1) a sensitized microstructure, i.e., chromium depleted grain boundaries, (2) high temperature, high purity water containing dissolved oxygen (and possible contaminants), and (3) total sustained applied plus residual stresses exceeding a threshold value, i.e., resulting in microstrains capable of continuously rupturing the passive grain boundary film normally formed at the metal / water interface (surface). Figure S-4 lists some of the potential aggravating factors explored by
- the Task Force and describes the situation existing for each factor at NMP-1.
Slightly higher pressure stress level and the absence of pipe / pipe butt welds may have been aggravating factors. These resulted in a very high percentage of welds with SRI >1.2. This may have contributed to the large number of cracked welds observed. Although the pressure stresses were fully compliant with all relevant codes, on the average they are higher than those present at recent BWR plants. Thus, the presence of higher SRI values at NMP-1 would explain the higher incidence of intergranular crack initiation since the SRI is a measure of the probability of crack initiation at a given weld HAZ. The SRI has not, however, been correlated with the total number of multiple nucleated cracks at any given HAZ or with the depth of cracking (which is a function of the through-wall stress distribution). S-12 l l . . - , . w w p ; , g . . .-
.._ m m ,. . . , . g, , ,_ , .y.,
i ___ l In addition to the high percentage of SRI >1.2I welds, the implementation of decontamination prior to most of the U.T. examination appears to have r:sulted in the reporting of a significantly? higher. percentage of welds with 360* intermittent indications than for other plants examined without d; contamination. ' -- The Task Force considered the relevance o,f the previously observed NMP-1
~
piping materials performance on the long term performance of the Type 316 Nuclear Grade replacement material. It is concluded, based on a review of the Type 316 Nuclear Grade literature and available test results that no IGSCC is to be expected over the remaining NMP-1 operating lifetime since one of the three necessary conditions, i.e., sensitization, is not present with the low carbon (<0.02%) material. Thus, regardless of SRI value. IGSCC will not initiate. Further, the Type 316 Nuclear Grade material ha's' added resistance against potential off-chemistry transients that might inadvertently occur over the remaining plaat lifetime. 4
\
s e J ! S-13 n
Suction Loop Number 11 12 13 14 15 SEKibow PT (LOF) . PT kOF) PT,UT PT PT, Leak in SE, IJT Elbow / Spool PT (LOF) U ., PT,UT PT PT,UT Max. Depth = 05 in. Max. Depth by UT
= 0.55 in., GE Met Exam m
Spool / Pipe E' No Spool lljT, PT UT-NO Ind. UT (2 Spools) T E PipeNah UT UT UT UT UT Valve / Pipe UT UT UT UT UT Pipe / Elbow PT,UT PT,UT PT,UT PT,UT UT,PT 0.250 in. Depth by Met ElbowElbow PT,UT PT,UT PT,UT PT,UT UT,PT PT = PT Indications UT = UT Indications NI = Not inspected Figure S-1. Status of Pipe Crack Evaluation at Nine Mila Point
Discharge Loop Number 11 12 13 14 15 PT,UT PT, UT PT,UT PT,UT PT,UT PumpElbow Ekow/ Spool PT-No Ind. _ PT, UT l .PT) PT,UTl W UT NI NI NI NI NI SpoolNaive Valve / Pipe NI NI NI NI NI vi b _ Pipe / Spool UT Spots UT, PT - PT PT, UTl Ni (2 Spools) PT-No Ind 0.320 in. Depth by test SpooiElbow PT PT FT-No Ind. PT PT-No ind. Elbow / Spool 5 Ni PT PT-No Ind. PT-No Ind. I h UT - Spool /SE Leaks in SE Ni PT,UT PT-No Ind. PT-No Ind. PT = PT Indications UT = UT Indcations NI = Not inspected saisi-u Figure S-1. Status of Pipe Crack Evalcation at Nine Mile Point (continued)
l l ll' Il l 2 A
/
1
/
/
nnn i ii f 550 f
/
3 4. 3 000
/
/ h d
/
/
l W e
/
/ 0 e l
/
/ s i
1 i P p T^ U
/
/ /
w
/
/
b o
/
/
l E
/
/ n o
r nnn
/
/ i t
iii 550 -/
/ c u
242 /
/ 8 S
- ////
i 000 3 1 p o V L o d i W e n a i n s n A
/ i o
/
/ 6 t
a
/
/
i i c
-/
/ d n
/
/
I T
/
/
P
/
/
f o
/
/ h c
/
/ t e
/
/ 4 k
S
//
i f . f 2
// S
//
/
e r
- /
/
/
/
fV 2 i F u g e p
. i I
P
.y .
l l llllll llfl!I \ lllj\ i>l!
Percent of Pipe Approximate No. Welds With Approximate No. of Welds With Reportable Plant Welds Inspected Crack Indications _ Indications Nine Mile Point 1 62 53 85 Monticello 135 6 4 Browns Ferry Unit 2 49 2* 4* Quad Cities Unit 1 9 0 0 Dresden Unit 2 54+ 10** 20 Millstone Unit 1 13 0 0 Hatch Unit 1 63 6 9 Hatch Unit 2 117 38*** 32 Brunswick Unit 1 36 3 8 Oyster Creek 31 2 6 Duane Arnold 63 0 0 Peach Bottom 2 3 1 33 Peach Bottom 3 97 15 15 (ongoing) Vermont Yankee 59 34 58 (ongoing) Browns Ferry Unit 1 33 22 67 (ongoing) Cooper 116 15 13 I (ongoing) - I i l l
- Presumed to be due to fatigue, l
** Includes one indication for 28" furnace sensitized safe end. *** Includes one 360 circumferential indication in manifold end cap 20-30%
of wall confinned to date.
+1 of 2 indications still under evaluation + includes 2 FSSE Figure S-3. ISI of Large S/S Piping Welds per NRC IE Bulletins 82-03 and 83-02 for Domestic Operating Plants Who Have Inspected (asof6/83).
S-17
POTENTIAL AGGRAVTING FACTORS SITUATION AT NMP-1 o DEGREE OF SENSITIZATION
- Possibility of furnace sensitization - Evaluation of fabrication records due to fabrication procedures. plus metallography indicates no furnace sensitization present. - Presence of localized and/or uniform - Evaluation of metallography indicates cold work. no undue cold work was present. - Excessive weld repairs. - Meta 11ographic evaluation of cracked welds indicates that weld repair is not a significant factor.
o STRESS LEVEL
- Potential for excessive stresses /
cyclic loadings.
. Pressure stresses. - Somewhat higher than in recent BWR's but still well within relevant 1968 B 31.1 code allowables. But it is not clear whether this is unique to the NMP-1 design. Therefore, no difinitive conclusions can be made on the role of pressure stress.
. Stress intensification factors - More pipe / fitting welds and fewer due to the presence of fittings. resultant pipe / pipe butt welds than for most other BWR recirculation systems resulting in high stress intensification factor welds.
Plant run with some loops valved - Added stresses were minor, out. Hangup at supports. - Added stresses were minor, o ENVIRONMENT
- Steady state water quality. - Avg. conductivity /slightly higher than avg. Cl'.
- Water chemistry transients. - 1971 & 1979 conductivity and chloride transients are unlikely to have accelerated crack initiation.
- Pipe decontamination prior to U.T. - Did not aggravate cracking but made cracking more visible to U.T.
Figure S-4. Potential NMP-1 Aggravating Factors i S-18
SECTION I INTRODUCTION The Nine Mile Point-1 (NMP-1) nuclear power plant is a Model 2 Boiling Water Reactor (IsWR) owned and operated by Niagara Mohawk Power Corporation (NMPC). The nuclear steam supply system (NSSS) was designed by the General Electric Crpany.
<- The architect-engineer function was performed by Niagara Mohawk who in turn contracted Stone and Webster Company to construct the power plant. The construction permit was issued in April 1965, with first synchronization achieved four and one-half years later in November 1969. Commercial operation began in December 1969, and full power was achieved two months later in January,1970.
Nine Mile Point-1 was originally designed to generate 500 MWe at full power. This was later increased to 610 MWe. The plant has five separate recirculation loops which are constructed of 28" diameter, Type 316 stainless steel piping. Through P. arch,1932, the NiiP-1 plant was in operation for aooroxim3tely 12 years, (78,000 on-line hours). On March 20 1982, the reactor was shut down to repair a recirculation pump seal. Following this repair, a hydrotest of the primary system was performed on March 23 to check the repair. During this check leakage was visually detected at two of the ten recirculation system safe-ends. These safe-ends, constructed of T-316 stainless l steel, had been furnace sensitized during the initial manufacture of the reactor pressure vessel. With this discovery, further visual inspection revealed three pinhole indications and a single 1/2" long indication oriented parallel to the axis of the safe-end(s). All the indications were located in the heat-affected zone (HAZ) on the safe-end side of the weld joining the piping to the safe-enda Ultra-sonic inspections were then performed on the two leaking safe-ends plus one additional safe-end, and intermittent crack indications were confirmed and verified by a boat sample to be IGSCC. Based upon these events, Niagara Mohawk decided to replace all the recirculation system safe-ends. I An NRC meeting was held in Bethesda, Md., on April 22, 1982, to review these l IGSCC incidents and to obtain NRC approval for the proposed safe-end replacement ! program. 1-1 I
Following the initial series of inspections, the non-destructive examination effort was expanded. Ultrasonic examination was performed on the HAZ's of the recirculation pump discharge casting to riser elbow welds. Code reportable indications were found in two of the five elbow welds and non-reportable U.T. indications were found in the other three. The indication orientation was ci rcumferential . Examination by dye penetrant on the inside diamter of the pipe confirmed the presence of cracks, and replication techniques . substantiated the intergranular nature of the cracks. Finally, a boat sample was removed from one discharge weld and destructively examined revealing intergranular stress corrosion cracking (IGSCC). Once again the examination effort was expanded. Ul trasonic inspection was performed on all remaining welds where radiation fields were accept-a bl e . Cracking was indicated in a large number of welds. In a letter dated August 6,1982, NMPC notified the NRC of this extensive additional apparent cracking and indicated the intention to replace all the 28-inch recirculation system piping. In early 1978, The presence of IGSCC in large diameter piping was not new. cracks were found at the KRB plant in West Germany. A total of six welds were found cracked in the HAZ on the pine side of the oice-to-safe-end welds. These circum-ferential cracks, found in 24" diameter T-304 stainless steel piping, initiated very close to the fusion line and propagated up to 0.2" into the pipe wall. This plant had also operated for a substantial time, approximately 68,000 hours. The discovery of cracks in large diameter recirculation piping was not unexpected after the earlier IGSCC incidents in small diameter piping. The necessary conditions for IGSCC: sensitized material, high stresses, and oxygenated coolant, exist regardless of pipe diameter. The NRC also concluded that there was no safety concern for cracking in large piping as stated in NUREG-0531. In early 1981, additional cracks were detected in the 22" diameter recirc-ulation ring header at the Fukushima Unit 3 nuclear power plant. These cracks, were located adjacent to the sweepolet which joined the header to the riser piping. Indications were found in a total of three tee joints using ultrasonic examination and later confirmed using dye penetrant. The header had been constructed of Type 304 stainless steel. Following detection, the joints were given an induction heating stress improvement to mitigate IGSCC, temporarily, and replaced one year later with T-304LC stainless steel as a permanent fix. 1-2
l l Based on extensive NRC review of the cracking in the NMP-1 recirculation piping and its generic implications, in Bethesda, Md., on September 20, and October 15, 1982, the NRC issued IE Bulletin 82-03, Rev.1, (dated Oct. 20,1982) which established specific recommendations to be taken by eight licensees who owned plants that were to have an outage prior to January 31, 1983. The recommenda-tions required inspection of the recirculation piping. These inspections led to the discovery of cracking in large diameter T-304 stainless piping in six other plants to date, Monticello, Hatch-1, Brunswick Unit-1, Dresden-2, Vermont Yankee, and Hatch Uni t-2. Following the discovery of extensive IGSCC at NMP, NMPC decided to replace the entire 28-inch recirculation system piping with Type 316 Nuclear Grade material. To provide added assurance that the 316 Nuclear Grade replacement material will not suffer similar cracking due to some unique plant specific cause, NMPC appointed an interdisciplinary Task Force on Sept. 29, 1982. The stated objectives of the Task Force were:
- 1. Investigate the cause of cracking in the recirculation system piping and sa fe-ends .
- 2. Determine the reasons why cracking was not seen before 1982.
- 3. Evaluate the adequacy of ultrasonic testing techniques.
- 4. Evaluate system pipe stresses and stress rule index values.
- 5. Examine other relevant issues which may bear on generic concerns.
- 6. Evaluate the adequacy of the replacement recirculation piping material.
The Task Force held four meetings to fully review the NMP-1 IGSCC situation l and to establish the necessary inputs needed to allow accomplishment of the stated objectives. This report documents results of relevant investigations, findings and assessments made by the Task Force. I l i l l 1-3/1-4
SECTION II EXTENT AND DISTRIBUTION OF CRACKING s Of the 76 welds in the five recirculation loops, 62 have been examined by ultrasonic testing (U.T.) or liquid penetrant (PT). Boat samples of leaking safe ends and cracked weld heat affected zones (HAZs) have been sent to General Electric in San Jose and J.G. Sylvester Associates, Inc., for metallographic examination. Full-circumference samples containing the weld and, in most cases, several inches of pipe on either side have been distributed to a variety of testing labs through the coordinative efforts of the Electric Power Research Institute (EPRI). EXA'11 NATION RESULTS The ultrasonic and liquid penetrant examination results are recorded on the recirculation system loop welding diagrams shown in Figures 1-5. Forty-four welds have been examined by U.T. , with indications reported in forty- three. The indications ranged from barely perceptable spots to lengths of six inches or longer. Not all of the U.T. indications were reportable in accordance with Section XI of the ASME code. The distribution of crack lengths as determined by U.T. is shown in Figure 6, which illustrates that the majority of the recorded indications are one inch _or less in length. Forty-nine of the 76 recirculation system welds were liquid penetrant examined in order to obtain additional data as to the extent of cracking, and to corroborate I the U.T. data. Most of this work was performed at Battelle Labs under contract with EPRI. Several welds were examined at Nine Mile Point by Niagara Mohawk and General Electric personnel. Indications were found in 37 of the 49 welds examined by PT. Weld rollout diagrams with shetches of the PT indications are attached (Figures 7 through 11). Photographs of selected PT indications are presented with the metallographic examination results in the next section. 2-1
COPRELAT'ON OF U.T. AND PT RESULTS Because indications were detected in nearly all (43 of 44) of the welds examined by U.T., but only about three-fourths (38 of 49) of those examined by PT, it is apparent that there are other effects producing U.T. indications or tight cracking not detected by PT. A comparison of the U.T. and PT results for the weld HAZs examined by both techniques is provided in Table 1. Of 51 weld HAZs examined by both U.T. and PT, there are 40 in which agreement (cracking or no cracking) exists. Two of the HAZs have PT but not U.T. indications, and nine have U.T. but not PT indications. The latter result, U.T. but not , ...aications, could result from oxide buildup within a crack, preventing the penetrant from entering, or from U.T. signals caused by geometric reflectors. Conversely, U.T. non-detection could result from an inability to properly position the U.T. probe due to geometric constraints of the pipe surface, or from IGSCC related attenuation resulting from extensive grain boundary deterioration. Tables 2 through 4 summarize the U.T. and PT data. Table 2 lists the number of welds with U.T. or PT indications in either of the adjacent HAZs, and Table 3 is a comparison of the frequency and extent of cracking in the suction and discharge sides of the loop. Table 2 shows that 53 of the 62 welds examined (85%) have crack indications associated with them. Thirty-f),r of the 62 examined welds are field welds, of which 29 (85%) have indications. Twenty-eight are shop welds, of which 2 '6%) have indications. And as can be seen in Table 3, the frequency of cracking is c .at equal for the suction and discharge side. Table 4 shows that 75 of 104 (72%) HAZs examined have indications. This is less than the percentage of welds with indications because several welds have indications in only one of two HAZs. As is the case for shop and field welds, the incidence of cracking of shop and field weld HAZs is nearly identical (indications in 75% of shop weld HAZs, 71% of field Weld HAZs). Table 4 also compares the incidence of cracking among the HAZs of pipes, elbows, reducing tees, and safe ends. Pipe and elbow HAZs, which comprise the majority of the population, also have nearly identical rates of cracking: 80% vs 74-77% for pipes and elbows respectively. Because it is not clear whether the indications next to Loop 14 SW-9 are on the elbow or the tee side, 0 of 7,or 1 of 7 reducing tee HAZs may have indications. 2-2
In either case, this is substantially lower than the incidence of cracking of the other fittings. Since the carbon content of the tees is between 0.054% and 0.075% (see Section V), and the stress rule index is greater than 2.0*(Section IV), there is no obvious explanation for this, other than the fact that tee HAZs
- comprise a small percentage of the total population of HAZs and thus have a higher probability of deviating from the mean.
For the HAZs with PT indications, the fraction of the circumference that is cracked has been estimated. The results are listed in Figures 1-5 and summarized in Tables 5 and 6. A size distribution of the PT indications is shown in Figure
- 12. Table 5 lists the average percent of circumference cracked for each type of HAZ. For the HAZs of shop welds, field welds, pipes and elbows the average circumferential extent of cracking is 15-16%. This, and the fact that there was little variation in the incidence of cracking (% of HAZs cracked), illus-trates that the HAZs of shop and field welds, and of pipes and elbows were equally susceptible to cracking in the environment to which they were exposed.
The average circumferential extent of cracking of furnace sensitized safe ends is slightly higher than that of pipe and elbows, with much of the safe end cracking axial and, in two cases, through-wall . The extent of cracking in each of the five recirculation loops is compared in Table 6 The incidence of cracking ranges from 59% in loop 15 to 94% in loop 12, and the average circumferential extent ranges form 6.3% in loop 11 to 22% in loop 13. It appears that the mostextensive cracking occurred in loops 12 and 13, but the leaking safe ends were in loops 11 and 15. On an absolute scale the incidence of cracking in each loop is quite high. It should be noted, however, that the degree of inspection performed on the Nine Mile Point-1 recirculation system was unusually high, and that Nine Mile Point-1 is one of the oldest oprating BWR plants. Other plants, until recently, have not inspected to the degree of Nine Mile Point-1. CThese fittings can be treated as " branch connections" rather than reducing Tee's. As branch connections the stress rule index value is 1.7 rather than 2.1. 2-3
I TABLE 1 COMPARIS0N OF ULTRASONIC AND LIQUID PENETRANT EXAMINATION RESULTS No. of HAZs No. of HAZs No. of HAZs Reci rculation No. of HAZs . With Agreement liith UT, With PT, Loop PT'd and UT' d Between PT & UT But Not PT Ind. But Not UT Ind. 11 12 7 5 0 12 10 8 2 0 13 13 10 1 2 14 8 8 0 0 15 8 7 1 0 51 40 9 2 1 4 d 4 1 i 2-4
TABLE 2 COMPARISON OF SHOP AND FIELD WELDS 1 No. Welds With Type. of . No. Welds UT 'or PT Loop Weld No. Welds Examined Indications 11 Field 9 7 6 Shop 6 6 5 12 Field 9 5 5 Shop 5- 5 5 4 13 Field 9 7 7 i Shop 6 6 5 14 Field 9 7 5 Shop 6 6 5 15 Field 10 8 6 i Shop 7 5 4 Field 46 34 29 = 85%*
'Shop 30 28 24 = 86%*
Total 76 62 53 =85%* l l l \
- Fraction of examined welds.
2-5
TABLE 3. Comparison of Suction and Discharge Piping Welds No. HAZ's With Indications / No. Examined Loop _ Suction Discharge 11 6/11 9/11 12 8/9 7/7 13 10/11 8/11 14 7/11 7/11 15 1 0/13 3/9 41/55 (75%) 34/49 (69%) Ave. % Circ. by PT (HAZ's w/PT Ind.) Loop Suction Discharge 11 5.0% 6.7% 12 11% 20% 13 40% 9.1% 1 14 18% 8.9% 15 13% 29% 20%* 13%*
- Note that this is not simply the average of the above averages.
2-6
TABLE 4 EXAMINATION RESULTS BY HAZ No. HAZs With Indications /No. Examined HAZ Type Loop 11 Loop 12 Loop 13 Loop 14 Loop 15 ' Total Shop Weld 7/12 9/10 9/12 9/12' 7/10 41/56 (73%) Field Weld 8/10 6/6 9/10 Sf' ' 6/12 34/48 (71%) ! Pipe 9/10 7/8 9/12 7/10 7/9 39/49 (80%) Elbow 5/8 7/7 7/8 5-6/8 5/8 29-30/39 (74-77%) Tee 0/2 0/0 0/0 0-1/2 0/3 0-1/7 (0-14%)
'? " Safe End 1/2 1/1 2/2 1/2 1/2 6/9 (67%)
15/22 (68%) 15/16 (94%) 18/22 (82%) 14/22 (64%) 13/22 (59%) 75/104 (72%) O _..o_.._
TABLE 5 4 Average Fraction of Circumference Cracked by PT HAZ Type Ave. % Circ. by PT Shop Weld 16% Field Weld 16% Pipe 15% Elbow 15% Tee 0% or 13% Safe End 23% TABLE 6 Extent of Cracking by Recire. Loop
% Cire. by Pt Loop (HAZs with PT Indications) 11 6.3 12 17 13 22 14 13 15 19 2-8
x
,e J
- n-i s.-i -
I
- s. _
V tt N-23 su-Is - C
- u. ,
8 _ s.
- m-3 FW-24 h 25
_ s..u s.-s u-as
}
n-. Weld HAZ UT Results PT Results % Circ. by PT FW-1 Sa fe-End --- No Indications 0 Elbow --- No Indications 0 SW-1 Elbow --- No Indications 0 ) Tee --- No Indications 0 l SW-2 Tee No Indications No Indications 0 { Pipe Indications No Indications 0 l FW-2 Pipe Indi ca tions --- - FW-3 Pipe Indications --- - SU-3 Pipe Indications No Indications 0 Elbow Indications Circ. Indications 0-5 FW-4 Elbow (Upstream) Indications Circ. Indications 5-10 FW-26 Elbow Indications Circ. Indications 0-5 SW-17 Elbow Indications No Indications 0 Pipe No Indications No Indications 0 SW-16 Upstream Pipe Indications No Indications 0 Downstream Pipe Indications No Indications 0 SW-15 Pi pe --- Circ. & Axial Indications 5-10 Elbow --- No Indications 0 FW-23 Elbow --- Axial Indications 0-5 Pipe --- Circ. & Axial Indications 5-10 FW-22 Pipe Indications Circ. & Axial Indications 0-5 Safe End Indications Circ. & Axial Indications, Leak 10-15 Figure 1. Recirculation Loop No.11 Examination Results. 2-9
- n. . . - - . . . _ - . .
q f ) N-5 5W-4 - 9 N-11
% 29 lu-18 -
E E m-4 _ m-7 Z m rv
~
5d-20 Sh-5 - mJ E g,p UT Results PT Results % Circ. by PT Weld HAZ FW-5 Safe End Indications No Indications 0 Elbow Indications No Indications O Elbow --- 1 Axial Indication 0-5 SW-4 Pipe --- No Indications 0 FW-6 Pipe Indications --- - FW-7 Pipe Indications --- - Pipe Indications 2 Axial Indications 0-5 SW-5 Elbow Indications Circ. Indications 0-5 Upstream Elbow Indications Branched Circ. Indications 35-40 FW-8 Elbow Indications Branched Circ. Indications 25-30 FW-31 Elbow Indications Circ. Indications 0-5 SW-20 Pipe Indications Circ. Indications 10-15 SW-19 Upstream Pipe Indications Branched Circ. Indications 55-60 Downstream Pipe Indications Branched Circ. Indications 25-30 Pipe --- Circ. & Axial Indications 5-10 SW-18 Elbow --- Circ. Indications 5-10 Figure 2. Recirculation Loop No. 12 Examination Results. I 2-10
p
/ )
- m-9 .
b6 - I Su 7 - m-32 h21 - C~ btt n FW-11 FW-34
~ '
nn
~
bt) w-s L () ( mx I
. FW.3, l l
Wald HAZ UT Results PT Results % Circ. by PT FW-9 Safe End Indications Circ. & Axial Indications > 50% Elbow Indications No Indications 0 , SW-6 Elbow Indications Branched Cire. Indications 45-50 l Pipe Indications Branched Circ. Indications 35-40 l SW-7 Upstream Pipe Indications Indications Downstream Pipe --- No Indications FW-10 Pipe Indications --- - FW-ll Pipe Indications --- - SW-8 Pipe Indications Circ. & Axial Indications 40-45 Elbow Indications Circ. & Axial Indications 15-20 FW-12 Upstream Elbow Indications Branched Circ. Indications 40-45 FW-36 Elbow Indications Circ. & Axial Indications 0-5 SW-23 Elbow Indications Branched Circ. Indications 20-25 Pipe No Indications Branched Circ. Indications 30-35 SW-22 Upstream Pipe --- Circ. & Axial Indications 5-10
'l Downstream Pipe ---
Circ. Indications 0-5 SW-21 Pipe --- No Indications 0 Elbow --- No Indications 0 FW-33 Elbow -= Cire. Indications 0-5 Pipe --- No Inaications 0 FW-32 Pipe No Indications Circ. & Axial. Indications 5-10 Safe End Indications Circ. Indications 0-5 Figure 3. Recirculation Loop No.13 Examination Results. 2-11
,M' W-13 lJ 58-9
( 5d-10 - N- 37
$w 24 -
C
" "" in- 25 m-14 m.15 g ...
N.40
~
c) SW-26 5 .n __.
~~
\,.)
u.ss Weld HAZ UT Results PT Results_ % Circ. by PT FW-13 Safe End --- Circ 6 Axial Indications 30-35 El bow --- No Indications 0 SW-9 Elbow or Tee -- Circ. Indications 10-15 Elbow or Tee -- No Indications 0 SW-10 Tec No Indications -- - Pipe No Indications --- - FW-14 Pipe Indications --- - FW-15 Pipe Indications --- - SW-11 Pipe Indications Circ. & Axial Indications 15-20 Elbow Indications Circ. & Axial Indications 15-20 FW-16 Upstream Elbow Indications Branched Circ. Indications 15-20 FW-41 El bow Indications Circ. & Axial Indications 0-5 SW-26 El bow Indications Branched Circ. Indications 20-25 Pipe Indications Circ. Indications 5-10 SW-25 Upstream Pipe Indications Circ. Indications 0-5 Downstream Pipe Indications Branched Circ. Indications 0-5 SW-24 Pipe --- Circ. Indications 0-5 El bow --- Cire. Indications 20-25 FW-38 Elbow --- No Indications 0 Pipe --- No Indications 0 FW- 37 Pipe --- No Indications O Safe End --- No Indications 0 Figure 4. Recirculation Loop No. 14 Examination Results. 2-12
i p l i : n :
/> ;
m.u _,I \ ' n.n C
~" ~
nn j mo i, - w.a u.i, ,
~
[ - w.a
._r ~
-., )
u.i. _- t i
~ ,
)
n.n I Weld HAZ UT Results PT Results % Circ. by PT i FW-17 Safe End Indications Circ. Indications, Leak 20-25 Elbow No Indications No Indications 0 SW-12 Elbow Indications Circ. Indications 10-15 , Indications Circ. Indications Pipe 15-20 ! SW-13 Pipe Indications --- - I Tee No Indications --- - FW-18 Tee No Indications --- - Pipe Indications --- - l FW-19 Pipe Indications --- - FW-20 Pipe Indications .--- - SW-14 Pipe Indications Cir.c. Indications 5-10 Elbow Indications No Indications - FW-21 Upstream Elbow Indications Circ. Indications 0-5 FW-46 Elbow Indications Circ. Indications 20-25 SW-30 Elbow --- Branched Circ. Indications 40-45 Pipe --- Branched Circ. Indications 20-25 i FW-43 Tee --- No Indications 0 i Elbow --- No Indications 0 i SW-27 ~ El bow --- No Indications 0 , Pipe --- No Indications 0 l FW-42. Pipe --- No Indications 0 Safe End --- No Indications 0 Figure 5. Recirculation Loop No.15 Examination Results. 2-13
A 4. TtTrAL WDOS
, FIELb WELA5 (N)
. __ sWeP WELb5 (.SW) o M~
C - D .. 5 8 20-if. . I m 2 - 10 - 8- - !l ,; 6-- ' I 4-- l i-
- g. ..
L - _3 ,
. . . 1 . ! .
j'
. i" 2
2' 5" 4" $" 6" >d" CRACK LENGW (INCHES) _ Figure 6. Frequency Distribution of Crack Lengths 2-14
r itAJOR RADIUS OF ELBOW T SAFE-END LOOP 11 WELD R4-1 ELB0W NO INDICATION 5 F W W W W . . . . . . . 12 2 4 6 8 10 12 ELB0W LOOP 11 WELD SW-1
~k TEE
,l
*Possible lack-of-fusion.
- 12 2 4 6 8 10 12 1
TEE LOOP 11 WELD SW-2 l PIPE NO INDICATIONS 12 2 k h 8 10 12 FIGURE 7: Dye Penetrant inspection results, welds of Loop 11. (Refer to Figure 1 for weld locations.) 2-15
1 PIPE 4 LOOP 11 WELD SW-3 ELB0W (12 o ' clock is mior radius of elbou.) w . w W . . . W F W W W 12 2 4 6 8 10 12 ELB0W (Upstream) LOOP 11 WELD FW-4 ELB0W (Cut through middle of veld.) T y . , . F . 12 2 4 6 8 10 12 PUMP LOOP 11 WELD FW-26 ELB0W (Cut through middle of veld.) 12 2 4 6 8 10 12 l FIGURE 7. (continued) Dye Penetrant inspection results. Welds l (Refer to Figure 1 for weld locations.) of Loop 11. l 2-16 l n-
I l ELB0W LG0P 11 tlELD SW-17 PIPE NO INDICATIONS (12 o' clock io minor radius of cibou.)
' W W W . . . . . . . e 12 2 4 6 8 10 12 l
PIPE 1 LOOP 11 WELD SW-16 PIPE NO INDICATI0_NS j 12 2 4 6 8 10 12 l i (12 o' clock is mjor radius of albou veld.) ELB0W W Wm m
- 1 PIPE (Weld vao cub'through Z/4 of aroun on the pipe aide with plaam torch.)
12 2 4 6 8 10 12 FIGURE 7. (Continued) Dye Penetrant results, welds of Loop 11. (Refer to Figure 1 for weld locations.) 2-17
ELB0W , A LOOP 11 j WELD FW 23 y ,s - <e - . y \M si PIPE (Halfucid. Plaoma cut on pipa aide of acid) (12 o' clock 58 ma5o!*ltzdiu3 of Cibo0.) 12 2 4 6 8 10 12
\ i
) I PIPE g NOT INSPECTED t , . ,
_ l LOOP 11 WELD FM-22
~
met / e gw ) )j a SAFE END (12 o' clock is section stampad 2-1.) 12 2 4 6 8 10 12 FIGURE 7. (Conti nued) Dye Penetrant inspection results, welds of Loop 11. (Refer to Figure 1 for weld locations.) 2-18
i. (12 o' clock io major mdius of cibou veld.) ELB0W Boat Sample t 8'
~
LOOP 12 WELD SW-4 PIPE Parted through middle of veld.
' W W W W . w . . w . .
12 2 4 6 8 10 12 PIPE l s t LOOP 12 WELD SW-5 ELBOW (12 o ' clock is major radius of elbou.) 12 2 4 6 8 10 12 (12 o' clock is mjor radius of elbou.) ELB0W LOOP 12 WELD FW-8 PUf1P Parted through middle of veld. 12 h k d h 10 12 FIGURE 8. Dye Penetrant inspection results, Loop 12 welds. 4 (Refer to Figure 2 for weld locations.) 2-19
I (12 o' clock is minor radius of cibou.) Puf1P LOOP 12 WELD RJ-31 ELB0W Parted through center of veld. 12 2 4 6 8 10 12 ELBOU LOOP 12 WELD SW-20
~ ~ ~ - n n. .
PIPE (12 o' clock ia minor radius of elboo.) 12 2 4 6 8 10 12 4 Trepanned sections for crack ei::ing
"" "#" U "* * "'*'
PIPE r ~ /Av d n M A, h,* , a s vh ~ xQ I
\
l 1 % ( } l , LOOP 12 1 1 WELD SW-19( ( ,
~,, - % ~ v , my .. s~
,I-'4 l-/ 1-a 1-3 PIPE (12 o' clock is punch mark on veld.)
12 2 4 6 8 10 12 l FIGURE 8. (Continued) Dye Penetrant inspection results, Loop 12 welds. (Refer to Figure 2 for weld locations.) ! 2-20
4 Plaam2 cut through edge of crown on pipe side of veld. ' PIPE v-LOOP 12 WELD SW-18
, - ,. ~
ELB0W (12 o ' clock is major mdius of elbou.) 12 2 4 6 8 10 12 FIGURE 8. (Continued) Dye Penetrant inspection results, Loop 12 welds. (Refer to Figure 2 for weld locations.) l l 2-21
l I l Only 12 o' clock to 6 o' clock (180*) of j SAFE END this veld is recorded on this sketch. lL A D '? - _ _ > - _ r_f - v J ,fi.f A s . LOOP 13 WELD FW-9 ELBOW (12 o' clock is major radius of elbou) F W W W . . . . . . . = 12 2 4 6 UT Depth measurements ( ELBOW I W'
.to~9 5 g . so- .
&# - _ _ _ _2 ss. ~ L Ll LOOP 13 WELD SW-6 jc PIPE .S~orrm sy 9t:4p:44
, W . . . . . . . . . .
12 2 4 6 8 10 12 PIPE
-ux k Y~~ ~ AM ... __.
LOOP 13 WELD SU-8 n
~~ , __
ELB0W (12 o' clock is major radius of elbou 12 2 4 6 8 10 12 FIGURE 9. Dye Penetrant inspection results, Loop 13 welds. (See Figure 3 for weld locations.) 2-22
(13 o' clock is nn,ior radius of elbou.) ELB0W
._ ~ ~ .- - . a :..
LOOP 13 WELD FW-12 PUf1P Section was parted through center of veld. F W y W . . . . . , , , 12 2 4 6 8 10 12 (12 o' clock is prinor radius of elbou.) PUf1P LOOP 13 tlELD Bl-36
- ~
ELB0W Section uac parted through center of veld. ! 12 2 4 6 8 10 12 ELBOW
~ u .1w .~ n a n,. . - ~
LOOP 13 WELD SW-23 m - - . - . ~ y _, - ._ PIPE (12 o' clock is minor radius of elbou.) 12 2 4 6 8 10 12 FIGURE 9. (Continued) Dye Penetrant inspection results, Loop 13 welds. (Refer to Figure 3 for weld locations.) 2-23
PIPE Not Recorded
- -- n \ ,
LOOP 13 WELD SW-22 PIPE Three 900 sections were received. F y y y . . . 12 2 4 6 8 10 12 l (12 o' clock is minor mdius of elbou.) PIPE LOOP 13 WELD SW-21 NO INDICATIONS OBSERVED ELB0W section was plasma cut at ocid crown on pipe side. 12 2 4 6 8 10 12 (12 o' clock is major radius of elbou.) ELB0W LOOP 13 WELD'FW-33 PIPE , Most of veld root and HAZ destroyed by plasma l cutting. 12 2 4 6 8 10 12 l FIGURE 9. (Continued) Dye Penetrant inspection results, Loop 13 wel ds . (See Figure 3 for weld locations.) 2-24
PIPE This section not recorded.
~ $\e ,e a r 3 _&
LOOP 13 WELD FW-32 v S AFE-END (12 o' clock is long seam of spool piece.) 12 2 4 6 8 10 12 FIGURE 9. (Continued) Dye Penetrant inspection results, Loop 13 welds. (See Figure 3 for weld locations.) l i 2-25
(12 o' clock at left edge of section
~
SAFE-END Pipe was received in four 900 sections. 4-=~-~~ ~ m . . x . .a. ~ LOOP 14 WELD FW-13 f i 7 a p s 4 t ELB0W 5 l Elbou scam celds m _
)
12 2 4 6 8 10 12 (12 o' clock at major radius of elbou.) ELB0W (?) LOOP 14 WELD SW-9 TEE (?) Section was cut through croun of elbou side of veld. 12 2 4 6 8 10 12 (12 o' clock at major radius of elbou.) PIPE
-vh e /~N /N\ ,
, w cw
{' I LOOP 14 {, j { } ($WELDSW-ll
-g,- 3
'J ~,_,
2-1 2-2 2-3 ELB0W Three trepans removed for . rack sizing. 12 2 4 6 8 10 12 FIGURE 10. Dye Penetrant inspection results, Loop 14 welds. (Refer to Figure 4 for weld locations.) 2-26
(12 o' clock is major mdius of elbou.) ELB0W L2 - _ A >A- - -- LOOP 14 WELD FW-16 PUMP Section vaa parted through center of veld.) r W y W . . . . . . . . 12 2 4 6 8 10 12 (12 o' clock is minor radius of elbou.) PUMP LOOP 14 WELD FW-14 H e x ELB0W Section was parted through center of ueld.) 12 2 4 6 8 10 12 (12 o' clock is minor mdius of elbou.) ELB0W
# ~~.-- u- - % ~ s '.n LOOP 14 WELD SW-26
- -~ ~ .. ~
t PIPE _ 12 2 4 6 8 10 12 FIGURE 10. (Continued) Dye Penetrant inspection results, Loop 14 wel ds . (Refer to Figure 4 for weld locations.) 2-27
PIPE (Upstream)
~ -.
LOOP 14 WELD SW-25 y . - PIPE (Downstream) (12 o ' clock at punch mark is center of weldJ 12 2 4 6 8 10 12 l 1 PIPE
.- ' - _ - _ - _ _ . - , s LOOP 14 WELD SW-24
-y--
ELB0W (12 o ' clock is at major radius of elbow.) 12 2 4 6 8 10 12 ELB0W LOOP 14 WELD FW-38 PIPE Section was plasma cut through center of welds.. NO INDICATIONS vere observed in remaining areas. 12 2 4 6 8 10 12 i l l FIGURF 10. (Continued) Dye Penetrant inspection results, Loop 14 i welds. (Refer to Figure 4 for weld locations.) 2-28
PIPE LOOP 14 WELD R1-37 SAFE END NO INDICATIONS OBSERVED l i . . . , , . , , , , , 12 2 4 6 8 10 12 1 l FIGURE 10. (Continued) Dye Penetrant inspection results, Loop 14 welds. (Refer to Figure 4 for weld locations.) l I 1 l l l s. 2-29
Not Recorded SAFE-END
% , A ,
LOOP 15 WELD FW-17 ELB0W 12 2 4 6 8 10 12 ilot Recorded E ,,,,y ggpgy ,g,55n
-- by U.T. r- 3 m LOOP 15 WELD SW-12 r
PIPE 12 2 4 6 8 10 12 (12 o ' clock is major radius of elbou veld.) PIPE s ~% - -. - u LOOP 15 WELD SW-14 An approximate l/4" thick flange was uelded on the elbou side of veld. The HAZ and veld were obstructed in several ELBOW areas from this welding. The pipe side was plasma cut at the veld croun. 12 2 4 6 8 10 12 FIGURE 11. Dye Penetrant inspection results, Loop 15 welds. (Refer to Figure 5 for weld locations.) 2-30
(12 o' clock is major radius of elbou celd.) ELB0W x __-__ LOOP 15 WELD FW-21 PUl1P Section was parted through the middle of weld. F W W T - . . . . . . . 12 2 4 6 8 10 12 (12 o' clock is minor radius of elbou.) PUMP Boat Sample LOOP 15
.- .- ~ w_
[ WELD FW-46 ELB0W Section uas parted through the middle of weld. Also a boat sample was removed in the elbou HAZ. 12 2 4 6 8 10 12 (12 o' clock is minor radius of elbou.) ELB0W LOOP 15 HELD SW-30
- - - _ . ,_ n -_ ,
PIPE Section was plasma cut near veld on pipe side._ 9 12 2 4 6 8 10 12 FIG'JRE 11. (Continued) Dye Penetrant inspection results, Loop 15 welds. (Refer to Figure 5 for weld locations.) 2-31
l PIPE LOOP 15 WELD FW-43 ELB0W NO INDICATIONS 12 2 4 6 8 10 12 ELBOW LOOP 15 WELD SW-27 PIPE NO INDICATIONS 12 2 4 6 8 10 12 PIPE LOOP 15 WELD FW-42 SAFE-END NO INDICATIONS - c 12 2 4 6 8 10 12 FIGURE 11. (Continued) Dye Penetrant inspection results, Loop 15 l welds. (Refer to Figure 5 for weld locations.) 2-32
I l 0-5 15 - -
% 10 -
s 5-10 5
~ ~
, 5-10-15 40-45 25-30 35-4C i 30-35 45-50 55-60 1 \ l l TOTAL % OF CIRCUf1FERENCE CRACKED (BY P.T.)
~ 'ONLY INCLUDES HAZ's W/P.T. INDICATIONS Figure 12. Extent of Cracking / Frequency Histogram.
2-33/2-34
SECTION III LABORATORY CHARACTERIZATION OF CRACKS INTRODUCTION To characterize the extent and morphology of cracking, boat samples were removed from four weld areas for laboratory examination, and one quarter circumference of pipe to elbow weld (loop 15, shop weld 12) was shipped to GE Vallecitos Nuclear Center for a comprehensive evaluation and failure analysis. An additional boat sample, containing a cracked region of a safe-end was cut out and shipped to Battelle Columbus Laboratories for detailed examination. The examinations included visual examination, metallography, and scanning electron microscopy, with the objective of characterizing the mechanism of cracking. OBSERVATIONS AND CONCLUSIONS
- 1. The mechanism of the through-wall cracking, which caused the safe-end to leak, is IGSCC of furnace sensitized Type 316 stainless steel. The mechanism of the pipe heat affected zone cracking was IGSCC of weld sensitized Type 316 stainless steel.
- 2. Shallow transgranular cracking, incidental to the IGSCC, was found on the 0.D. surface of Boat Sample "A" from weld FW-22 of Loop 11 leaking safe-end.
- 3. The cracks of Boat Sample "B" from the same Loop 11 safe-end were found to contain contaminants. Analysis of the material showed the presence of sulpur and chlorine and a number of other elements.
The origin of contaminants cannot be determined from the examinations perfo rmed. It's possible they could have been introduced subsequent to the cracking of the piping. No contaminants were found on' the fracture surfaces of Boat Sample "A". l 3-1
- 4. All safe-end thru-wall cracking, which resulted in leakage, was observed to be associated with axially (Transverse) oriented cracking.
- 5. All cracks examined were intergranular, with an inside surface intiation, in the weld heat affected zones of both field and shop welds.
- 6. Electrochemical potentiokinetic reactivation (EPR) measurements on several areas of pipe and elbow base metal gave readings near zero, verifying the base material is in a solution annealed condition. The weld heat affected zones have low values indicating a narrow sensitized region. This was also verified by metallography.
- 7. Carbide precipitation was evident (by metallographic examination) at crack initiations.
- 8. Some cracks propagated into the weld metal to a maximum distance of .050".
- 9. Some cracks had propagated across the longitudinal previously solution annealed pipe seam welds in the heat affected zone region of the butt weld.
FIELD EXAMINATION Introduction In April,1982 ultrasonic (U.T.) indications were discovered in a 28-inch Type 316 stainless steel elbow, welded in the field to the discharge side of the No.15 reci rculation pump. (Weld FW 46 in Fig. 5 of Section 2.) Cracking was confirmed by subsequent visual and dye penetrant examination of the pipe I.D. In-situ metal-lography was performed to determine the cause of cracking. Replicas of a visible crack in the HAZ were examined by the General Electric Co. in San Jose and confirmed I GSCC . 3-2
R!sults and Discussion The affected elbow is located in the pump discharge portion of the No.15 recirculation loop and connects the recirculation pump outlet to a vertical run of piping. It is welded directly to the cast stainless steel recirculation pump. The weld was made during field fabrication of the recirculation loop. Access to the elbow was gained via the recirculation pump, as sketched in Figure 1 Circumferential cracks were visible approximately 0.125 - 0.250 in, from the weld fusion line at 7:00 and 9:00.* A third visible crack, reported to be located at 12:00, was not investigated. In-situ metallography was performed by a Niagara Mohawk Power Co. employee working with a General Electric Company engineer and consisted of:
- 1. Polishing the surface to a 600-grit finish.
- 2. Etching electrolytically with oxalic acid.
- 3. Reproducing the surface structure with cellulose acetate replicating tape.
The visible crack at 9:00 was chosen for examination. A composite photograph of the replica is shown in Figure 2. The cracking is oriented horizontally across the top. Grain boundaries are not visible, but the angular, branched appearance indicates that the crack is intergranular. Areas of interest are indicated by the arrows marked A, B, and C. At location A, branches of the main crack appear to have surrounded several grains. At B and C it appears to zig-zag around the grains. Short penetrations into the adjacent grain boundaries are visible at location B. The intergranular crack morphology combined with its l location in the presumably weld senistized HAZ leads to the conclusion that the mechanism of cracking is IGSCC. Subsequent U.T. examination of other welds in the recirculation systems, resulted in the identification of a large number of indications considered to be cracks. All original recirculation system piping, between safe-ends and pump nozzles, for both suction and discharge sides of all five loops was removed for replacement. As will be discussed in a later section, the original piping is being replaced with a Type 316 Nuclear Grade stainless steel with a much lower susceptibility to intergranular stress corrosion cracking (IGSCC) than the orginally used material.
*This reference system is defined by looking into the pump outlet (against the flow direction) with 12:00 at the top.
3-3
Twelve sections of welded piping which had been cut from the recirculation systemand taken out of the drywell were examined visually and by liquid penetrant. Each section consisted of a 10-15 inch ring, containing a circumferential weld, which had been removed from one end of cither: 1) the 90 pump suction elbow (reactor vessel outlet) which connects the outlet safe-end to the downcomer or,
- 2) the 76 pump discharge elbow which connects the riser to a spool piece. (The spool piece is welded to the inlet safe-end.) The position of the elbows is shown in Fig.1, Section 2, a sketch of recirc. loop No.11. In loop 15 the positions of the shop and field welds of the 76 discharge elbow are the reverse of loop 11.
The results of visual and liquid penetrant examination are given in Table 1. A total of 20 heat affected zones (HAZ's), not including safe-ends, were examined. Visible cracks or crack-like PT indications were found in ll. Three HAZ's exhibited cracking over more than 30% of the circumference. Figure 3 photographs show circumferential PT indications on the elbow side of the loop 13 discharge elbow to riser weld. Approximately 30% of the circumference is cracked. Figure 4 shows indications on both sides of the elbow to riser weld from the pump discharge side of loop 12. Note the short axial indications in the HAZ on the pipe side of the weld. The most extensive PT indications were found in the loop 13 suction elbow-to-downcomer weld (Figures 5 and 6). Approximately 50% of the elbow HAZ and 40% of the pipe HAZ were found to contain circumferential cracks. Several cracks were visible without liquid penetrant. Boat samples were removed near the 9:00* and 10:00 locations using a hand held cutting tool. The excavations that resulted were 1/4-3/8 in. in depth. Fi gures 7 and 8 show circumferential PT indications running continuously through the excavations, indicating that the cracks are at least 1/4 in, deep. A larger grinding tool equipped with a strawberry burr was used to chase the cracks further into the pipe wall. It was found (Figure 9) that the_ crack near 10:00 on the elbow side of the weld penetrated nearly 0.50 in, into the pipe wall. Note that the crack turned to follow the weld fusion line after propagating radially for about .30 .35 in.
- The longitudinal weld seams of the elbow were designated as 6:00 and 12:00.
3-4
Ultrasonic crack depth measurements were made between the two excavations by a GE Test Engineer. The transducer was placed on the end of the pipe about 0.50 in. from the weld root. This technique was made possible because the rough plasma-arc cut surface had been machined flat. Measurements at six locations from 9:00 to 10:00 ranged from 0.20 in to 0.45 in. (see Figure 10). Good agreement i batween U.T. and grinding was obtained at the 10:00 location where the crack depth measured 0.45 in. by U.T. and 0.50 in. by grinding. The deepest crack measured by U.T. was 0.55 in., in the loop 15 suction elbow / pipe ring, on the elbow side. The corresponding PT indication was light compared to an adjacent indication which measured 0.30 in.in depth by U.T. Many of the twelve examined weldments were found to contain fairly dark circumferential PT indications which hugged the edge of the weld root. An example is shown in Figure 11. Boat samples removed from two such locations revealed that the indications were caused by a superficial lack of fusion between the weld root and adjacent base material. In most cases it was not difficult to distinguish indications of this type from crack-like indications. 3 To determine whether the base material of the recirc. piping was sensitized due to the possibility of an inadequate solution heat treatment, EPR measurements were conducted on the loop 13 suction elbow. Measurements on both the inner and outer surfaces, well away from all circumferential and longitudinal welds, resulted in Pa values less than 10-3c/cm 2 , indicating that the base material is in the solution annealed condition. This finding is consistant with documents stating that elbows fabricated by the Crane Company were heated to 2,000*F + 25 F for one hour per inch of thickness and water quenched.
)
3-5
METALL0 GRAPHIC EXAMINATIONS Boat samples and piping segments containing cracks were removed from the recircu-lation piping system and sent to laboratories for metallographic evaluation and failure analyses. Listed in Table 2 is a description of the samples removed for examination and an identification of the investigating laboratory. The results of the metallograpic examinations performed on the boat samples are described in detail in the following sections. Table 3 is a summary of metallographic examinations. A. BOAT SAMPLE "A" Boat sample "A" is one of two samples removed from the outside surface of the leaking regions of the safe-end of recirculation loop safe-end weld No. FW-22 of the Nine Mile Point recirculation loop Number 11. Each sample contained the outer surface portion of the visible cracks at the 12:00 location. The orientation of each crack was nearly transverse (axial). Figure 12 is a schematic showing the location of the safe-end weld samples examined at Battelle Columbus Laboratories (BCL) and at General Electric Companies Vallecitos Nuclear Center (GE-VNC). Sample "A" included part of the safe-end-to-pipe weld. The loop 11 recirculation loop discharge safe-end is one of two 28-inch safe ends found to have had through wall leakage. The safe-ends were fabricated from Type 316 stainless steel (0.054%C) and were used in a furnace sensitized condition due to the RPV post weld heat treatment. Visual Examination A low magnification photograph of sample "A" is shown in Figure 13. The thick (lef t) end was cut partially into the safe-end-to-pipe weld. The thin end points toward the reactor vessel. The visible crack had not quite propagated through-wall in the plane of the cut surface, but there was not much material holding the specimen together, making it easy to break apart. Itetallography The sample was broken into two pieces, identified as A-1 and A-2. Section A-1, the thick end, was mounted on its 0.D. surface to enable examination of the surface cracking. Section A-2 was mounted on its side to reveal surface and sub-surface cracking. Low magnification photographs of sections A-1 and A-2 after 3-6
BOAT SAMPLE "A" (Continued) polishing and etching are shown in Figures 14 and 15 respectively. The safe-end to pipe weld can be seen in Figure 14. Figure 15 shows another weld approximately 0.25 in. away, which is probably the result of a repair weld or weld metal build up during original fabrication. A high magnification photograph of the fracture end of section A-2 is shown in Figure 16. This is the lower left portion of Figure 20, 0.06-0.08 in below the 0.D. Surface. The primary fracture and the secondary cracks behind it are inter-granular, and the clearly visible grain boundaries show that the microstructure is sensitized. It is apparent that the intergranular stress corrosion (IGSCC) cracking initiated on the I.D. surface, and Figures and 15 provide corroborating evidence. This cracking occurred as a result of the combination of sensitization, high temperature oxygenated water, and a tensile stress (residual and applied). l The upper left region of Figure 15 is shown at high magnification in Figure 22. The top of Figure 17 is the safe-end 0.D. surface. Two transgranular cracks can be seen propagating in from the 0.D. The one on the left, marked A, is 0.005 in. deep in the plane of polish. These cracks are incidental to the leaking cracks and may have resulted from a chloride containing substance being in contact with the outside of the safe-end. It is possible that they initiated when hot reactor water began leaking from the through-wall cracks and interacted with the piping insulation. Figure 18 reveals that the tip of the crack in section A-1 penetrates approximately 0.006 in. into the safe-end to pipe weld. IGSCC will occasionally penetrate a short distance into a weld before crack arrest occurs because dilution of the outermost portion of the weld results in a local region low in delta ferrite. This phenomenon occurs during welding due to the mixing at the weld-base metal interface. Hence depletion of the outermost weld metal of ferrite forming elements can occur. Scanning Electron Microscopy (SEM) SEM was used to examine the 0.0. surface, and the fracture surface after the specimen was broken open. A composite 50X photograph of the axially oriented crack on the 0.D. surface is shown in Figure 19. The crack length is about 0.22 in., but this figure is an approximation because the crack becomes very thin, making the tip difficult to locate. 3-7
BOAT SAMPLE "A" (Continued) A typical area of the fracture surface is shown in Figure 30. The fracture is intergranular and covered by scaly oxide deposits. Energy dispersive X-ray analysis shows that the scale deposit is Cr rich, as would be expected of the oxide layer. No contaminants were found on the fracture surface. i B. BOAT SAMPLE "B" Boat sample "B" is the second of two samples removed from the outside of the leaking regions of the safe-end of recirculation loop safe-end weld No. FH-22 of recirculation loop Number 11. This sample was examined at the Battelle Columbus Laboratories (BCL). Visual Examination The boat sample shipped to Battelle was a small wedge shaped piece approximately 1 inch long and 1/4 inch thick. Figure 12 shows the schematic sketch of the safe-end and location of sample "B" which was examined at Battelle Columbus Laboratories. The sample was visually examined with a stereomicroscope. One crack approximately 1/2" long was observed on the outer surface. In addition, crusty crystalline-like deposits were observed on the outer surface, along with some rust colored stains. Figure 21 shows the appearance of the specimen in the as-received condition. Scanning Electron Microscopy The as-received sample was examined with the scanning electron microscope. The primary purpose of this examination was to analyze the deposits on the surface using the energy dispersive X-ray analyzer. Figures 22 and 23 show the appearance of deposits observed on the outer surface of the piping along with the X-ray analysis obtained on the deposits. It can be seen that the deposits are very high in silicon along with iron and chromium and nickel. In addition, trace concentrations of sulphur, phosphorous and chlorine were observed. Semiquantitative analysis obtained on one area of the deposit is shown below:
\
Si 37.2% P 0.625 S 0.55 Cl 0.47 Ca 0.064 Cr 35.57 Fe 23.38 Ni 2.04 Cu 0.4 3-8
BOAT SAMPLE "B" (Continued) Metallography Subsequent to scanning electron microscopy examination, a small piece was cut from the end containing the crack. This piece was mounted in a metallographic mount l and prepared for examination of the transverse cross section. The examination revealed one long thru-wall crack with significant branching. Figure 24 shows the appearance of the crack in the etched condition. It can be seen that the crack is entirely intergranular. Analysis of Material Trapped Within Crack Examination of the metallographic specimen in the as-polished condition showed the presence of a significant amount of corrosion products or contaminants in the crack. In order to determine the nature of the material, the specimen was examined with an energy dispersive X-ray analyzer associated with the scanning electron-microscope. A number of areas in the crack were examined. In all cases varying amounts of chlorine and sulphur were observed along with a number of other elements. Figures 25, 26, 27, and 28 show the appearance of the contaminant and the results of the X-ray analyses obtained on the material. Also shown are the semiquanti,tative elemental analyses. l obtained. These two areas shown were found to contain the highest amount of chlorine and sulphur. The origin of the material cannot be determined from the examinations { performed. it is possible they could have been introduced subsequent to the cracking of the pip 'ng. SEM Examination of the Fracture Surface The fracture surface of a portion of the specimen was examined with the SEM, after the fracture surfaces had been separated by a saw cutting through the unbroken li gamen t. Figure 29 shows a typical area on the fracture surface. The intergranular fracture morphology is clearly apparent. This fracture morphology is typical of that observed in the intergranular stress corrosion cracking mode in boiling water reactor stainless steel pipe. 3-9
C. SAMPLE 3 - PIPING SECTIONS - SUCTION SAFE-END AND ELB0W WELDS l Sample 3 are piping sections containing a part of the suction safe-end to elbow weld FW-17 of loop 15 and a part elbow-to-pipe weld, SW-12, just down-stream from FW-17. As described earlier, leaking was found at the 8:00 o' clock position on the safe-end, and numerous U.T. and P.T. indications were observed about the circumference of the weld on both the safe-end, and the pipe sides. U.T. and P.T. indications were found on the pipe-to-elbow weld SW-12. The portions of the loop 15 suction safe-end-to elbow weld were removed and shipped to Battelle Columbus Laboratories for examination. The section was located near 7:00 o' clock, (Figure 30 ). The second sample, also sent to Battelle for examination was removed from the 12:00 o' clock position of weld No. SW-12 of loop 15 (one weld down-stream from FW-17). The piping section is shown in Figure 36. Several metallographic samples were removed from the two piping sections for optical metallograhic examination. The cutting diagramof the portion from the 7:00 o' clock position is given in Figure 30. The arrows indicate the surfaces that were examined. Figure 31 is the cutting diagram for the portion of the pipe-to-elbow weld removed from the 12:00 o' clock position. Figure 32 is a 4X view of sample"A-4" showing the crack in the safe-end. This etched view was prepared by grinding through the pipe wall thickness beginning on the pipe I .D. surface. This is the inner surface view of the axially oriented through wall leaking crack in the safe-end. The 0.D surface view of the same crack can be seen in Figure 30, on that sample marked "A-4". Figure 33 is a cross-section etched view of the samt sample of Figure 32 No cracks were observed in this sample of the safe-end-to-91 bow weld. ( Figure 34 is a sectional view of sample "B-1", renoved from the portion of the loop 15 suction elbow-to-pipe weld SW-12. A short cra:k can be seen near the weld fusion line on the pipe side of the weld. Sample "B-2' , shown . in Figure 35, is a 3-10
i sample removed from a location just adjacent to sample "B-1" of Figure 31. This polished and etched view reveals a crack near the fusion line on the elbow side l of the weld. l Sample "B-3", shown polished and etched in a 4X view in Figure 36 is a cross sectional view of the elbow seam weld. No evidence of cracking was found in this view. l
?
1 3-11
D. BOAT SAMPLES "C", "D", "E", AND "F" l Boat Sample "C" This section was cut across the weld SW-6 between an elbow and pipe on loop 13 (see Fig. 3 of Section 2). Figure 37 after penetrant examination. The large indication at the left is the elbow side, and a smaller indication on the right is in the pipe side. Details of the crack on the elbow side are shown in Figures 38 and 39. Figure 38 shows the i crack location with respect to the weld. The cracking is completely intergranular, and carbides can be seen outlining the grain boundaries, indicative of light sensitization. Figure 39 shows enlarged detail of the center and bottom of the section. The carbides at the grain boundaries are more evident at higher magni-fication. This crack went through the entire section, about .200". Figure 40 shows the crack on the pipe side. It is %.120" deep, and stops immediately at the circumferential weld. The crack is in the longitudinal seam weld of the rolled and welded pipe. Since this weld has been solution annealed (prior to welding), the ferrite content is extremely low and the material behaves similarly to wrought material. Figure 41 shows the end of the crack, and it has arrested almost on contact with the higher ferrite circumferential weld. Boat Sample "D" This section was cut from the same weld as sample "C", but at a different circumferential location. The as-received sample is shown in Figure 42, with a large crack visible on the elbow side. During handling, this section separated. Figure 43 shows a portion of the fracture, and a second adjacent crack. Figures 44 and 45 show more detailed. views at the beginning and end of the crack. Grain boundary carbide precipitates are readily visible. The grain size of the elbow material is larger than normally encountered. The crack mode is intergranular. The section which separated during handling was examined by scanning electron microscopy (SEM). The entire fracture surface was intergranular. The presence of heavy oxide obliterated < any other fracture features which may have been present. 3-12
Boat Sampele "E" This sertion was removed from loop 12, shop weld SW-4, shown in the sketch of Section 2. A penetrant indication was present on the elbow side of the weld, as shown in Figure 46. The indication was not a crack, but a very small lack of fusion area which acted as a crevice and retained the penetrant. Figure 47 illustrates this condition. Boat Sample "F" This sample was removed from loop 12, field weld RI-5, between the safe-end and elbow, on the suction side of the pump. An indication was present on the elbow side, as shown in Figure 48. This indication proved to be slight lack of fusion between two weld passes as shown in Figure 49. No propagation.had occurred. 3-13
Section of Elbow to Pipe Weld A 90 section of elbow to pipe weld from shop weld SW-12, loop 15, was examined in great detail. (See Figure 5 of Section 2.) The section had been cut through the edge of the weld on the pipe side, and again about 10" into the elbow. Looking at the inside surface, the entire section was 22" long (circumferentially) and 10" axially. It consisted of 5/8" of pipe, the weld, and %91/2" of elbow material. The inside had been decontaminated electrolytically, so the surface was clean and bright except for large areas where slag from the plasna cutting operation had adhered. To simplify handling, a circumferential cut was made about 1" on the elbow side of the weld, resulting in a ring section 22" long and 13/4" wide, con-taining 5/8" of pipe, plus weld, and 1 1/8" of elbow material . This section was cleaned of slag, and examined by dye penetrant. Masking tape marked at 1" intervals was fastened to the inside to map indication locations. Seven indications, varying from 5/16" to 1 1/4" in length, were found on the pipe side, for a total length of 51/4" or 24% of the total circumference. One indication 7/8" long was present on the elbow side, 4% of total length. Typical indications are shown in Figures 50 and 51. Assuming that the center of each indication should be the deepest part of the crack, sections from the center of each indication were polished, and the crack depth measured by microscope. The cracks on the pipe side ranged in depth from
.100" to .275". The crack on the elbow side was .137". These measurements are in good agreement with the depths estimated previously by ultrasonic examination of .1 to .2".
Metallographic sections were prepared from one crack on each side of the weld. The entire crack on the elbow side is shown at low magnification in Figure 52. Crack depth at this plane was .110". It was intergranular, and penetrated into the weld about .015". !! ore detailed views of the beginning and end are shown in Figure 53. Significant grain boundary carbide precipitates are evident in the vicinity of the crack. The section was repolished and etched with Kahlings reagent to emphasize ferrite in the weld. Figure 54 shows the crack arresting when it reaches an area of the weld containing significant ferrite. Figure 55 shows the ( pipe side of this section. No cracking is present, although grains are outlined by carbides, which is caused by weld sensitization. l 3-14
. -- - _ = _ _ _ . _ . _ _ _ _ - - __ _ _ _ _ _ - _ - - _ _ _ _ _ _ _ _ _ --- __ . .-_ -__
l The crack on the pipe side is shown in Figure 56. The crack is intergranular, and has propagated into the weld N.050". The beginning and end of the crack are shown in Figures 57 and 58. There were very few carbide precipitates at this location. Most of the propagation into tte weld has been in a low ferrite zone. Because of the apparent lack of visible carbides at this section, further work was done using SEM and also dark field illumination. Figure 59 shows the pipe heat affected zone at 800x by SEM. Figure 60 shows a similar area at 800x by dark field illumination. Almost continuous carbides are present in both photos. I s % k 4 e 9 g a 4 3-15
~
f COMPOSITION Chemistry & Structure Since the heat numbers are unknown, chemical analysis of the elbow and pipe side of loop 15 shop weld 32-SW-12-W was performed and the results are given in Table 4. The compositions including carbon contents of both the elbow and pipe materials (.065 and .049%, respectively) are within the normal range for regular grade Type 316 stainless steel. SENSITIZATION MEASUREMENTS _ A. ASTM A-262-Practice A In order to determine the degree of sensitization in the safe-end material, boat sample "B" was repolished and subjected to the ASTM A 262 Practice A Sensitization test. Figures 61 and 62 show the structure observed on the specimen. The evaluation of the etched structure revealed that the sample was primarily a dual structure. Closer examination revealed that several grains were completely surrounded by ditches. Hence, according to Practice A the sample has to be evaluated as having a ditch (sensitized) structure. B. Electrochemical Potentiokinetic Reactivation (EPR) EPR measurements, a method for determining the degree of carbide precipitation in stainless steels, were made on pipe and elbow base metal and heat affected zones. All base metal measurements resulted in readings of Nzero, indicating a lack of sensitization. The values for the weld HAZs were low, indicating a narrow sensitized region existed. The low EPR values compare with the metallographic examination, where carbides were evident only in a narrow region of the weld heat affected zones. Al-though most experience with EPR has been on Type 304 stainless, with some procedural changes, the technique is also applicable for use with Type 316 stainless steel. The EPR values of welded Type 316 S/S are lower than for an equivalent carbon content welded Type 304 S/S (see Section X. ). The EPR tests verify that the base material is in the solution annealed condition. I 3-16
HARDNESS A hardness survey was made on the weld, SW-12, pipe and elbow weld removed from loop 15, with the following results: Location Hardness RB Weld 90-92 Fusion line, pipe side 90-92 Fusion line, elbow side 88-91 Pipe, base metal 85-88 Elbow, base metal 84-88 The hardness appears normal. FERRITE SURVEY Measurements by Ferrite Scope were made on various portions of a weld section (SW-12 of loop 15). Location Ferrite % Main weld, center 10.5-11.5 Consumable insert 1 Fusion line 3-6 These numbers appear reasonable except for the consumable insert (root). Since none of the cracks appear to have involved the root pass, the low ferrite has not been associated with the cracking problem.
, 3-17 l
TABLE 1 Results of Visual and Liquid Penetrant Examination Suction l l Loop Weld Resul ts 11 Elbow / Safe-End PT indications attributed to lack of FW-1 fusion at weld root. 11 Elbow / Pipe PT indications attributed to lack of SW-1 fusion at weld root. 12 Elbow /Sa fe-End Superficial lack of fusion at weld FW- 5 root verified by boat sample. Elbow carbon content =0.045% 12 Elbow / Pipe Axial PT indication in elbow HAZ, SW-4 superficial lack of fusion at weld root verified by boat sample. 13 Elbow / Pipe Extensive cracking in both HAZ's: SW-6 Approx. 50% of circumference in elbow and 40% in pipe. Max. measured crack depth =0.5 in., by grinding. Base material not sensitized by EPR. 15 Elbow /Sa fe-End PT indications in safe-end only. FW-17 15 Elbow / Pipe PT indications in both HAZ's. Max. SW-12 depth by U.T.=0.55 in. Quarter ring sent to Vallecitos revealed cracking over 24% of circumference of pipe and 4% of elbow. 3-18
l TABLE 1 (Continued) Discharge Loop Weld Results I 11 Elbow / Spool Piece Short circumferential PT indication FW-23 in elbow. Most of pipe HAZ obliterated by plasma-arc cut. I 12 Elbow / Riser PT indications in both HAZ's. Carbon SW content of elbow = 0.063%. ! 13 Elbow / Riser Circumferential PT indications over Su-21 approx. 30% of elbow. Elbow carbon content = 0.063%. 14 Elbow / Riser PT indications in both HAZ's. Elbow
; SW-24 carbon content - 0.054%.
l 15 Elbow / Riser No-PT indications. FW-43 I I r 3-19
~. . . . - - - .
TABLE 2 , SA!!PLE DESCRIPTION IDENTIFICATION LABORAT9RY 4 SAMPLE' Boat Sample "A" Axially oriented leaking crack - General Electric j
- 1. '
safe-end side of weld FW-22 of Vallecitos Nuclear Center loop 11.
- 2. Boat Sample "B" Axially oriented leaking crack - Battelle Columbus Labora-safe-end side of weld FW-22 of tories loop 11.
- 3. Piping Sections Sections containing axial Battelle Columbus Labora-cracking - suction safe-end tories to elbow weld FW-17 of loop 15 and SW-12, elbow to pipe weld of loop 15.
4 Boat Sample "C" Cracked - both sides of elbow- General Electric to-pipe weld SW-6 of loop 13. Vallecitos Nuclear Center
- 5. Boat Sample "D" Same as Boat Sample "C". General Electric Vallecitos Nuclear Center
- 6. Boat Sample "E" Penetrant indication - elbow General Electric side of SW-4 of loop 12. Vallecitos Nuclear Center
- 7. Boat Sample "F" Penetrant indication - elbow General Electric side of FW-5 of loop 12. Vallecitos Nuclear Center
- 8. Section of Elbow - 90 section of elbow-to-pipe General Electric To-Pipe weld from SW-12 of loop 15. Vallecitos Nuclear Center 3-20
TABLE 3
SUMMARY
OF METALL0 GRAPHIC EXAMINATIONS MATERIAL sat 1PLE(S) RESULTS Safe-end-field weld HAZ e Boat sample A & B from e Thru wall crack is axial FW-22 of loop 11 on both I .D. & 0.D. of e Sample FW-17 of loop 15 sa fe-end . o Material sensitized (by PWHT of RPV). e IGSCC present. e Trace contaminants of S.P. and Cl found. l Pipe & Elbow Shop Welds e Sample SW-12 of loop 15 e IGSCC contained in weld HAZ's e Sample SU-6 of loop 13 f elbow material and pipe ma terial . e Sample SW-4 of loop 12 e Many P.T. indications are due to lack of fusion. Spool piece - seam weld e Sample SW-12 of loop 15 e Solution annealed condition. e Sample SW-6 of loop 13 e Cracking by IGSCC only if weld material is in HAZ of circumferential weld. ) Pipe & elbow field e Sample FW-5 of loop 12 e Some P.T. indications due welds to lack of fusion 0 fusion e P.T. of Fil-23 of loop 11,
, and FW-31 of loop 12 II"**
e Other P.T. indications due to IGSCC.* ONote: This not confirmed by fiet, work. 3-21
TABLE 4 CHEF 1ICAL ANALYSIS OF LOOP 15 (32-SW-12-W) SUCTIO!1 ELB0W/ PIPE MATERIALS Pipe Elbow Weld Al <.005% .006% <.00f% B .003 .001 -- C* .049 .065 .050 Co 0.14 0.23 -- Cr 17.36 16.64 18.90 Cb 0.05 0.05 0.04 Cu 0.15 0.23 0.08 fin 1.64 1.68 2.00 > Mo 2.70 2.49 2.42 Ni 13.67 1 3.01 11.04 P P .033 .027 .024 S .017 .014 .020 Si 0.46 0.65 0.92 Ti .006 .020 <.005 V .06 .04 .03
- Determined by LECO carbon analyzer 3-22
l FIELD tlELD 46 0F LOOP 15
% m -,
FLOW BAFFLE I LOCATION OF CRACKING A 0
)Y RECIRCULATION PUMP - TOP VIEW 1
4 4 , 28" e l FLOW BAFFLE - END VIEW I Figure 1. Sketch of Loop 15 recirculation pump showing location of HAZ crack indications. 3-23
- :. j 2: ,s ,. w;~
- y. : +[
r.< m 7 { wi,..;Q
.: -y ~'
nb t :g . y . e p . D yN ,
,,,cy,g .{g:
E L W n o g t
.S. -
?j.
f O E G Z d y e;3 e 7
.r y' .
D E
= t e
g rs
- c hu 1,' e 4 , _
g
** yK6 ?.
$ ~ f _
,+?.4 f -
==
nf' . ..,.6 / . Mp . s{' A -
,. . f. , 2rw '.
aS t 9t .
!. a
.b.
tg. y e s _y %x _
)' gpM?w P.u< . M y ,{'4%~,
H
.(m. -
' k,. d v gs.
a*> 'x p. . t - l e p~ g r y %, Q}f. i
's:
.? ae
;g
?
W n 5 .. m h**3 ;:.c. i sU- Q,i.. Qiyw ely...
. hp43,.5 g .
a \. w. ;: 4.M'g$r.: -tg k c ge D- a N . t,y. a.g, 1
$rz,f,:o..-
[.r - E<.... .
?
h;tw yg - C r r
$ . 3
.. a
.p
$["~c_ w3.,7.Q,W% 8 5
E l u n a
.,g S
O [u a;g' .p~r. P R U r g y P r C 4- e (s%?. k w - ;
< ..&g . I H
P A R t n 3gq ;_u
- G I w*:.
O R 5,g%* P f E 57 bg . s s F. o R . R h O ht
- 3. . y F
% pu E p g : 5 am
~mQ-pO..T4' 7
O ri T
, gz O
- .. N E C
U t oA
? 2y D E o0 3W. Q.j~e^
R 1 M O h0 P : S 9 e tt i a s o) pZ mA oH C(
' 2 e
. r u
g i F 3,.
. .N .
. wINp..
l l l l t 1 ELB0W s
.:. . .. .s
/ : s.
~
. ,., i? =",
. f,
+
gr.
. p
, b =, <i s,** w 4
. g'T
'A g
$ ,"*'*?
. . . .4.
w
,i. \ ,,,. A , o y,
-f .
, .*w s . e ., . -
. s.,
; 1:
a
.t ; nc c.xx , ,
a
. g c..
s ec~..,
. g. , s.d h'
.. ...[.Ch. ,1 m' d a . -
. w .s t 9-a *
g,ed d ' 3.,
- m. - ~
snl** ,
,g, t
#'.i,,..- -
/ **
;p :-- '
; -a "'
Q. ; , . . ,. ~ W "'-W%fWW . +~..: c,,u. .s i .,
,4 .. . ,..u
,.. ,-s m2 .!:<aj
##)
1s 4 _u PIPE YMYd3kf-: I N ? % e % ,_2E D[J' *u d M i'** d* * ' kid dId Figure 3. Loop 13 Pump Discharge Weld No. P32-SW-21-W. Note all indications are on elbow side of weld.
ELBOW
.i . '*
PIPE Figure 4. Loop 12 Pump Discharge Weld No. FW-31. Weld HAZ indications on side surface of pipe and elbow.
. s. ELBOW
-y
. ; =.~.~ ,. . . ;. ,y.; s;7,, u . .;
is ,
.~g; p
r c :. , ..,A-v.
)
?e 17i . -
g;
. . ,. m g>-
PIPE e- -l - _p m 3
. . Y**
_s d ,,, , , , , s -- . , , - . Figure 5. Photo of Indications in Loop 13 Suction Elbow / Pipe Weld (Weld SW-6). 3-26 L ._ _ _ _ _ . . ___ ___
. . s. c , .r .m.. , . ,
- j
~-
f_..
- 7. _ .. . yg. -_
s_y.e w-ELBOW 13 - sS
- r. o . r 7 .. . .
lM Yl c $p' a;cj.L , 4 ..
\ll ;.-r,' . e ., . .,
- _> , f . A s-g[** ~ , *
~
'l ,
;bh*;,,g - . _ , ,l .
)
.sq
', . Tty e : h' %,
< 4 . ; ; :giQ- :?j PIPE w
., e. _,
&' . gg I .[q T d#
g g. lk1 :s n ,t .._..._. - j F U 4 w 3-. .a s - i -> * . g 9; '- . ,o
+ ,v, . ;:v ~
. ,>. v. a. ;.
ah,!~.m :,
. a u..:: . -
Figure 6. Photo of PT Indications in Loop 13 Suction Elbow / Pipe Weld. (Weld SW-6)
&sy.. l>l;[-lf f, TDG)]fJn a . i :Q -
s .<,
-, t .; A <- r ';w j;j~/i 9
, . . ' . :f ..,;g_ ,
r- 3, m y;, ~ 4
;>w , Lu -> .g ~ a sy ?
'O
. :. . '.m Jg
. , - r_armas Figure 7. Excavation Remaining After Boat Sample Removal at 9:00.
3-27 , .~.. ,. .o p :: n : . . ;; ;.s .m. .- v ; . :. . a . e .- r,- : .e ...-e- -a
>p l
l e. 7 s# ELB0W z W ~ ,. ~ . st .
., g .z t
WELD PIPE w. l Figure 8. Excavation at 10:00. T .
-b -@
h@ ; JI: , A .5 ' ,
.5'[M',"ff .!,
y- \ J.-l; PIPE ELBOW
.$ . O a; 4,. ' d. ,
n en
,+ - Ur..
k ,
Ui[(*{#
sc ; ;$ s '
>(kjffi N3
/
5; I
. . - p >.a .
wr, .r--- l Figure 9. PT Pattern of Crack in Loop 13, Weld SW-6, [ After Grinding. 3-28 l---
l 1l1 l g 2 n 1 hi td pn ei . Dr s g t
" n s '
e t 5. by m n e e m e r u 0"
% r u
s a s " 3 e a 3 5 0 M e 4, 'l h M *' 5 x t p h e t 3s , D p " i, e 5 . D 4 , , yl ' T. T W U U , f
- o
- s n 5 l t
s u s e R g n i w o h S d l e h W e p i s_ 6 P
/
w W o S b
' l D E L
E n W o i t c u 4 S 3 1 p o o L p_ 0 1 e r u i g i F 2 1 W 0 E B P L I E P I'=
) ' EGG 1 SAFE-END i c j Figure 11. PT indication due to superficial lack of fusion at weld root. l 1 l r 3-30
Nozzle I 1
/- \
i f\ \ '\ \ $ Safe End to
, si i
\ 'l Nozzle Weld 3
s s j 's, . I I Safe End Transition ! (Bevelled) . l
;/
l l 1" Long x 1/4" Dp. 4 Sample B - Examined at BCL
' l i.
/ Sample A
_ _. i Examined at G.E. 1 . s ..\
, -i.-*.sy \\\
Pipe to Safe End Weld
\\ < FW-22 of Loop 11
'\ ,. , .
s f*\ ,
,D i OD ID Pipe i
- 1\
\
Figure 12. Schematic Showing the Location of the Safe End Weld Samples Examintd at BCL (Section View), and GE - VNC. 3-31 ,
1 0.D. SURFACE l I l SAFE-END se
, - w .. . -
TO PIPE WELD
> s._%._
Nc. : )p'% u-v t-Y _= t, !!!!! i s Figure 13. Photograph of Boat Sample showing partially opened crack, 3.5%. 3-32
l 1
'i SAFE-END w /
,~
l TO PIPE
.\
WELD - 5 l I f
- n. ,
l . t
,L 1
i l i l Fi gure 14. 0.D. surface of Section A-1 after breaking open to reveal fracture surface, 0xalic Acid Etch. 15X i l 0.D. Location of Fi g. 17
't C-l, ,/ v Location of J ' <: ds,
.f.,
~ '
Fi g. 16 .
't' Figure 15. Section A-2, 0xalic Acid Etch,15X.
l l l 3-33
l mvrw~.. ? a :~c m .pn.: - -
;- ?~ pzp-: .
- ~
x.- u 9_ ,- y
., N ..> ~; -^ ~
...- ~, .
f5 ,' , 3.? ,
,- ri
$7-U .; ,_ - [ , . ..
.,f - /'
l h., t~ . I ',, , .
., j - '
i
~._ .. 1 1 * '
, - . ,/ . . .
.g., y-
~s
.('
L, s ,;, ~ y;; ~. ? \ , 4 , ~f _ N
. . =
,, 7.
Figure 16. Lower left region of Fig.15, showing intergranular cracking. 0xalic Acid Etch,125X. l l l i
0.D. Surface "A" "B" w,,-
' ' '~'
/ ll ,
\
f _ s i . 4
't a. .
1 . I'
- u ._
! j m , ,-
- s. .
^- / ,,
l
- . - ~ %_ ,
Figure 17. Outside-initiated transgranular cracking on upper left region of Figure 15, 0xalic Acid Etch,125X. 1 i 4
= *
.(' = 4 P
,'.( 4% . ': ' ,-
.s, T'
1 % g
-v. . -s
- y. -
..w
(> , :,: - s.. . ;;.. I a, g,>.'- z - _'. r,
,. 4 ,., # 9 s. s, ,
. , ,. + .;2.; .: >. ... y.
v..,;.
.t.* s e r%j- ~ , ,
g '( g.6_ y'.',# j'i
..j. . , .
T4 *e [:* -
,.s, . -
a...,.N.,- .w,.. a_ . - ,,s. s #g U,' > a[' * * "yN4 [ , *'t _y.,,.'_, ' ' ~ ,< ?r~ %*'
,.a. ". w , . 6 - .;= ; . . ,,*;,,~~%'-s y ,
<. ~e . . ,. .- .
,g .. s . - c,
, :5 7. . n, . v y.. .:: + -' ,:::, - u :
+ - 1 ,.>.;9.: :
. . nu.
; , c +. . . - . .:e > -
~. **'w~.,
.. .ms' 3
-~ yf.';,
) s' ,s* ,
,..,.r e.f: w< 3 er l
..-a~
l
,..s.,s-.
.v.....,.,.-
n,.,~... I i, I I. m 6 Il i Figure 18. Crack arrest in weld materia,125X ' 0xalic Acid Etch. l I I I l i l! 1 l l l 3-36
*% 'N
, *=u.
wN, >
, ~ R w.
., ,. . e
,y,,
e'sw' ~ % g,,
. . ;f'.ky1 4-
- w ?,W. ,,
. - ^
y' ..,
' , .y -
. ,A-e.,.
( -;' -' - 4
- +
x..L-,rg-
- g. ,
\
. .' , ~ . v?. , / ._g
' k .1 4 %,%yy . < ,v ,
j? . v.
.,, ;.; i" .
s
.,s 3s ~ ::
[4g
. .,Q s 4
- e. \'
,v -s, ~. .'W;n. %,
/
8, .- , . . v - ,.~
>W
".4 ,
~. e t: -
7 .. m-y';%_,
, . , , ~ -
y v n,
- a. * . ' .
, g ,.A 3,
-,r a 4 ..
P -
~ , ,;
~, -'
. e l,g Q s
, t,. -
y .
,. 1 -
. t ,
a% % . - ' J t Figure 19. SEM photograph of 0.D. surface crack, 50X.
?, .. .
j, 'vi 7., . 4 -- . g.
,h.
E: l b w;
]
G.,ill ^^ k s
- Figure 20. SEM Photograph Showing Intergranular Fracture and Scaly 0xide Layer 300X.
l N
~ _.
1 i C9990 s2.5X l i Figure 21. Appearance of the Piping Sample from Nine Mile Point Reactor in the As-Received Condition. l 3-38
l l
. p :
., ' f6 N .
.'~
s w~ - l l l SEM1023 20X
, FIGURE 22. APPEARANCE OF THE SURFACE DEPOSIT FROM ! THE NINE MILE POINT PIPING SAMPLE IN THE AS-RECFIVED CONDITION.
9 3-39
' ' ~ - r- gry rt i m -im 1000t,,_ ,__I.
'"~1 i *
!/ 'FT' ~
"~
y$! :p o..n.nn"d- "! J >r I I ~' ~ i' ' "'l' '" l' I ~'
-A,- ' m" y,I ; ~ (z~ ' i q
.i i
p;d "7
.j 1 , 9
- h m --l F 1 . ,
p:;- 1 Si 37.213 i l. t.
!I . ! I i. ! 4. .{ . .h. P 0.265 ' Q*
- l' I, . .'.
{ "!. S 0.547
- i. . . ,l . . .' . i i.. i
, I.-
.. .o ' ,
..i. + 4: Cl 9.466 :a
'. '" . i n,. ..t'. ', '.
!. .ii .o
, [p-..(. .
i t. ! .t i .,- i q l. , ..p ; i ; em 0
. ". l$. .; j.n: . Ca 0.064 s
' (.. f .. e i . i..
Cr 35.567 m
- ir m
.; a H a i
' . i In m .
' .n y" q g
- j'
- i -
, ;.+k 1 ::n y! $ Q p'! $,- Fe 23.378 ?
.(9, lj f..p.
_: a.; .j.. .t . l. . J. -! J .;o d :!: .p ". qi ., Ni 2.035 A. ,
.' ,e3 .. . $. .
.h .;. i 0.395
. 1 o.. .:.. ,
- i. y.
d
,- at p. .!.' S .,, .
L, ! ,- Cu .
- ,; I;
.p .. ."4 ,
. " . ..;.p: .
a
- i. a 1. . i . n, !. , ..., i . p. !. i.i . ,.
- 3-p . .. ..t. ... . :; %. . . . . . . . . . . . . . ' ,
i 1:. i.p. ,
, e I; I. . c i p .,q- q a. ..n d' ;h - ,.:' t;l- mit t'.
'j;
. r c .- ..; j.
n f. . ft:',h- l .? .j p tu pll 'C'
'. I l:. r: ' ;, .lj %t .,; i 'n . o. i C : h; , , *
.i .:l- ni : t' 't
- n: ..o -
t
' l . ,, ,
'ti' 9; .:tIi 1. *!i p9 hij !!d. l !!q{::l:: nii
.'. .l. :. .q, ih.
I it L . r
.' U. 'o. : t .h,a.. ,n.. -. .i I i u:
.. . _. ..t. .' .I, . .
?' ...'- '
y, pl g..
.p: m;I . ., l ly: lj!! l . : ;%, hl ;;dl i'h:
'j rl f jh pi ]: II,l H !
f: ;d. *j Wi
- .i d i i c ',
' ;:; '.E n 0, . hli n;j ;J. I- h ,l tt I i O .:" ijn gj' y l i
.i
- l 9:h ' ."1; ::I M':
i; ,
- . t I .
*i o; l 'il lI : 'l I P;'
jd
'U ..;
T
-9
..t; .... 2.i.
...I d;.
. .:p:
.u.
n
. i ::. .: .1 .
3.. a. 4 a'lh ..L ;Ilf I! up .i l 3 "i j u" - I.
- h q
.i
' i .v
@ Ll'
..t t
i! hy Hj! .U+i Ol,p.; : i !Pl
'I i ii ;!!: ' p" ' ' ' ri 4 : .g.. i,. 4:j j'i. Mi 90 ih! H jI I: '. I
" . ' :.. n l.pd .: 11. lipi D io i ;ij.*- t
- in
;d': :..
':4 :l :&!pq[ ln' . 7 'y . a.g .j i!al:- I h yn :
; *.g Ut: l n
.! ,li h? lhl l l li i;d
,; D .: , !i yi hhl
'ipi *
!.L d 9 I i h.a; $".l:l qH :ii ip: [>i h p: y:s.
l p .' Mii lU il
.; O.
p Cil ..: i.:. .; i.: :h. . . .t . o! . . u o i.0 p !! o: $.1a d: .t.i :
.g
- .L r. .:... .
;:. 1 ... . . .
.b lh' !I ijf; d4 'P!
E '!E i.. ~ q' .; : . :n' d!. .. . 10 'o . p' !
. . ' h.
id . . l!. .:.. lhe
' op
'!. ' i ..I . h. 'i hI : l ! Jiji ! j':l !,
l1 I:. i
-' h.lj '. . !' I .l!j
; ll l .
. l
- t.
li
. ;fi-l 'l
' c. . -
U
. ' '. b' 'l , 16 l' 'i i '
l- i
- 'li l l li
.g, i j 4
- l
'. s(( i l';i c-I,cf' , :
w r to . !' -
- 3o. : o'!
n o. m
' . ; ! .1,1. ...
P l .i . !.c.q i n.
..I 1.s .h. m '
. .l . J
. 1 .J. t. .' , .
. .,I .ii.. .: .l i,..1. , :
h Nj '{h,
'.;". I l b !l,.!! $ h:
$ ". s $l-h Oy E5 O o
.:; !b
;, . t
- 3:-
- ;:k 4
- k. I,' I ikeN::,l.! ! I
'l,.
l
?
I g f I i { t 11!: . i: . ;!! J
- d. b. !it?
N! i
., :ig; i; $
j.. i :t;. . l
!.!l,l !ll.!
;.l-; :. 3:1;
.I ; * '.i 4 l i : .iji !. .'; , g .
'. i
(, ) ,l. l 6 6! 8 .
,1.. . . 1' 1:. :;l, ..'
l l 1 ;. p' ' i ', 91 l'
.' ..l !l ,* ,
i ',i.' ig lg . . : !. ' i ll ; j! . l t 1 g t . l .'.,g f.,. i,1.
.,l.7 il , .4l
- - .;'l. i .:
- ' .. 4 i.. I I Ii l l j l I Oll .ti, g , , ': .,
; '. ..:. up .q'h
' . ' '!; ip. 3 , i J .' I' ' Y ,;;; g; . ..q !i .j lgl. r l. : !;.l 01 :p!
p{. ...' ' . ; ;.n:;. .c .,;. .... 8 1 ,d !ltijg i q: ,. h., a l ' nfj . a.. . .
- a. . 1 . . ,, , eq -
,. i u a
.t.
- 1. ,
. o.;i.
- jp '. ;* ..c . h . . ;, . : .:1 .1 n .a j. . .. i a:.P.. UD. ..l. n i It 1%
U d;.; pI :.n .. . . . y l-n I.' ii:, '
.< il d: ni; 1P ;G l
;.h .
I. 'nl 4,li .d.p!
.:.. :.l. ,1:!
I H I -
! i
.j . n y . ...
o ,.. t o h':q. e .p l , i
. ". .a I . q. !:*
p; m -
..h ,; q;
' g!j i i p4 , .
,n p, d " i .H I ,ju
, . . h. .. ..!.,
J u,. :y ll, .
- l. .l : . '!
.a.
.a t
.t. . $ a .l .$ , .@ ...
!a 1
.l5L . .
... .i .n. L .i
. . . .. ..e
- 1 a i in .jp n,. ! n. :h ;;q iii
- o y I 4:' h.i N 6 j:i 3 ! !
- i i .
4 if T "n n: hl.
$ @. i' "rh.l! o p l [p*
id y;. 4 m
- pe-q. . .; ~F :r h !n H :1 : h jo ! l
- 7. . Ll.4 1
.] p q .i p s
.m : .1 .p ; I !!o qP .
i0g h p- +,
.h. .
. . i .: .. : .
i.W "o - - tp {.g .: pi i,:,,
. i, i . . o
- m.
- la. ie i !.
! .I .! :
l r. . ,
,c.
- . p- eq i
j..o 9 i t. 1.: n" . 1
;;p,
.y q
+
c.n
- n. "!il. :;; :
i
,n 4*
, 1
.u. j, i
i
- j .t . lj 1 -I li, oij .q: lh"
- . r:
In ep 4 n or i
"a.,
3
,. l.- % :,,gq. ib o-e >
q, . , n;. .1..!,e .i:- n-: 1j :
- m.
- n. . 2 1 n; ,,
9 h.i . e . . ,7 h li, h, . .
, .i
. li 4
. n.
- ll.
- ,.a. ! L, .:n ,. - n. o.n.
~. u
- p- .
..o. . .l i o.
, o
.e . o.- : 11 . F. .
_i..a.. .... .,4, a. . .i z. a .J n, ... i , a- n, ,g,
.:o 9 u n.
o
. i
. i .e. .
1!! i Ip: :o iU 08' t i ao. .iy a ; i ;1:0
- .- .i ;
iil W qn H: y pc n.i th ti j gp iN 'n #: . ; . qp;W : ! "..lg: y- a rp ,;b! ;l - a- f :d i l Jh T :U ; d:4 D
;f rp :s !!non @ . ! - o dE i !t i-tl{ii l .f
'.W gM'! .N; ! !!' 1 H.. L hl;l dr 4 4i I F U p t p M bjl ~!r
. ai. it id !!u h:- i:
r i.
- 7. . . - *
, op 1r is: !i..' !tlil! .
T; Pt b"N . j "v as :: q:- t lj ;al!
'I! [k. h 9 h N;i!:- M u.! f.. ;!;t; F :
i l [n! $ j[7 .p - 4 --
-h ",,T'L .
fn S
'. t $ % .
.o h d h@p uj
- 'k:1 i1 l!,,Il h" i
di.r!: %..c F-i h - m;. o.1.a. :n ;.q .a: . :!.!l . lla:. j -
- g ; i 1, , o ;! ; i q ! i t
.Jn . .: .i. u ,in ni t lalh d.h, Il di! : .- .
pL I7. . .; ., a : i G i .q L :. !
. i 1,.
n 4.. i : .. t : .. ... a.. .. , i 4, m m;p +
.daal m r o. -l 1'g: 1. 1 p :aa n a.mu n. . e. a : . n. : ; m i ai i o .jf. ia ., : i. p. :
,a , ; p ;- :, .,, ".; .u i
!!I, [, .. ]a q.
y: ! j o.: u. . ", : , ..:
;p e. , s . . ...;
4 ., ,
- i. .n i 4
v . i,I ,..c .e
.;7 , -
J o. s i.p.
- n. na on:e;,g.n :d;
- a. n f. a. a, r 1 .
m:i .m'. m, ..a o g. . d, l ;DiRg l.
. j.. .. . .
0 mn.*m e
' i
.aL n. t m :- l ih! . ll Hh tid 10 4A' .h i.h .
1.0 ENERGY (KEV) 10.20 FIGURE 23. ANALYSIS OF DEPOSITS ON THE OUTER SURFACE OF THE NINE MILE POINT REACTOR PIPE SAMPLE SHOWN IN FIGURE 22.
aa_ me - - _ _ - .. m .23_ _a__a... _--*_.a.r-e. _ e_ __ ,aw . .,_6_..A4_.-. .asw w._.. -w_a._.--- - -_. __ _. - - . _ . . - - - - -- -- 2:C
, r,- .
\
p t
- - > 'd g ?s .
rf
,s.
'( ,-r^: y{ , .
- f. :
)%. =
x s _f,y :. y f. @ , &
,; - 1 ,
< ,;s ,
i v:- Npfg -
*b 7
- L, u
.t s ' s c ' y. .,
t , ol . .. 32 ;. c
* , y '
. . i:.. .
{Q ;,y- .c-
- 2 IA 4
gegTgA7 and Bc ST948
- 50%
gE 24. APPEAR ^" p TBE CRACE \" COMDIT10H-3-Al
- -' ' ' ~ " ' ' ' '
r'
- r. A/
, J L.a a: -
w SEM1034 600X FIGURE 25. APPEARANCE OF THE CORROSION PRODUCT OBSERVED INSIDE THE CRACK. 3-42
~ T T un 1: rp: 7'h . " n, 7 I %R ' F M"" ' " " - -
2000 '
!' - )' l ' '" m'>r'~1 "o ;IF !: ..! di; W SE "q-QUANTITATIVE A lI"M" ' ' SfS"
. l ' /6- a. i 4, .
1 1
! t .'. !-
@nm.
W. . .. d, s f >d> . n:. + g "[' i
- f .i :i,. l- p ;, ,
f M S 8.721 r j: l . . C1 29.531 l .' I t l 0 ' 1~ .. f L 0 .+. .l..L l .1 p 273
.a, 1 .i, ..- '. . . 4.m .
% r i
. Cr 2^. : i ."...
.. .!f.
Fe 26.933
! ! ..: ?: F 7 ....!!U p- @: 41;
~! ! .
T elj:, .p; y. .. 3
! i:
ia i Ni 1.565 d liji
; n '
.g .n :s. k,; 4 q .:[. g. ;;5
,.,i 0i b. .bi . K 5.712 .0 p N.
p]g i >.n , i i h l . 3.979 i 1 p i! i lh, h !! [fi! ji! r g
' Ca e] i i .. h y yp. h f. (. ..h q g' :.l. 3
... .l., '
r i !' ..,,,o n ,;,- .on.y' t ;.;
'r a j ., ,j .. .. .
!. i
- i. .:,- ...
+
- n. . .
r aJ 3 1: e'pon , .
': e ip. .,
,3i.
'l ...h i
.; i
! 4
! ;d; 0
lp:i e I li" nL ipl
- il 3- ul i o-ii. I i
;{
i j; g,.! i l', :hi
,i!.
[e: h;
- )nt. pl
...p :.;.
.? y;. Li: .! ,
.r.
! : ;4j ! l t; j it la
! j ik! I,: g].n..
i O' !. .
} pl
- ,l
' I ;, : i il
.' p :. !l1 ;, ji i t
': , lel; h,
ai pn Jj; I 4;:'] J h } a d:
.p.
n d
.c3- .:.j n
p;..
- l- -
[, ',1 ,t j ;di :: I h; ,,
'l .J. U p; : , I t :
l
' , . jl*: l ,y j, i!
q .i, . j, :.;; j4 p.[; I l' ,n 4, lh. lj l j j 1 l l l c a ,n y- ' ..h'
- aj; al l ni t
- i. 4; ir . .: . 1.. ; : 1
'am dh
- 4 4
.J I- '
- T . ' -
. . . . n, :.
d- ..'i l .t; Ni l i .' [1 - ,:l. ni f .[: ,p t l..; l i II :t e q] jr. I' : i i o p - i .: -
- ,; i..
,4 , 'j. :i t u'l p'qi ' 41 l ,. j . 'a ; .l '
ni ;! .y I- di' ! ! . l.l l f; H. ji
..i
. n. . 1 j .; .i !: .q l ,:
. ; pp l:i i h .. i
.I n;. ;ji' i.-
.r.
n i, , ,l p. i:. .y ; j q' .,
- p. q. . ,li
.c 4: . :g,. .o .P ,
h,;: gu '. p :; - :q, }q i. I
!,n p j
, , . . FL ;qJ '
- m. a ,
d, i e ; d: ;i ...: h: 1 j n .d. d., l l j :/ 2 . n.. j;. i . P'
-: j ab @. ;;:. ;;pl .h,., , .
4: I ,I ;, , l'
, l e .
- . .n .., . . '
i i 1 i . I I ,
- .h I:;;-
ig +
.ig p! I .
pl j,, i yo l! , ' :gil 9 #;, ; l' :- l
. ; f' t i 1 .g[! l g lo'4: #
.!, 2 #
6 ' .h -
,., 'e :., f f J. i ! l
{... i i
;; i
,p.r ..
' llg..l di ' lo lj j l:
,1 ag i *
,"- pg , !,
- 4" H p. ;
ln p, ;pl l, p i ; . l
- en ; ,!D: l4; :;.
o ! se @-
,j- m -
d - Nl
. j
- 1. ; .- !p M !
h.8 .' HI U ". " f lUl h};l t l i
.: ' I d 'f d;
- p' t r4 L .o '* y;: t 'I E: t: tiJ t IN 'i b '. . i i f[";{ l i [
L," ! . .
, ' y. { : *.,
? l':: .h- . :! @h"
. .' , dr*,y jq h '.I1, f -l; j l I . I ld! , l!. ]9 !i *i'
' li .j.i
- U. ,
r*!r p pi. .
"I. .! r 1' :; .l, y in; .ij ,[; d' ;i ho l . l i;n ll i
iN:l l i.lN j :i ,
.I I lin . ..lj .
f p- 'L . .hr *:i.. sp .i .
't . l l i ,;
- .i, i.
p; ,
, e. nU lp h. i i.
.- ' i p 6' 1 i I
.dh
. c. .. h; !.ri /> ;.,,'n :
..- i,1
.l. :d'
- . , .P.
*i h h.
- o. a l... 1.s
. . a. .* 's h'
. ! , .U. l . .... ?di
,. .. . J i j
.M .
(.4
- ',' "I Hij j y;. @i, i
.i .
H i *
.' o y !! 4;; !i Pil ' . . : h* ! p ! ji ql;' d; [j'i p,j f i , .. . .
pq blj ! ph y
- ;- ,$.. ::p .i' d :;; . 'p p [ (.
in: pi ;p j
- h. p. ; d! h'l h: 1 .:
- p. [a.y
$a3 J- <
.e i 3' ,n-f ..h ,
,n, e I, -
I . i,. q . .,
. +
Je ~
.J~ . 1. . ..a. #
... , , . '* H
't- !
d u ...u.- .
..i...:.- , .
- 4. :. :c; h.u a d'h' 4'
u '4'9 p' .
.::; ']' ; !'H- !
'Jh'm~ c: '[. h, '
..l- s. i'. m. *! a
- ,, ' H ,
_:n h' o
. ': 1; O i '
". , : ' , n:: : np .;:
7
;r . 'I ,
o u
- f. I ' ,
- ": nt
..i. [.. a ia q ;' ., ;
L,: ail h , ll
.j. ,- ; d':!v
.m .!:. d; !,- Illi ly e. e ' hl
'. t - ' t: ; l ;. , l
. 8,;
,i t
.d; !! !I ' l 1
; j- i e :i'3i ?. a ' o;, i. j i, j h, :.. ' j .
I I it,i
.' oc ni l: i .
9i y" n. ;;,l I. ep g! a [.!p be,
~,+ 1' c :: -
j j
! t m : ;lp r. :m r "t
I tr i .: i Ic
-a .
-. 5 L.- .
, q.
- i
- p'l.a lp ll.i ..
.:. i l"
.n. 4 ]h. y j i,9 j N i ! : I i
l g, j l';. i .i. .o n. a : :,1; a; i.: 01
. ui d: n i .' -l;. p l::
4 . .
- .i.A '
.c
~~
' :, j !
1: "!! ;i Pi ll ; I p Uh T. i j' 'ii ". S: n;!
, .j p jp
' d.,::i;:
i
.a . .:ji o- :
ay t o.L- ,e [%! i h a. .m .u a,! .J p; :!i c,a 2 .a
.,. , a , . .
q; ,. g ;- p- po
- &.h
. ' q : :. ,q' l p
; .q t , I in a, p, 4
. ; : - 11 . - l, -
- l: l- J lo q:r; i ; p. .i h ;. i
.H d o, "i i ; j ,
),
i
. A . n. i
,pn ...y.-
ji p . . . . e . 'i . . mI l r ,
; L; '
t'
- .1
~ 7 pp is, p
J m J driq !.r! L Hj.! Ie ij L Uh l, ! !4 'l ll:u p/ j'M: i , li ;
"4 #m
@4(i fh j !4 f e! I ;d
'. 3 .r ih HH '
fi ;. ; ': o' .p::4'
.ll 'f in i id M lli i
h h - i ! lt !'
;i, e je lgp i. l.p!
- tA Up ' i A
j; ld ."d .n,
- L - l'a ih in i j
- y. ); i ju ! [,. Ud "o "p!l
. i
! 7...a l, j,.. 1 pI p,.ih> :n ld. i., p1 i i ca l l
r i:r ' , : :i a, .e; i- , ll g .g 1 + c 4: i nu
'! ...i d r
q Ja . .d o p i t.n e q io i l l
.y y g /' h
'. /'
1 [3 ". (. l,.I
.;;. ; i) 9;q j j 'y: , ,. :4! ,,!
g(:,j:"h: i, d fu ,.
"', l4
'.i M I l T !I l h.' Fl
'h- ni '1 i ih-
,A, p ,i. 1 - -
q-
.a. n: .n
.o a '
!L
;~ n n a. -
o h 7 .. h. in '111 1
+
.m I I: I , li .
'l i h ~
3 p; 4 ti n'a
.: : ad 4: o . :'p .p- d.
p.
- h. . +. u 'p ' .I, !;g t i U;p ' h Hj P ; j .a ,i i' i i d;, h' I li di, . 1. l7. '. ... I.
i i i ' Lp r !" J, He
.' . 1 .lt I P.4L:I. bJ h'l h' lu a . . .m.:. . I,I t.i ,' 1. . 3 F..
i .
.i t .i ,. n i op. ha.
" . s.
r.: i.
. a. ..o .n., u. .' : .r.r' h....! . .=:l ..i.
a .
'4n. i; j li; i ll pl h; ijh % ,n!Puy ! !p
... 7
.i>l jl ..;m :.. j dj t
i'i :;;J 1 '! il. ;:.
- ii. pH O 1,. Eb.. HJ .h{.; l >
, 4 o M N U ,+ - !!1 -
e a, 1
, rn J c ; ..
h . :ll0.
- i a ; su o i ....e;, ,. r .
J,
,j i '
i4 :.l ip i @. . !,i i '
@n .'l !jl nl p
- i. :q1 ip' ;m in ,f " , " '
I hh .. 4p 4U @p
~ ; . i; m
.. !' !Q n.
!,j lI,
- u. .a.. l.: I::
u N g . I:!d.
- p u : by ll
}l .l au :,n a .- .:.. :.n .
dp - +
;W:je n..
'e,
.. ~
gg ]' i, i gn; ,a gij j g ['y ng 9 g{g" j 4' p m , du, it y i4 g.g il I p. . i: - yJ % ,ib, p. 4.f.nNIRg. .n a,mta Ye.(KCy l. iip pm. ,a g; ml, ;, q. .
,y ln, w
n mi !!nj3 !p i A ig: an J 4 0 y : 3,3 l. n ! ..h . 1.0 FIGURE 26. X-RAY ANALYSIS OF THE CORR 0SION PRODUCT PARTICLE SHOWN IN FIGURE 25.
~][0 ~
es. e
e n f
*f1 h~_
p . I k w s, j w b f
=
*~' 4 1,w ,4.81 420X FIGURE 27. APPEARANCE OF THE CORROSION PRODUCT OBSERVED INSIDE THE CRACK. .
I 3-44
1000 & ~~ r i : F i l :i - l L ;i' F ; '. l l (#e. f l '"
.'l i l' N. g
; i s'
5 2.573 . l . .' . l i i ; I i - l
, i j
i. s'" Cl 71.130 I
- i. . .
,! ' l l .
; K ') . 255 .
9' i , i F i i ; i, i, i ' ei
~
- n. .
Cr 9.853 Ii - i l l 1
.Jr !' Fe 15.844 -
i i .i ' l a l . l
' : 1 ' ,. .. p %e ,. Ni 0.393 l . ! ,. l. t ;l '{. is t
t 1 .I 1 . p'
.a
,;1 ,i ,
3.;j'! q
~
I ' i ll t 'i' *l-.
; ,i t
. A i
l f' f ,l., ..!... ..,'a ..i; 4 '*!
.p ~ gh b
tt i
*.j !?:
~'" ~ '
t,,, 4 til; If ;i ,i', l ,-. ;; [ jf ; ;. .]i' ! !'{ ; - f-l i all1 g, :lL; l. q:.' i .a,; j s : i+ 1
..e
,i
*7 ,iql . g f;!
l, i.c ';l: , l ,, y :i l io p'n j ,g: '
~
- 3. . i : .. . ., p, , ,. .
. ..q ij'.! .i . l .,lU tIl 1'1. :...i .; ,it.1 l .i . l'! i'1;r 11 a I - --l ,
,.i.
' 'i- ;l: pa *
! 1'e -
.i.
li:
+
**i ,,.*
- m n-- ' "
e .c!; n
;- i:.-
q : 0, n a ,i
.i
; ;l; ,
'o. -
.,l.
-;: 'o p. '.':l: .4 .i n. 1 ;
.i...
e I i i- ' t: .
, l In!! ';p
;t i . j:' .; ., I i e .t ; ! . ; t' ;d', ;j lt'l i "!: r; r ,:i. '!F ;h. :: !
i
' . .h,:: ..
e. l' j '. e [. J, . , ~. p .j..
~
i ";
. .c ., ,
;. . . j' l ",. ,1' j*:::a!
. . I . .: .., .
- ;h. n
'.l- . - cl
". c;j. ,e
. , i...
. :; .e 1.. 3. , ' a q: .;;. .,' Il - -
lei c' OU ':. i I
.i.- m. ; i
':$r- :.;p!
i ^ !
- b. y .I p !4
... I:..- e '
6
]j l
. I- T i -
.i[ q O' .;j j- :!;, . ,f" dt. a p. o p ll 'i:, j "li' %
[ i
;pp i
.i.. .u. @. .o .. :.:
';- li, ,,iees
.i. . . . . g.
; t.* ;-
i.. . p. .t.. .
.] .gi e
;i-
,y-
; 6 ,,
r af ;.... Il
-t
! i g . , i'+ a t ;-
o o , l...
- i. ,,
.l,
; . ~,
.. , .l
- h' : . .
p,: r. . .,.*i ,
- li ' p ,,
ni. p i h, p!. t-j.. 4.i l.
,.. i l,: jl. ls:, .t p. . ,j, i,-
- l. ..- j , f. .
l.--l- l
, i i i' : .
i i i. i . ' . i. .,.
- l ,1.. l,ij' .*e i.
il. ' d g- [, ..{ .lJ. ' t 1- -6
- j
" o} . i p'
o 1 a . - r i' c ,; , , , ;; .ji
'4- :i : .h ,
.. h ;l ' (.l j' q[ ;,, g 7 o, .
n v) g .. :h.
. m. !
,,: -o: y :..
.,^
n :: - il n i pj t,i , f.g e it ! ! ll p,lrr U tp i i n,. i, U 'h;;
'i m-
' f. p' t
L
- e. .l m.li Ud n r ! :D ,!' s .l b ,,
.c ~:i :L' d !,h. l.ti, : ::!'
2' -l l L p '
- h. -
,il y a i .'
. g .
r ]' a n o na y . m- . p' ' p c
. ,p i 4.,: .,
. .j .ltl o
u i
- . vi,: ,a i
. nl
- i.
' 1. ' ,
i; 4 ;. ' i
, .'I' H f' ; , M @t!;p? -
h it '
, ',:! 'l ap
,. .d i. ;$ Lt n i p iy i 1 L Fit.i h. i ly i
L }d i... ' ;;ili - I !.l' :l- . l a: p- i ;- ' hi i 'i c 'n- '-
-l.
+ - -
.. p.. ..} Aj 1 ;.. a .i .l la: ! . . y jo:
4
.I. ' i
,i'. :.t i. .:. ,i p-
' o -
gp;- j.l 3 a
.q.. . t;. ; -
- t-
;l , .i i
.g y q' q .ap y, ;;p p p q ; i ;
jl . p i; j g Q', , 7:: l jl i 1 ; ! i. ) I c! i.< L .p a:l p;( ;F L.i ,.t. d nF % p!l . ; V: '.;, Pi T [' L :!!.. F. i : i p,i i,
- l ir e- -p;- jQ.e..
m .c i in
- ii l i-
. n, e.;. n. #,. .
.h .t.
p i. , i
-i
'[
.. . 4 l
; i ! ] ;': +* . r;! i j
.;! h q!', .pl dj :l;, p lp pi f.:
l' i d M ^ o!
,. . S .
-] -.
.J-4 i l
- i h,b.1 .!
+'
ii
"" ,L!'
+
!!-l lal 'l.
i lj Q! t li. :p lii o; i.i - i o.Uj..p a ':d
){
, ; i
- i. -
i .i- it. t . o. 4.l .- a 4 . q, .t. q
.!. .l ~
t~ . , ..L ...
,gg.. g l{
- l. y h p.- g n.
p p
; e.
q ; n, j
. . . ni 4,
7 i ,
.t j :" l:
j , ,. .t i. ' :: . Ui . ji W. a. IW i:'. i - 4,
! ;P i.
, : ,~
p ,
.! . , . . . i-
]
i l 1 , dj 'g g.I 'ijl de l; li! 7,: .i: !oln i.n 1I '
,' V a P 4 d;~ ', L p~:
- i. W:lpb b'
. W. 4 m o ;p n
.j., L q m 1p, o; ..llr
; ., i t ; i ; n
.o p. : i g. is to .c. m '- ,,a -
A:y !@n -f ql: H- .jo" ,' i
]~
't ,0
, 7 ,.
, i
- '1 io 40 l 44 lj l' '!, i:. 1:; l
.% g-l hl; ji [ 4
! [,. il ti;. .
.] .
.bQ 4 , [j
! .k.
b,
!l. h l [ , Y W'
[{
, i o
i i [ c: ]
.l 1;:,i
,i llM -l - m e, n.
. - i l . . , . .:
4 ... Ny t N b, .
,i 0
h. pL "I"i ."
' "4" j *K i" ."
'd. i :.
6h
" " '! p" '"' -. "'~
i- % d : :.' . i, a' "-j * ,!?
+g.l q l" ";"l
.: Et ER3Y '(KE{h $
ti p"
;j , ' jj, 4 % ,:j
'b d .
jj, : M e, Shj,jp i!F hvdMM 1.0 FIGURE 28. X-RAY ANALYSIS OF CORROSION PRODUCT PARTICLE SHOWN IN FIG. 27. 10.20
i 1 1 [< E e<."
'...i n <-
g* .g f n.
-e .
j 3 h, k 85X FIGURE 29. SEM FRACT0 GRAPH OF THE NINE MILE POINT SAFE-END SAMPLE SHOWING THE INTERGRANULAR FRACTURE 4 MORPHOLOGY. 3-46
l ELB0W L -l IE ' SAFE-END , l l A423 i FIGURE 30. CUTTING DIAGRAM 0F A PORTION OF THE LOOP 15 SUCTION SAFE-END-to-ELB0W WELD.' SECTION WAS LOCATED NEAR 7 0' CLOCK. ARROWS INDICATE SURFACE THAT WAS EXAMINED.
)
4 3-47
*=4, ,@
.O p' .' # ."
A m -
. -r \;
q 'f ELB0W l
~ +-D:#; ; . .I 4- ,
; gs
.a. s c ,y ,
h' # ', tg y )
.: ,) hf - ,
g ig-" - ' f * !;
} ; 4 +,_ )n, '
li 24J 'j i i h .@jpF
, g i '!, 4 .i? G } ?; .
I.i - $ x, *Y .- .27 I qli~, %$ 'S PIPE 1 A424 FIGURE 31. CUTTING DIAGRAM OF A PORTION OF THE LOOP 15 SUCTION SAFE-END-ELB0W-TO-PIPE WELD. SECTION WAS LOCATED AT 12 0' CLOCK. ARROWS INDICATE SURFACE THAT WAS EXAMINED. i e l 3-48 l l
ELBOW SAFE-END
. .- #: c y ~ _ . .
-- $ 4'
' , _ . ,-g.., '
s.',, ..,% ,4 - <
.g. . .
,..,ft . . . ,
.4 Mg .., A '"b . j- .'-- h'(.
~
*[ 7 g y n ". w y'. * : . , -?.
- l3Qg; ...-gM - * . ,
_~ . 4. y:p.g
- f.
m; - ,
.s , g z__
, g. ,.
q3.-. .-.
' [ - ME,3' . ' i,f.) .. . ~?..
. 4
. ,$)
- ' %. 'X:l n;
-Ny . ..A m' .. . . ? .l ' _ ~' .
r
.:y. .-&:p. w :;u. ,'<._ g44 f h. -~g. - .p.g)y.n,-
y
;,n
. ': .;f 3 s .y 6.;.9.. -*
=' = " ey.y/,.g .
.= .t .
- f. . _ _ , pf,ga -
,j . p,-:. ;; , ';;7.-y ;<a, ,y.4,Ly
- .m ~.: : . . .
4,. _ :- - g -x ...: .r g , h.N
~~ '
h 4.[.";a.hYs-di:,.h'- h,. . i[..k
. -. : s ...h,. m:, . ., e-s : +a.n
. .. .. , . :: , . ,_ ; c I,'Y " ., ;;*: ; :- S w
I'.f '" 6l. . T'.z _'. -__ h_ ,.. . MA dQ1'.(%dk . . 9.' ' / V.W . -
/w- -
. - - ~ - .
3.?O,:i ~N .':w ..:Q.;;ik -
&pgg ):.fc.. : : 1 p .n x. a '. o ^, , - ,
&iyjgy.y; ggsty.% WW.p -
... ,. m q q. qk%.3]u . e.*.
w, A '
= . .. n. . 4 ~ 5 :p m ,;..:; ur -
a . qw g.g s. v :
~ , . . . . -
gl;z.}g. g l. . <'ffy i. . . ' ?.f. d. ' .ljf;.Q_ , ' i
&.j) hh. _S h4 < f,.l s, u .
9K371 4X 1 FIGURE 32. SAMPLE A4 SHOWING THE CRACK IN THE SAFE-END. THE SAMPLE WAS GROUND THROUGH THE PIPE WALL THICKNESS.
X
,: * ' T. 9 l.'. A _e ..
i . e *
.* . .. 6* )k'
._ g -
; ,. , , 9, 8 ^ _- / A* .s
. .O'm.g4
'. v.
h* : y ., , ' '
'y . y. . .
':.,.$$p=,,f,.
.Y ms;, '*
~ - ,.
- - . y_ . : -
E 4.-s a.
;, gj., g,;=.
R
. 4 .
N ' .
.- s E
.u
, _ ; .~ _
xY . W
.'.r.
, +
.. I" .* ;
.j. .-
. S
., ;** */
K r s 1 . . ;< . _
.~- C A
R
- nc.,e. y, g. '
._ L
- C
". y . w, x.:
*
- l . .,
O
.. ?' .- ..;....-
N
." '! 3'
? ;~,., a 7 .. . .
. .~
,g
- D
.O . \ ,' . ,
[.. ;3 . .. -- L E _: _ . W
~
.v
.* 4 W
i.: .'
~~-L. - O
- . ~
* - . - .. B
.. . ;- .. - L
.,, y E
- ^ ' -
O
' A~ .
. - T
'q/- D
- =
N 4 . . - E t .. . -
- ~ .
^
E F
. ; T ,. .
. .__. . . i> '3 y^ A S
%~ ,,
, l ' .. -
E
'p> ;
7 s. Q w# . [. H T g'y * 'f. H
- . p,Q,.s, 3 ,w5 G
.~
;';, ~ f' '
U
;jrwu&'. O y)* .
'a ,
I
, 4 a,_ .
R H
'%I Qi. ..h
- T
^ q: e.,
. /y ._
)
, ,%'% N O
s, ?y . , < ; 5 I t.~ n.' e- ,'-
},
t w- 7 2- ' g.h' a .xn,. y 0 .Q, . TM;f-.
'Q ~
.y T
C E S T y .f S
*p; . f v. , t . i.hj.
S OE
",h. .
- s
*s?.
t. s
, -#r T
-, > Q,p- . y. 4
<( .y'- .
RL CP AA M r.,*- Qk,.. <. -. T' I ,.
- 3 S
.\ G t% s. , e, v -
NS 2 *Q{" t I I p>>,. ,-
' WH n .' .> OT H
SN
' . - ) I w a.W .' , 4 AD
- E EV
<;.. LR PE
- v. , MS AB
. t . SO w.a.
.* . j 3
- 3 W .
"" E R
0 . U B G L . - . I E ' , F
, - * ~
. . 9 6
3
. K
. 9 go
ELB0W PIPE
* +
U .%."1,_E W f , R 's yf;. , Ns-- ;. -
.s
; p f-hd4.if.. .a
.- s. r . n.e..
+
1 e
- e. .. .
, y1 .
?L
;;.s. };
+
. s .
a4 '
- m. . ., Qf) .h, \ 2, 4
- m. . :. .,~ .g., .-
, s, e.
, , ,i y .- _ .s . .
. *: <g . , ,j . R, . ' . ' ';: . . ?; g . '. '-
g . . * .
$Q $-sD.,.\' : ~ Q +y:.' 'r b' W.f ' ' ' c , ,
f*.g . > ,, . . . ,
... - A. .._ . :, -f4 i
! _'...:y. . p, z e n.
f[;}n.
- s. ; -
., ,,. p. m ..
A,
- * * . k
~,. .
^
;; f _. & 7,g g [,f.. ul~
_M Q ]+ j f 8-a _f } ln[:[;i-f _y..y.-f f
,..r.w.c_ w m.w_.
e, p . fe , l '. [m, .x,~:- ::. .;... e. v.x..
. s.a. - .. f g
.\ \ . '
.. i';;, .
, f .t, :n
~ A p h,%a , M .
4.' <y
,,(
f . ...'s., fs~ 7 '. 5.[ [ q',
. . ,f.'
o .
-y ._ g;
.*yw . / , ' . , , '-
g
.<p,.,. i,h '.
t'
~,
Q j ~ ,yI IN . '.
. _ p n'
'\ , -.Q{f, ,p Q% . . :$*y: . lj j, .Sj : *% - e
'a .
. . f.[t k * ~' :,.. [;;- ~ ' -
J/ q. Ii [# h '. [ ,.5 . ,. [ j $ ~ <,. ^. 4 Y;;i. . ~. 0 ;^
., . . L .% q r
, l ~ ^ f fll,t y ,. ; y g @...
.l '.?
A ;; j :_. lip &-[.?_ % ~ @, b y1
/ ..' .v:.i.z+Lf.::'k.P..g' 7 .
Q: 5
,n,;.. h y g{ , & g ~;
.&(:..fg*;g n - "^~dk':
b.
. :nu.f 2
fQ., n:l <. :~.
..[y
[.,
.. .f,;/. .g&.. ,. .?,3 ^. . , _
w g.. ,i. h:.'.y:,R[jh&y.
- w. , zp. . p
.. c ,_ .
w, n . . .u. . .
~
- m., -
. ; WN,\y ,..ecC y ??. ?,.., Q il..':!.[. L. $.-}x.Q _*', %
~ fl.lly. ,x L ~. .q. , , t g.y. .. _ ; .x -[W'5..w
~
s -i ...~., h85.s,: sr 5.h.h
. .r x, e%. p.
.S: Q, . . x
;.tig ,
-+. wp :
, .. . - ; .1 .,.g.g ,
$m h.h.?h? k.h.. . e ..: .:)s-~ ~l SY w3.,A(%;r.
- 4. v .~ , t. . ' .,, .e y n,.u.3; t.
. ;s n.,...< .
y :. v. 2go., 2,,. -rj g. , g. .:. c...z;. c.g c
..,c, .w ' , ..
.: . .m .
4.< .,.q,,
. j
.. \ tf
%.h%rk,%'.yj$u),-..y;- ek . y.
3;c%.n.q.~N:.,M;I%,%. % %, y.;z,,,. %.. . W,: e. q- a
., . _ g . . ,, ; :y.v. . , . _.
- w. . ; ; +:. ,,$ : _. +: :-,. o: - .
..-n
.. ~. .
. y, :. ,, ~-
~
- b. k f
,f.:..* f~ f,,
.Q .y . 2%.a
. . y. .
.hl.f.. . _h].[. m. Y &k;l. :fet, [ya : :f. , ,
-l_ ..
- t n.J1M: ; . ,yd ..y gy;y _,
cq.,p gg s.j gy:_ .'.,.jc .v ,.
.L. .
4
. .g i ,.;. p g g' y ; g g 4 -a p + .
- e. . ~
2._.;e,.4;..r,,;R.g>..j , ;., 1_. $; ,,. :x; . m~ , ; g' ,.j.a_. , ;., ; j + + .2. t ,
-' .' a.; :q y .; .i .g . - Kw
- n .. g 7 m.
n' ' .. ;.
'g-j.
9K372 FIGURE 34. SAMPLE B1 SHOWING A SHORT CRACK NEAR THE WELD ROOT ON THE PIPE SIDE OF THE WELD.
--- . - - - - - - - - - __ - - - - - , _ ^ _
, l 1) l X
- N . 4 h::.
- c ^ - '
J
.e Q 1:-
eg '
,d. $
.~-
- W O B n.#.
- m. :#;
D L L E _sptp - E W E E
-1 . '., s H
- T v;'(
9
'9
, O _
T _
- .i .s E _
D I _
.
- S _
~ - W -
k: O v . E B L A f.* , - e E H rg T _
- q. y
~
.e N
;j
~
. .p - O E
'n ,: .
-j s- N I
L g eg
.%t. .
O N Wr4,g I S
' s. s
' U F
y; 3.3 .. E
- H
{ -s T (
. s R
A E g ,, g n y'* yk$wmg4p,- N v
*i f ,
K f
.{
a ,* a e C
- A
*I j v I w. *:
o
- R C
' - e A
,+
g%:y g j N(feg"pkd fp, G g
. - i y -
N y 4: I W O nyse H
., S a , 2 B
. j
'., E L
' .' P l' M
/ A S
r4 $ , y ' ,-
? -
5 D. 3
/., (I .
E R U w . G I d . F er 3
. . 7 3
K d, . - .
.h 9
$i
,=
.. . .. ... .n ._ .q .
.g ..
. 4 TM. M. r,6 M ' ' .gF. .p;.
. .7 . . J -j'984 ^^. : iq ;'.g
- rc- *
...t'A
.- m'a ' ' M(ar . . , .. , ,
- gee c .v n..... y y b ; c w r < y ,. (^ -
' y p ,':.. . .
.:' '. ? . f,.f.f : . . R .. . h. ..u:
; m.,h ' .;& .sy 4$RQ,.. sq T
. _4 -ni p;,pJ 4...
-. .g.<.f r; .f e.m L.
y g g . 344:hs p.a
~t!l.;:.h _:g. . .g. ki hN g .
t . fhh . [f , [: I
; n n:,.;;;; m e.g.annu .. y 3 .- . . .;.y 3 +m;..: .
3 y_y .. . ; ;;. e4 - m t. -} .o P ,y..,.
...a.t :, a .:c . :.3.ev,csgw 'v .. : ..
n=.f.; ; .w ~.w ,
. ,- ] y v .u ~ <,o. g?. . , f,;.-
~ 9,.
. .,. j. ...:.. .:.e. m..g : ;.., _: m. , 1 . . :2. . u.
.- ..-.. y,.. . .;.;s..;. c . ;n
- c. . ...v.,.. .
, _ . , . , . y. ..
~
. $' . N' 7 1 .
a
.; . .yu y 2&,f.f &.g;. .4;f.w.p.pgyfL; :+ -
3;, _
- k.
- p Q y ; y . , y ,. f.p, T e ?:; , ms_ :" ; n. .. ,..
. x
. v .n : .
3
- J. b . ~ s. :
3 : y. n._v ' y: 5 43 f ~./pp .; q .. ' ] ; Q j,. ,f.t R .ti.;:;. .* .'5 g :z, h...y g :.:.2% g. . ;;. y - . , .-
.g .
;y ' [y.
+ ,, g..ag . 3. y,...g.
w 's cf. _ w
,~. - .. v o
- m. .. u.. e. ..: .p&.Fr w .a
- c. a 3.a s .u. g.. u,, y w 4'" ? -~$y 74 ...,: ... p:y:.g.
;.: ; +. 3 ,g. r'.~. ,.:..; ; :;.$i Oy,. o, M , f. w .. . ?-
q' M}.j .f.~%,, .. j..
..Y Q L . Q .f. M . Ha v' {- f;;k'* 5g sq ,l. j .t p. .g 'f $m W g g w / y ' ,. p a n +N}' y q
- $; n,n. p,$p p y. ggh4ss g kn,y .;
..e v w-a n,g oO.: r.>.yu e v4sn..+ .y; g+ .. -y r , .
. g,. m.:.7 w.s , 4.,m.. . gV ,n:f. a .- : /w x
.....x.. u 4..
.. . w. g . m n >: p d'- ).
R.. . . .Q .r. r,,.y p .~ . !.... e
.9
- .. g . .. , .
- c. .1..s.. e .
z.. 5 .. , *
;Y f.
.h:.l?f,. ,h'g.,y
[,h _. '* * . .l k ^
- . . Q. , ^ :
,..,f .
- l k. . b- c. < .. -e . ,
. . ,....,j,,.....p.,
h [b?' . . J,, , . c %.
.....(. ..
. , s
,4 , , ,
., . , , . )
~
. j,. '$, . . .
~
4
.*./ . - - -
- . . .,;..,. ,. 2 9 K374 4X FIGURE 36. SAMPLE B-3 SHOWING A CROSS SECTION OF THE ELBOW SEAM WELD.
F l l l l
.'r'
~ * $
ELBOW W~ d
..; u .,,
vi . N s.
<T 'x 3';r i
) 2 3 4' i i 4 Fi gure 37. Penetrant examined - Boat Sample C, removed from
- weld SW-6 of Loop 13 i
1 fj 3[k.h$i@E " I;! I r
,- -Q[
^~ ..
%.27-w;g+-}
, .. .~:s ', .
f *%. 3 i i a 3 . '
' I
;3-
- 3, a
# P 50X Figure 38. Etched Crack on elbow side - Boat Sample C.
t 3-54
D k
'\
- 9 r
i ~) 4 9
~.
\
b\ s 4 f
/
e ~.,
; 'Ns
,j'
/
9 e s s. f $
, c \
9\ @ T o.ttcweQgodtcgG9
,to * * * ' ce* y 90 sio*-
4 3-55
, s,...p y,.;o .,, Sra%m. ./, ... ;,.
, :r.. w$,. v ;;-
. i., ~ ,.-c.. .
,. r ,..s
.a, v,:-
c
,. - m.
.),
.,m u, ,7 ) v.. O#., s ,>, $s' . : .
s
.: 3'
. .:w. , p s 4, ,
ac .. g<. , y
$ys
.h-Y.ihh'i'N.
a;, v.n.. , ;
. . '$ b,..?&.Y!^.*I: .y &., e h f5; f
- j k bl
# m.n h : y. f ,ycM N est f' s :.'*S'e . **'*f.# . , = .
" ~ '- "
.s. %2.Mpj ; ,
..y<.
. n. M+ .,
, %'u s c- ' .
^:g":,
.. o .
?
],. , f -
f _.M r y. . s
+. :..4. .. ,
- 4..
weeG _ ~_. ~ .- Figure 40. Etched view of crack on pipe side. 33X
. .. y .w
..s s.. .*
* . . . , . . *y r
. . .a .
. ., :.(. .> . g( .;
". '. ,3. ;'.gi..: ;,. ..,k,g-)
.. ,. q. .. .
s - -
,..J, ,,
c .,,4 ..s.,,
.., ;- *., *. .+
- . i,
- ? . c . *.s .,
n
'~
c.
.~ .
.'*: .'\ '
+ ..' w < , ,
.. ' ~ . w42*.? .'
b,a..!,te ,.M m ,._:f*.-c. P. - 5:' .. .. . @
- p. . N ,- ;*,' %, 3:r-[
., y, , ,
.*,e- . .: a .- *
. ; % .. ~.
.ew f .p, c,%;;;d% .%...*~
.?; -
.. y
':~,
?
'. 3 'g.'v
. u '. . . ,..$o '7 L - g'.,:.y
., q ~ ~ 's l, %.~ h %.
'....... . . . . . '~ %..w
. 1. s. - x.
- s. . .s s... ... .
..i.
..t
. . n i.,.*-:.s
*.J. .#um
. y.
.~..,,...m .h.s-
. s a.
s
. ., i ' .. v. ..
N..-. g.v . . '. ;
'. v ,.=g eN
.IM.
s.
..... m
. ,< . .,ne
.. . s.
s : +, .
.' ,..~....-+..t 1
<.s. ae
. >.<. . w.t, . i. .
-lk, b.,[ ,* , , s - .
..[
Fi gure 41. Etched detail at bottom of crack. 250X 3-56 , 1 1
~
2 8 4 Figure 42. Boat Sample D removed from SW-6 of Loop 13.
.iql~
,/.
1
/ -l
~
Figure 43. Etched view of Sample D, containing 33X portion of fracture, and adjacent crack. 3-57
. is -
- 77. ~ ~)
** N- ... ,
x ..- : L 1; 4 . z, s l.
,88, ,
N, n I
. - . 1> ,
./
I N \^
. j
,1 .. s x "
, . . v.
' 1 -
f- . n.. -
\ .ss s\ t-m , i l
, \ s - .,
\.. - /(, s . ( -
) N;
. s ,
s ~
's
'a,, l
% k~\*i N / ..
Figure 44. Etched view of Sample D containing 128X beginning of crack in Figure 43. y ., g, - 7 n y gg.vg
- . .)'- ' 5 * .
,. ( ,e -
k..: .y;
-[~ . _
- *~. f ..
m s
.N. <- .
\.
7.- ~ s. y '; , ,;i N.: _J > >
,f y W.
- 9. x~< ,
.~~. ,
y'
,. gf c.
~ .. . _ . .
)
.. s. . .
.a
)_ Y.
4 'y< ., ,
.,e .
. f. .-
,. y .
'/
v .- .- .-
,~ -
i .
- q. . .
< :, _ a /:K' ,\
s
-s, +
3
\
.3, . .* ..w Figure 45. Etched view of end of crack 128X in Figure 43.
l l ! 3-58
I M .7.q; yt. ') Jr 's g[{ 1 " 'v s. b( g:a :! ; iy~ l l 2 3 4 Figure 46. Penetrant examined view of Boat Sample E removed from SW-4 of Loop 12. i
~, .
4 a i .- ,. l
. l
. . i l
l Figure 47. Etched view of indication at 128X elbow side of Boat Sample E. l 3-59
l [ l l l l l l l l .
@y: :
l ,
- vp s,p, 1.,
a- [ l$l , ~
. 2 3 i
Figure 48. Penetrant examined view of Boat Sample F removed from field weld R4-5 of Loop 12.
, 7 + , . g:., ,
3-., , ,.
' : 1;;;
' l, .: ;
,\ ,
': h, ':S >;. .
b l l I ! l Fi gure 49. Etched indication of Figure 48. 50X 3-60
v
,f .,J ,
. 's : ' :. g :;_: -. ,
L L
. ; p _;
_ ,s
. . I I
l , , # .-- % , i \ l l . I ! Figure 50. Penetrant examined view of 12X i indication on elbow side. Section ( was removed from SW-12 of Loop 15. 1 i i
.. 2
, s - c
, t;,, . /
l> jy'; J \ ,' J ,. h,. -
~Y " lflQ'$ -
1 y ,~, my%J'?
.m o ,
i
%T
.s . < ,
- j j . ,
j , i i Figure 51. Penetrant examined view of s2X longest indication, pipe side of pipe-to-elbow weld SW-12 removed from Loop 15. 3-61 t
7,
,. 17]
7;n u.
" 56)
. y x; y.7 ysw. 1rhn,
,t.- a't#m.. '< '%,
Figure 52. Etched view of entire
;w i
- }Ef%yfM1;;$
li
- Qi crack, elbow side of y
L *. a .
~o xPfEI'r y%{ SW-12 of Loop 15.
i [' {h,3 1 %* nyreg N +gx[; z n QQ:,:.
- f. _, . 6 *-::;
' ,R -
S .. - ?k.: - g 'n + , ,,. y
~
. s i c.w*.n;m:yfaQ, ddhkk' j, yiy 5.gy r'( t 3 i[rn5N 1
,i
' g, - a
?
4, .
..,Dff;#?!
r , , j
,n >
#s -
'% 4 33X
., 4, _
j ~ 3. '.? (m
.m ,. .
,7 ,,
./
'- } -
-7 ,. )-
, p .
') -; .,
(, 1
' .- 'j,
. .: q l
j'4 > ;
...M j %'-)' ,av?.&fls.?[
t G ,MTu '
, cd, . .u.hsd[NEh6.@w n.
\
. . . x :.
K~ ~ .,
.;!~::> j ;:?*; l. m@tc.~f,
.:.~. a. s :s.: ., n? l ,:.',* &
128X Fi gure 53. Etched view of beginning of crack in Figure 50. 3-62
~
...~._
- s
}@sM.:F. s, )\
., .t.? .
N, u3. :s p%- , u -
.s
,c. v .
3 9.% u .- b, ? 33Q +, :J.'@. r
- y
.&g s, ;. 4{;% . - f ;-[,,~$ Q.iY~v hh
\'. s(;.%
~
'a
- f 'gi',
- /2, W' Mo .y . ' + j , -
- +
1 x ,
/ ,.,,4 r;d p - . . - ,,. , - , .,,, g, .
4.
, ,/ k. r-(di'.jsJ ,.W Ll ', ^ -
3 w ). g . i.
.K .' , -
- y. m
_ -) n5'if g~' J. u.a.? - - .
.. rls . ,.jp . 'g! yh)A;A ::
, k .., ..
e 4 y.% -c.: > s-'g.<
-t .
,; , .~..:'x j
-.s ,
'a7 h)is
,s
,..W.... 5 ;..d.p c m.. r; ,A. 3 ev, ,:.c yl
"* e 4 >
~
M 7 s: :
- n.e, g s h,di ;[ $ . 3- '
y $'! ?
@Mba<c[9', e.t y e. . :;s ;
4 4, N-s.[h:k.w.w-kh
- q. q
- y. 0. m' p,"
~
'.y.g('rp.gn. , y n,%,, o m-2-
%g. - n,,y.t '
.w;:,$'),
- cc.J.N,.
,# .=...
s.;d;M: .[::.. isT -U Figure 54. Kahling etched view of crack 128X extending into weld. This view shows crack arrest in weld metal containing significant ferrite. ap. _ e- .,.C 4 ,. + .- e
.i i -
. ,J-h " E D, . ,.
,, a .
j, yzr'. . ** ..
...'y ,
's .. : e s.
. g.- gou. -i.,
. f7 ( ' ,. iI ,.,> f Tb hy% *
'T .f( v-o' l\ .s .
.3 i .k,
- o qh*N..,[Yk.-)
i , i
.%g'A -
c, s
.w, s.,s- i, 4,, ,y.eh .,,,.
y y,. '. ,j w =
*' ~ ,Q k.tfri e 1 1T ';' '.
~- 4 -vf.3:{(.cf.. ,. n*- .
l ,
~s.~ * . , ' *, ,t,t,P. ,,* ' ,
\ ,
,, b -g _ ' #1 ,,
..- . .* .. u i'.- .
b a .,,' -' ...c'c sSg,t,
- i %. '
e q
"* 7 sd ':
6 -[gg h; , ? _ * Tj'n.[g * ',..,
. r+. g
- c. w ,,y w :p . - , ,
a A... y,3 - c ,. s s t,. . s cp y. e--r , u, .:
. 1 r.
s . .a,_. - Figure 55. Etched view pipe side, same 128X section as seen in Figure 50. 3-63
l
> 4, t
_ , L.,.
-.r 1
. ,,t.
i e 1 r n -: 1
-'ge'
% Ss I
l
'"9 M% t -s
~Z ~
Y@ 4 , 1 r *.
-- /\, ,_
--r ,- s ig;.4 p
_p. 54
-[ . .y6;)
3 . .. .. ys d _- N
. .. % r?fj@g r
_ -s> s e. q q, f n Sp a.ep: ..
-[. 'd ::s
! - se ' 'j .,
s
- L;;cM.]
ygp;
, J .g
*d j 1
., .d. m,,. o.
-k Q s g_ .. * . 's)
.v
- r?
11'i . i -
~.~T _q A ", .'L' /a /' .F n .. s .
W n.;\ i J
.n -) .
c ./ ~
~
.], f _: h -
.t k'aj ' -
'^
A,b .
. J
- y.
d 33X Figure 56. Etched view crack in pipe side. Sample was removed from shop weld SW-12 of Loop 15. 3-64
- n. 4 u .
4 ' i ,
?
w - .l - ,
/ ,
v s { i p u i ,
/
l > 1 '1 ) Figure 57. Etched view, beginning of crack 128X i of Figure 56. i.. i l l
- m; ;n;sy a= ~.. v;m
.. m o,,,.ip .ss . . . .: y -- , ,w
. .o . .
- , 3. . .
t ..<.--
....%.g e
,. w k
.w . . -+ . . :n , ?
'-.c:.-
w- ..# ^
?,: - [a . .. .
Q %. ' @.:
'. w ;
.m
,.,..e...
s, c. m . m ., . . ,
-..c..
t., . .
. n. . .f . 3 c
*.b, j: . 1 #' -{-
- g. . .'t . H..y .. '. . *. -
;. e ,, ,~.r
;. / 4 ; v.'s.
.s
*e.s . ,
- y4
- 70 T - ri em . . . - n( . . A. ,-
& -' a .-
- s. ao+ %!:?'.t
- i ~ ,
my
-.n.
vp .- ..a. *
.L.n*.s. s .. ' ..,j. . .
ffg*y , s. --
~ .
- m. .,m . . ..,.;e- .Fi~ ~ ,
- s.v.D.u
, . . s r
~
.s~
h,,&. s% }{ . _1 s-g%
~
..Qa's E;- , , :h .
.^ : .y
< :7 ..,
g, ,
..~.C
- ' r,.
Figure 58. Kahling etched view of end of crack 128X , of Figure 56, 3-65
.,.~,g
.--- 1, i' . - - /
.,s
+ F #e - .. s
/ .- :
. ,/ ,
4
. / .
t l , i,
~-
*/ y,
/ 2 m
1 ,- - i ! 800X Figure 59. SEM view of pipe side HAZ, showing carbides. l h l i l , l ; I eg ,
- e. '
- 1 .
i e ,
. g
- 4 l
- 6 j
h 800X Figure 60. Dark field view of pipe side HAZ, showing carbides precipitated at ' grain boundaries. l 1 l 3-66
I% v* , _
.' '7f m ! .
I nd' . );. .. / Ig , ,
+ n, e~g :s. a 4nyi
' 1 L.
+ -
4 .
~
;. )
J7 ocj, 9@ ,
,:1 ~] .
h
)
. 'iti - .
E *. y.a. . s. , l - , v: - , v.M i i i ' g
*"6 p 4 pk -
- N 4
. , .;, +. . -1
.' : - ' . , \
y [8 , n t ' i
- e 'e 'O a
. 5- '
P. ,, .. .
, .7 , _,3 - .
{ s.g . # sN a .
?
, . {
, /: 4 t, . . > ' M u9
- m. ~ ~ U_ , . -
c L . . 200X FIGURE 61. APPEARANCE OF SPECIMEN MICR0 STRUCTURE AFTER THE DEGREE OF SENSITIZATION TEST (ASTM-A-262 PRACTICE A). 3-67
e 'W
,.g
./:4y,1 @
O W th',;"y,. -' @; j
.. gfda fh,%l
. . l,', ~ ,f ki '{}Q f}.
v.w. $. ,l.V ? <y.;;m. :.4%
. m. '
, ye h
%3;. ., :* . ,
.. , 1; , , ;.
* .i-e38E , e . .:
(.; Of - - x , ,9 . ', ..
\ '_ . lg' sg -
s d
,d v .
j!
- o
- i. . m , -,
/ . .
t I 4 7
,4
.{
J e-M 500X FIGURE 62. APPEARANCE OF SPECIMEN MICR0 STRUCTURE AFTER THE DEGREE OF SENSITIZATION TEST (ASTM A-262 PRACTICE A) I l I 3-68
SECTION IV l STRESSES AND STRESS RULE INDEX i Based upon the investigation as described below, stresses in the system are of nominal magnitude with the pressure stresses somewhat higher than more recent BWR designs, and there is nothing with respect to operation, support of the piping, or restraint of thermal expansion that have contributed abnormally to the apparent extent of cracking discovered. BACKGROUND In an attempt to pinpoint the cause of the extensive cracking, a detailed investigation of the pipe stresses in the system has been performed comparing the as-designed condition with the as-installed in-cluding any variations from the originally expected operating modes for the system. The piping of concern here was reviewed to investigate pipe stress due to pressure, weight during normal operation, thermal expan-sion for all operating modes, and residual stresses. The stress rule index was also computed. In addition, the as-installed support system was reviewed to assure proper operation for all components. The compo-nents included here are the snubbers, spring sway braces, and constant support spring hangers. Teledyne Engineering Services analyzed three of the five recircula-tion loops with their associated branch piping.2 All recirculation loops are similar in geometry with the exception of the branch piping. The range l l 4-1 x
of operational stress level of the recirculation piping was bounded by analyzing the geometries of recirculation loop 12, which contains no branch piping and loop 15 which contains the largest (14-inches diameter) diameter branch. In addition, the two loops which were found to have through wall indications were modelled. The two loops are loop 11 and loop 15. Stress levels were consistently well below allowable for the l three loops. The effects of the branch piping on the recirculation pip-ing were as anticipated (i.e., marginally larger stresses than the non-branch loop). Therefore, the stresses reported within this document are mainly for loop 15 which contains shutdown cooling branch piping. 1 WEIGHT DURIflG NORMAL OPERATI0fl These include stresses due to the weight of the piping compo-nents, insulation, and contents with the support system in the as-designed condition and in the as-installed condition. Investigation of the as-installed condition by NMPC personnel indicated a possible dis-crepancy in the spring hanger settings for the as-installed versus the as-designed condition. These discrepancies are tabulated below. Load No. Load No. Hanger No. As-Designed As-Installed - H-3 Loop 11 S 20000 lbs 17900 lbs H-3 Loop 12 H-3 Loop 13 ) Loads are per hanger. Two hangers exist H-3 Loop 14 at each support H-3 Loop 15 s NOTE: The above hanger load information has been verified by NMPC personnel by calibration testing. See reference letters: R. Pavlik-ITT-Grinnell Corp, to Niagara Mohawk Pwr. Corp. dated Jan. 25, 1982, and February 7, 1983. l 4-2 l
~
The maximum as installed weight stress for the recirculation piping is approximately 375 psi at node 270 isee Figure 1). This is the same as the as-designed weight case (i.e., the spring hanger load reduc-tions cause only an insignificant change in stress on the recirculation s loop piping). To assure proper (exact) weight distribution on the new system, new constant support springs will be installed. An additional non-operational deadweight case was investigated which includes a dewatered pump. This represents a condition that may exist during plant shutdown while maintenance is performed on the pump or pump motor. The pump motor was also assumed to be removed. The l constant support springs are not pinned. The weight stress from this case is approximately 1200 psi at node 270. These stresses are low and will not be a problem for the recirculation system. THERMAL EXPANSION During normal operation of the plant, the recirculation loops are at the same temperature as the reactor due to the high flow rates through the loops. That is, with the reactor at 5500F, the loops are isothermal at 5500F. This represents the as-designed normal operating condition. There have been times during operation of the plant when' a loop (or loops) have been " valved-out". In this event the pump is not operating and the valves (upstream and downstream) are closed leaving a stagnant leg of water below the valves. The stagnant water in the loops cools off to approximately containment ambient (1000F) awa.y' from the reactor. In addition, a case may be postulated where the branch piping 5 would flow into the "volved out" retirculation loon riser sr.d r:e thc reactor pressure vessel. The results would be a totally hat leg from the recirculation puna isolation valve to the reactor. The stresses 4-3
resulting from the " valved out" temperature distribution with branch flow into a loop have been investigated and found to be generally higher than the normal operating condition stresses. The maximum intensified thermal expansion stresses during normal operation are approximately 3000 psi at the branch tee. For the " valved out" branch flow case maxi-mum intensified thermal stress levels occur at the same location and are approximately 9200 psi. According to operating records at the site, the following loops were " valved-out": Loop 11 66 days 9/20/72(32 days)and 8/27/79 (34 days) i Loop 12 32 days 8/17/73 Loop 13 31 days 2/7/82 Loop 15 720 days 6/21/79 to 7/3/81 The " valved out" stress level increase (see Table 1) does not represent a significant increase in the SRI values which are in the l range of 1.2 (except at the tee welds, see Table 1). Further, Table 2 does not demonstrate any significant difference in crack location or the number of cracks for the piping systems which have been " valved out" for different durations of time (including no " valve out"). It therefore appears that the " valved out" case does not contribute to the i recirculation pipe cracking. I OPERATIONAL CONCERNS l , A. Snubbers: The recirculation pump is on a " floating" base. The base is supported by spring and snubber type supprts caly. A portion of the thermal expansion of the piping is accommodated by the floating pump base. During the past (1979) outage the snubber configuration of the pump base was i-changed. Also, the hydraulic snubbers throughout the recirculation piping were replaced with mechanical type snubbers. To assure tnat binding of the mechanical snubbers from displacements due to thermal expansion were not a l 4-4
problem the vertical pump base snubbers were instrumented with LVDT's. The plant operational LVDT displacement data measured at the recirculation pump bases indicates that the snubbers were functioning properly under thermal load. The measured snubber deflections were comparable to the thermal expansion analysis deflections. To further verify proper operability of the snubbers, a representa-tive sample of the snubbers were tested by Teledyne Engineering Services. These tests indicated proper operation during normal plant conditions as well as during dynamic loading events. This testing is reported in Teledyne Engineering Services' test report number TR-5923-1. B. Pump Vibration: Pump vibration was not considered to be a problem for the recirculation piping. Nine Mile Point internal correspondence 55-01-013 indicated that the worst case impeller mass imbalance would result in a maximum stress of 313 psi with a maximum deflection of 0.0074 inches. Nine Mile Point has tested the pump vibration instrumentation (3-23-82). The test data indicates that the maximum pump defleccion is 0.9 mils (0.0009 inches) and the pumps vibration sensors are set slightly above the maximum tested deflection. The tested deflection is significantly less than.the maximum deflection obtained from analysis, indicating that no vibrational stress problems exist from pump operation. C. Sway Braces: The vibrational sway braces existing on the recircula-tion piping risers if properly installed and operating are designed to have no effect on the operational deadweight and thermal stress levels. A repre-( sentative sample of sway braces have been tested by Teledyne Engineering Services to assure proper operation both in the past and for the future. (See Teledyne Engineering Services' test report number TR-5923-1.) l l 4-5 i L .
STRESS
SUMMARY
The recirculation piping was reanalyzed for the operating con-ditions discussed in Sections I and II only. The resultant maximum stress summary is listed for recirculation loop 15. The additional in-formation for loops 11 and 12 are listed for the same point to compare relative stress magnitude between loops. Recirculation loop 15 Node 265 pump discharge riser tee , Operational weight and pressure stress = 8220 psi
- Thermal expansion stress range = 9138 psi Recirculation loop 11 Node 265 pump discharge riser tee Operational weight and pressure stress = 8054 psi
- Thermal expansion stress range = 1676 psi Recirculation loop 12 Node 265 pump discharge (no branch line or tee)
Operational weight and pressure stress = 8014 psi
- Thermal expansion stress range = 947 psi
- Axial component of pressure stress.
The pressure stresses may be somewhat higher than other more recent BWR.recirc. loop designs, due to the pipe wall thickness. The pressure stresses are still within the appropriate design code allowables. RESIDUAL STRESSES Residual stresses present in the weld heat affected zone contribute significantly to stress corrosion cracking and are therefore, accounted for in the General Electric stress rule index (SRI). A residual stress value of 27,000 psi for the 28-inch diameter run piping was used in calculating the SRI numbers. This value is an extrdpolation of data based on the envelope of maximum axial values measured by GE and others on schedule 80 pipe sizes ranging from 4-inches to 26-inches. e 4-6
For RESID = 27,000 the approximate contribution to the SRI is 0.374. STRESS RULE INDEX The stress rule index was computed for the recirculation loop ) using loads and stresses for the as-designed cases in addition to the ' as-installed (including " valved-out" conditions) case. l The stress rule index has been computed as follows: SRI = Pm + Pb,Q+F+R ; S y Sy + .00ZE where: Pm " t Po = operating pressure (psi) Do e out:id dia:--tcr (in:het) t = wall thickness (inches) P b " B2 f Ma Ma = resultant deadweight moment (in/lb) I = moment of inertia (In4) 82
- Primary loop stress index Q + F = (K1 C1 - 1) + C2 K2 Mi + K3 C3 EAB (aA TA - *B Tg) 4-7 i
OF Po2tDo + B2 Do 2T HA SRI = Sy (K1C1 - 1) 2 +C2 K2 Mi+K3 C3 EAB (oA TA - 'B T) 3 +R l Sy + .002E K1, K2 , K3 = local stress indices (NB-3680, Reference 3) C1 , C2, C3 = secondary stress indices (NB-3690, Reference 3) Mt = therral (operating) resultant moment (in/lb) EA g = Young's modulus (average for dissimilar materials) a A. a0 = thermal expansion (in/in/0F) TA, Tg = operating temperature (OF) R = residual stress (Reference 4) (psi) Sy = yield stress (Reference 3. Table 1-2.2) (psi) E = Youngs mdulus (Reforon e 3. Te,1e I-5.0) (psi) The Stress Rule Index was computed for recirculation loop 15 and is presented in Table 1. The primary contributors to the Stress Rule Index are pressure stress and residual stress. The higher pressure stresses here appear to drive the SRI values somewhat higher, since the residual stresses are assumed constant for a given pipe diameter. This may account for a higher incidence to cracking. 4-8
Table 1 MI hx 21 Node Description As-Designed III As-Insta11ed I2) 101 Elbow 1.239 1.391 105 Tee 2.155 2.291 115 Tee 2.132 2.283 135 Valve 1.181 1.262 155 Elbow 1.188 1.287 156 Pump 1.262 1.397 200 Pump 1.182 1.260 205 Elbow 1.179 1.339 225 Valve 1.185 1.282 270 Elbow 1.429 1.444 NOTE: (1) As-designed - including normal deadweight + 5500F thermal case effects. (2) As-installed - including normal deadweight + " valved out" pump with branch flow + " valved out" pump no branch flow thermal range case. (3) Tee indices used although the indices were designed to give stresses in the crotch region not at the weld. l It must be noted that the recirculation loops at fif1P1 have less butt welds than other BWR designs. Butt welds have lower Stress Rule Index values and, there f ore, would be less than 1.2. This f act would make the SRI values, as a whole, higher than for other plants with more pipe-to-pipe butt welds in their designs. 3 l l Table 2 is a comparison of the maximum stress rule index and the number l of cracks in a given heat af fected zone (l'eference 1) at a speci fic location. I These data do not correlate very well leading one to believe that the carbon content, previous cold working and heat treating of the pipe at a speci fic i location contribute more cracking than the magnitude of the stress in the pipe. The actual locations may be found in figure 2 for loop 15. Samples of the Nine Mile point recirculation piping have been analyzed for carbon content. The results of this analysis indicate that the carbon content range is between 0.042 and 0.075 percent. The carbon content is greater than the 0.035 percent upper bound therefore indicating the pipe is in the susceptible region for IGSCC. 4-9
It can be seen from Table 1 that, except for the tee, the SRI values are all very similar at a value of about 1.2. This value is equal to the threshold with respect to stress corrosion cracking as observed by GE for a large number of plants / systems surveyed. It must be pointed out that the SRI values have been computed and presented here for reference and comparison by the reader with other plants, etc., keeping in mind that the relative values only are of most importance. The apparent higher values of SRI for the tee may be appropriate for the branch pipe weld but not for the run pipe welds since the stress indices (ASME Code) depict stresses at the crotch near the branch weld. Therefore, the SRI at the run pipe welds would be about 1.2 also. Indices apply here v T c Indices for a straight pipe would be more appropriate here The stress of importance with respect to IGSCC is the sustain-ed tensile stress on the inside surface of the pipe in the heat affected zone of the weld. A similar situation exists for elbows in that C1 and C2 do not c;;1y to the F.A?. CI and/or C2 represent maximum shear ' stress at times. In an elbow C1 is representStive of the hoop stress in the crotch and Cp 4-10
l l l l r' presents hoop bending at the side. C1 g-C2
\
1 Table 2 Node Max No. Cracks *,* Loop: Total Number Description SRI 15 11 12 13 14 Cracks ** 101 Elbow (I) 1.391 0 0 2 1 0 3 103 Elbow (J) 1.391* 6 3 6 3 0 18 105 Tee 2.291 4 3 NA NA 0 7 115 Tee 2.283 4 0 NA NA 0 4 135 Valve 1.262 4 4 4 4 4 20 140 Valve 1.262* 3 4 2 4 4 17 150 Elbow (I) 1.287* 4 6 7 7 4 28 155 Elbow (J) 1.287 2 2 2 2 2 10 205 Elbow (I) 1.339 2 2 5 3 4 16 f l
- SRI assumed equal to SRI on opposite end of component.
** Represents the number of multiply nucleated cracks at the indicated node number.
l 4-11 .
LIST OF REFERENCES _
- 1. IGSCC Task Force Meeting Minutes November 23, 1982.
- 2. TES Document No. 5838-4.
- 3. 1977 Edition of ASME Boiler and Pressure Vessel Code, Section III, l with Addenda. l
- 4. BWR Owners Group and EPRI Joint Presentation to NRC, October 15, 1982.
i b [ t 4-12
'veTF1FnYNE ENGINEERING SERVICES
, Utn t uti
). cg to e
,),,q UY. ,DATC 63 3
.1,,
CHKD. E3 Y ATE N"8h
- Y
,- .,, , o vtuc wr
,a ; -
2 to a'w -r s o:,,
?
Jo be a f*p , ,,;
"" a 36. nv o 3)*Db p.
7, N.,,, /
/ "
;., pd n- an o , ,nn __
, c
.Y M ' '
4
,j,J, 38 us-2 32'*E'C " E Y
x MM J2 te .446
/ .e
\
3g.,g 3:-,:-a [
/ m o
,3 a
/h t k 1 M AZ Q _
u . :. [ sz ns zeru $ .s2-os-n+ A 4 e 92 ,5-otz f a J :. e mic p3-NG-c3 '*' l . ar *:nt z d, t d, e
- n. k .b h- C 1 i e w . aerv n.c.,a 1 3:.tr.s.:1 J2 8f d'h A Ar ica k
\ I ..,M "' . e 32 is- nic.
" 1:-r: . w
):.oc,+,e , _ l -
_4-- k32M-23 eJ 51-rY>-30 nn Figure 1. Node Diagram for NMP-1 Recirculation Piping System 4-13
I . c MATCH LINES e e = 3 (9 '3 g :$2 C 8 8 [ g ==t-g , , 8 e s 8X
- l. j T
U 8 8 Figure 2. Node Locations for Stress Rule Indices Calculated for Loop 15. 4-14
SECTION V MATERIALS AND FABRICATION HISTORIES The five 28-inch recirculation loops were constructed of Type 316 stainless steel pipes, elbows and fittings. The straight sections of pipe consisted of rolled and welded plate and were fabricated by the National Annealing Box Co. The elbows and reducing tees were fabricated by the Crane Co. and were also formed from wrought plate. Following fabrication all of the wrought pipes, elbows, and reducing tees were solution annealed and water quenched. Shop welded subassemblies were prepared by the Grinnell Co. No additional heat treatment was pe r formed. The National Annealing Box piping was produced from four Seats of material which contained 0.042% to 0.055% carbon. The elbows and reducing tees were fabri-cated from nineteen heats of material ranging from 0.050 to 0.075%C. In all but a few cases the heats from which specific sections of piping were fabricated is not known. The ten furnace sensitized recirculation safe ends were made from a heat of 0.054%C material. RECIRCULATION SYSTEM CONFIGURATION The five 28-inch diameter recirculation loops are shown in Figures 1 through 5 of Section II, " Extent and Distribution of Cracking". Each weld is identified as either a shop or field weld. The weld numbers correspond to the recirculation piping isometric diagram. Each portion of the system that includes one or more shop welds and is bounded by field welds is one subassembly. The subassembly welds (shop welds) were made by the Grinnell Co. Loops 11 and 14 are identical but differences exist among the remainder of the recirculation loops due to the number and location of reducing tees and spool pieces. 5-1
.- - . .. .. . . .- . _ - _ - - - - - = __ - _- -
I The wrought portion of the recirculation system was constructed of 31 lengths of rolled and welded piping,15 90* elbows, 5 76* elbows, several reducing tees, and ten safe ends. The valves, pump dishes, and the elbows directly beneath the pumps were all made from cast stainless steel, and will not be discussed further since IGSCC was confined to the wrought piping. The remainder of this chapter consists of a discussion of the material and fabrication of the wrought recirculation system components. MATERIAL AND FABRICATION A. Pipe The straight segments of piping were manufactured by the National Annealing ! Box Co. and consist of rolled and welded Type 316 stainless steel plate material. The applicable ASTM specifications are A358 for the plate and A240 for the pipe, in accordance with General Electric purchase specification 21 A2103. A total of 31 straight segments of pipe were produced for Nine Mile Point 1. The part No. , length, material heat No, and slab No. for each section are listed in Table 1. ] This data was compiled from the recirculation piping material Test Certifications, which are found in Appendix A. The parent plate material was produced by Allegheny-Ludlum. Four heats of material ranging from 0.042 to 0.055%C were used. 3 The chemical analyses of these heats are given in Table 2. The corresponding mechanical properties are given in Table 3. Although it is possible to determine the heat of material from which a particular pipe number was fabricated, the distribution of the pipes within the recirculation system is not known. Hence a correlation between pipe cracking and l heat number cannot be developed. Table 4 lists the number of pipes from each heat and the corresponding carbon content. The longitudinal seam welds (one per pipe) were performed by an automatic submerged-arc process per National Annealing Box procedure 7840-A. This procedure required that the as-deposited weld metal' contain a minimum of 5% ferrite as determined by the Shaeffler diagram using the weld metal check analysis. The chemical composition and calculated ferrite content of the weld metal is given in Table 9. 5-2
. - - =- .. -, . - . _ . _ _ -- . _ _ . _ - _ _ . . _ _ _ - . _
following welding the pipe was solution annealed per National Annealing I Box procedure 7844-B, which specified heating to 1900*F-2050*F for 1 hour per 1 inch of thickness and water quenching. This procedure required that cooling i between 1700 F and 1000 F be accomplished in three minutes or less, and that
- rapid cooling continue to below 800*F. Electrochemical Potentiokinetic Reactivation (EPR) measurements made at the site and at the Vallecitos Nuclear i Center have verified that the pipe base material is in the solution annealed condi tion .
B. Elbows and Tees The 90 elbows (three per loop), 76 elbows (one per loop) and reducing tees were supplied by the Crane Co., Midwest Fitting Division. These components were manufactured in accordance with ASTM A403 Type 316 plate material. Note that each i loop also contains a cast 90* elbow located directly beneath the pump. I
- As in the case for the straight pipes, available records do not indicate the heats of material to which particular elbows belong. The chemical analysis of the l parent material is available however. Table 5 lists 10 heats used for 90 and 76*
elbows, and Table 6 lists an additional 9 heats identified as having been used for reducing tees. It is likely that some of this material was used for plants other than Nine Mile, but the fittings used at Nine Mile were fabricated from among these heats. The range of C content is 0.050% to 0.075%. On-site examination of certain sections of recirculation piping revealed heat j numbers stamped on some elbows. Originally it was required that heat numbers be stamped on all of the elbows, but it is likely that many were eliminated by weld-prep machining or covered by the weld crown. The elbows on which heat numbers were found are listed in Table 7. Of these, cracking was discovered in all except one (loop 12 suction), and there was at least one cracking incident for each heat of material identi fied. Available records indicate that all of the elbows and tees supplied by Crane were solution annealed per Crane Co. procedure MWP-WP316-AN1, which specified heating to 2000 F 2 25'F for one hour per inch of thickness followed by water quench. , Cooling below 800*F was required in three minutes or less. EPR measurements have l l ) eerified that the elbow base material is in the solution annealed condition. i l 5-3
C. Sa fe-Ends The ten 28-inch recirculation safe-ends are Type 316 stainless steel for jings per ASTM A336. Although solution heat treated following fabrication, the safe-ends were subsequently post-weld-heat-treated with the reactor vessel, which caused them to become sensitized. All ten safe-ends were fabricated from the same heat of material, which contained 0.054%C. The chemical analysis of the safe-end material is given in Table 8. WELDING j The GE purchase specification for the Nine Mile Point I recirculation system ! required all welded joints to be made by the inert-gas tungsten-arc process with internal inert gas purging for at least the root pass and second layer. Shielded metal arc was allowed for the remainder of the weld. Weld filler metal was required to meet ASTM A298 or A371 or equivalent for
! c ther processes. Austenitic stainless steel weld deposits (all pressure boundary welds from safe-end to safe-end) were required to contain controlled amounts of fe rri te . Both 5% ferrite minimum and %Cr>l.9% Ni were allowable acceptance criteria. [
A summary of the austenitic filler metal used for pressure boundary welds in the ! recirculation system is given in Table 10. The filler identified in Table 10 as having been used by National Annealing Box and Crane was for longitudinal weld seams in pipes and elbows. The Grinnell filler metal and consumable inert rings were used for shop welds. The heat input during welding was estimated using the equation: H (joules / inch) = Current x Voltage Travel Speed and using maximum values for the current an minimum values for the travel speed. He;c input estimates for typical shop and field welds are given below: 7 1 Typical Grinnell Shop Weld GTAW Root Pass: H = 36,000 joules / inch SMAW Second Pass: H = 26,250 joules / inch i SMAW Cover Pass: H = 79,800 joules / inch 5-4
- 2. Typical Field Welds GTAW Root Pass: H = 36,000 joules / inch SMAW Second Pass: H = 25,000 joules / inch SMAW Cover Pass : H = 44,550 joules / inch 5-5
TABLE 1. SUP9tARY OF NATIONAL ANNEALING BOX 28-INCH PIPE SLAB PART NO. LENGTH HJE 7835-1 7 ft 9.25 in 340041 4 2 3 4 5 I 7844-1 22 ft 9.00 in 29151 2 2 22 ft 9.00 in 46550 1 3 19 ft 3.0 in 46580 1 4 15 ft 9.00 in 29151 2 5 13 ft 5.50 in 340039 6 6 7 7 7 8 6 9 12 ft 9.00 in 340041 3 10 12 ft 1.00 in 29151 3 11 11 ft 8.50 in 29151 1 12 340041 3 13 29151 1 14 340041 3 15 9 ft 11.50 in 340041 3 16 9 ft 8.50 in 29151 3 17 3 ft 3.75 in 340039 6 18 7 19 7 6 20 g 21 29151 2 22 1 ft 4 in 6 23 7 7 24 ( 25 6 I( 46580 1 ( 26 5-6
l TABLE 2. s CHEMICAL ANALYSIS OF ALLEGHENY-LUDLUM PLATE l l HEAT C Mn P S Si Cr fli 'Mo l 340039 .055 1.60 .023 .010 .55 17.40 13.00 2.42 340041 .049 1.62 .024 .01 0 .55 17.10 13.50 2.46 l- 29151 .049 1,63 .01 6 .01 7 .68 17.20 13.52 2.39 46580 .042 1.66 .01 8 .013 .45 17.62 13.40 2.77 i 5-7
l l TABLE 3. PHYSICAL PROPERTIES OF ALLEGHENY-LUDLUM PLATE j I i HEAT SLAB YS (psi) UTS (psi) ELONG (%) RA (%) BHfl i' 340039 1 44900 86700 60 67.9 156 2 45700 86800 55 70.1 163 3 42800 87600 50 69.6 153 4 44200 86300 60 69.0 156 5 43000 89500 60 69.0 156 6 42900 88400 55 69.4 156 7 46000 88400 52 65.7 170 29151 1 43400 83700 61 71.4 159 2 39200 81900 64 69.3 143 340041 3 38900 82100 65 69.5 143 4 4100 0 78100 70 71.; 143
? 41300 81700 63 73.0 149 29147 4 46200 85800 55 68.0 159 46850 1 42700 32900 56 68.8 149 TABLE 4.
PIPING BREAKDOWN BY HEAT NO. HEAT NO. PIPES FRACTION E 340039 12 39% 0.055 340041 9 29 0.049 29151 7 23 0.049 46580 3 10 0.042 l 5-8 . I
TABLE 5. CHEMICAL ANALYSIS OF CRANE CO. ELB0W MATERIAL HEAT C M P S SJ Cr E g 45015 Mill .061 1.60 .026 .022 .49 17.13 13.37 2.50 Check .065 1.62 .029 .026 .50 17.11 13.34 2.58 65122 Mill .060 1.77 .025 .011 .52 17.32 12.76 2.39 Check .059 1.79 .024 .014 .51 17.36 12.70 2.40 45109 Mill .053 1.80 .025 .01 0 .59 17.26 13.02 2.29 Check .055 1.78 .028 .012 .60 17.30 13.00 2.26 44839 tilli .057 1.53 .020 .021 .59 17.04 13.13 2.24 Check .060 1.50 .020 .020 .60 17.08 13.10 2.30 44842 Mill .057 1.70 .01 9 .010 .45 17.13 12.86 2.24 Check .061 1.70 .020 .01 3 47 17.10 12.80 2.19 45012 Mill .061 1.46 .020 .020 .51 16.90 12.96 2.57 Check .060 1.45 .021 .020 .50 16.89 12.97 2.55 45110 Mill .060 1.48 .022 .01 6 .53 17.28 13.15 2.27 Check .062 1.48 .020 .015 .50 17.30 13.13 2.25 65120 Mill .061 1.55 .022 .020 .64 16.88 13.44 2.45 Check .063 1.59 .021 .020 .66 16.80 13.48 2.40 44840 Mill- .051 1.57 .021 .021 .63 17.18 13.22 2.25 Check .057 1.55 .022 .020 .59 17.17 13.18 2.20 65077 Mill .052 1.55 .024 .01 3 .'61 16.95 12.90 2.28 Check .050 1.58 .021 .012 .61 17.00 12.88 2.30 5-9
TABLE 6. l CHEMICAL ANALYSIS OF CRANE CO. REDUCING TEE MATERIAL i HEAT C M P S, SJ ,Cr r E g 65121 Mill . 056 1.85 .022 .020 .61 17.17 13.09 2.20 Check
. 060 1.80 .020 .024 .59 17.11 13.10 2.22 44838 Mill . 054 1.80 .022 .017 .62 17.32 12.88 2.30 Check . 060 1.81 .01 9 .021 .60 17.29 12.95 2.31 Mill 058 1.74 .024 .022 .53 17.27 13.00 2.23-65123 .
Check . 061 1.70 .020 .020 .55 17.30 13.04 2.23 45153 Mill . 059 1.64 .020 .01 3 .57 17.26 13.00 2.30 Check . 060 1.65 .022 .01 4 .55 17.20 13.08 2.28 45030 Mill . 065 1.65 .019 .011 .62 17.13 12.92 2.47 Chect . 066 1.64 .020 .01 0 .62 17.20 12.89 2.42 45139 Mill . 060 1.70 .025 .028 .54 17.35 13.04 2.25 Check . 061 1.68 .020 .026 .55 17.30 1 3.01 2.24 45157 Mill . 057 1.71 .025 .009 47 1 7.35 13.06 2.27 Check . 060 1.70 .021 .011 .50 17.33 13.07 2.30 45043 Mill . 066 1.69 .024 .020 .45 17.69 12.97 2.30 Check . 062 1.70 .020 .020 .46 17.60 12.99 2.30 64788 Mill . 068 1.44 .020 .01 2 .63 17.25 1 2.91 2.01 Check . 075 1.25 .008 .023 .52 16.47 13.43 2.04 4 5-10
TABLE 7. LIST OF ELBOWS WITH KNOWN HEAT NO. ELB0W TYPE LOOP SURROUNDING WELDS HEAT JC, 90 12 Suction SW 4, FW 5 44840 0.054 76' 12 Discharge SW 18, FW 28 45015 0.063 90' 13 Suction SW 6, FW 9 44840 0.054 76' 13 Discharge SW 21, FW 33 45015 0.063 76* 14 Discharge SW 24, FW 38 45109 0.054 TABLE 8. CHEMICAL ANALYSIS OF SAFE-END MATERI AL HEAT C SJ _P_ g S NJ r C_r M E-5349 .054 .50 .020 1.69 .009 10.84 18.00 2.37
- (
5-11
TABLE 9.
SUMMARY
OF WELDING FILLER COMPOSITIONS AND CALCULATED FERRITE CONTENT Weld Metal Heat % Ferrite Supplier No. C Mn P S Si Cr Ni Mo N Schaef fler National Annealing .036 1.47 - -
.45 18.70 12.70 2.70 - 5 Box Company 1.60 .54 18.80 12.65 2.75 5
.045 - - -
.041 1.64 - -
.64 18.80 12.35 2.78 -
6h
.057 1.56 - -
.58 18.40 12.20 2.64 -
44
.048 1.64 - -
.60 18.80 12.15 2.78 -
65
.044 1.52 - -
.64 18.80 12.25 2.62 -
65 u, d- Crane Arcos C8036- .053 1.24 .062 .01 6 .51 16.84 11.18 2.17 - 5 Mid West T316 Fitting Division S-16-5B7F Grinnell Company Arcos 4N105A .028 2.16 .014 .008 .29 18.97 12.00 2.37 - 6 , Chro-mend 31 6 Chro- 7776- .011 1.88 .004 .011 .39 19.49 11.94 2.39 .045 9 menar 316L Consumable .028 1.58 .021 .01 9 .48 19.68 13.42 2.36 5 Insert Rings
4 4 SECTION VI j WATER CHEMISTRY AS A POTENTIAL CONTRIBUTING FACTOR i I It is known that significant at-temperature operation with out-of- l { specification water quality can lead to enhancement in IGSCC susceptibility. i Therefore, it is important to review the long term and transient water chemis- l
; try history at NMP-1 and to compare this history with that of other BWR plants with significantly higher and lower incidences of IGSCC.
EVALUATION OF GENERAL ELECTRIC WATER CHEMISTRY DATA BASE FOR NMP-1 1 i ' This section summarizes the relevant water chemistry data from the Nine Mile Point reactor. These data are maintained in the General Electric Plant Chemical and Radiation Technology time share files. These files are sub-j mitted monthly by each utility to General Electric to demonstrate compliance to the contractual fuel warranty operating limits. The forms are sent to key punch operators who transcribe the data onto magnetic tape. While some verification effort has been made on the data from each plant, there are l undoubtedly a few transcription errors which have not been corrected. The chemistry parameters which are maintained in the data base and are plotted in this report are as follows: ] i j 1. Reactor Water ! A. Weekly Average Conductivity (uS/cm at 25 C) This is a numerical average of 5-7 daily readings in a given week. , i i B. Weekly Maximum Conductivity (uS/cm at 25 C)
] The highest value from A in a given week.
C. Weekly Average Chloride Concentration (ppb) This is a numerical average of 5-7 daily measurements in a given i week. 6-1
'I
D. Weekly Maximum Chloride Concentration (ppb) The highest value from C in a given week. E. _ Hydrogen Ion Concentration (pH at 25 C) This can be a weekly average or perhaps only a single measurement 1 in a given week. F. Silica Concentration (ppb) ! This can be a weekly average or perhaps only a single measurement
! in a given week.
- 2. Feedwater ,
1 A. Weekly Average Conductivity (uS/cm at 25 C) Same as reactor water. l
- B. Weekly Maximum Conductivity (uS/cm at 25 C)
Same as reactor water. l C. Dissolved Oxygen Concentration (ppb) This can be a weekly average or perhaps only a single measurement in a given week. Maximum data are provided when there is more j than one analysis in a week. D. Metallic Impurities (ppb). l :There are many variations.in presenting these data.
- 1. Insoluble Fe
- 2. Soluble Fe 6
i
- 3. Total Fe
- 4. Insoluble Cu
- 5. Soluble Cu
- 6. Total Cu
- 7. Total Metals (Fe, Cu, Ni, Cr)
- 8. Total Fe + Cu Of these, the most important are Total Metals (dominated by Insoluble Fe) and Total Cu where fuel warranty limits have been established for these parameters.
Most of the data presented encompass the time period from plant startup in 1969 to the shutdown in early 1982. For better readability, each set of data is presented as two plots - 1969 to 1977 and 1975 to 1982. Large time gaps where data are not presented generally reflect a refueling or extended maintenance outage. A. Reactor Water Con.ductivity The weekly average and weekly maximum reactor water conductivities are shown in Figures 1-4. As a reference point for the discussion, the fuel war-ranty normal operational limit is 1.0 uS/cm at 25 C for steaming rates greater than 1% of rated steam flow. The warranty allows for an interval of 14 days per 12-month period where the conductivity can be in excess of 1.0 uS/cm; above 10 uS/cm the limiting condition for operation, plant shutdown is required. An eyeball average of the data would indicate a typical conductivity of 0.4-0.5 uS/cm, with substantially better water quality observed after 1979. In an EPRI funded review of the NMP-1 water chemistry l , it is pointed out that conductivity levels of 1-2 uS/cm were present. However, these higher conductivities were not representative of routine operation but rather correspond to the relatively frequent conductivity, mini-excursions in excess
.1 R. L. Cargill (NUTECH), EPR-09-101, January 1983, " Report on the Field Investigation of Recirculation System Pipe Cracking at Nine Mile Point 1" 6-3
of 1 uS/cm that occurred in the 1971-76 period, (Figures 3 and 4). The highest conductivity observed (30.0 uS/cm) occurred in June, 1971, and was I due to poor water quality in the condensate storage tank. The maximum chloride concentration during this intrusion was 232 ppb. Compared to conductivities of other BWR's. Table I, Nine Mile Point reactor water chemistry is about average. No conductivity excursions have been reported since 1979. B. Reactor Water Chloride The weekly average and maximum chloride readings are shown in Figures 5-8. Again as a reference, the normal operating limit for chloride ion is 200 ppb for steaming rates in excess of 1% rated steam flow. The chloride concentra-tion may be in excess of 500 ppb for an interval of 14 days per 12-month period. The NMP 1971 Technical Specification is substantially more liberal (1000 ppb). The average chloride concentration has been in the vicinity of 40-50 ppb with a few excursions in excess of 200 ppb; no power excursions have been reported since 1977. Like conductivity, the chloride ion concentration has generally decreased since 1977. A comparison of the geometric average of weekly maximum reactor water chloride level at NMP-1 with all other BWR's in the fuel warranty data base is presented in Table 2. Note that although NMP-1 is among the six highest plants based on the 1 standard deviation (upper 84% value) above the geometric average for this chloride concentration, its actual value of 89.5 ppb lies well below the 200 ppb specification limit and within 30 ppb of the upper 84% value of 59.2 ppb for an average plant in the data base. Thus, it does not appear that the steady state chloride levels are an IGSCC aggravating factor. C. Hydrogen Ion Concentration (pH) The Nine Mile Point pH data are shown in Figures 9 and 10. Routine pH measurements are not required in the General Electric fuel warranty document except when the conductivity is out of specification. The 1.0 uS/cm conduc- , tivity operational limit bounds the pH range from 5.6 to 8.6. The maximum 6-4
fuel warranty limit range is pH 4 to 10. Most pH measurements' are performed in the laboratory on grab samples. Due to carbon dioxide uptake by the sample, the measurements are not extremely reliable; the high purity of the water compounds the measurement difficulty. The pH data reported by the NMP chemistry group are consistently on the acidic side and are generally within specification. However, many data points in 1980 and 1981 fall in the range of pH 5.0 to 5.5. The average and maximum conductivity readings in the same time period cannot support the pH data, implying one or both measurements are in error. ; F s - D. Reactor Water Silica s
/
NMP silica data are shown in Figures 11 and 12. The analysis of silica l is not required in the fuel warranty document. The information is useful to provide guidance for the changeout of reactor water cleanup demineralizers. A normal operating maxima of 200 ppb has b'een suggested, along with a tighter administrative limit of 100 ppb. , Substantial improvement in cleanup system performance has been observed since 1978. The appearance of the trend is consistent with the cleanup demineralizers being changed once pe,r fuel cycle. Cleanup system operation before 1978 fluctuated significantly. E. Feedwater Conductivities The average and maximum feedwater conductivities are shown in Figures 13-16. While there is no fuel warranty specification on.feedwater conduc-tivity, a limit of 0.1 uS/cm was established for the common effluent of the condensate treatment system. A 4J,our time interval a'bove this limit is allowed for each incident. Routine operation is characterized by a con-ductivity of 0.07 uS/cm. There have been spikes in excess of 0.1 uS/cm, some of them approaching or exceeding 1.0 uS/cm. Most of these spikes appear to be the result of changing condensate demineralizers, since comparable spikes do not appear in the reactor water plots. 6-5 e;
F. Feedwater Dissolved Oxygen Feedwater dissolved oxygen concentrations are shown in Figures 17 and 18. Fuel warranty limits are 20 200 pob for operation greater than 10% power. The maximum and minimum values were established to protect the carbon steel
, feedwater piping from generalized corrosion. Normal operation at Nine Mile Point is 30-40 ppb. No excursions in excess of 200, ppb have been reported; in a few cases the dissolved oxygen concentration has dropped below 200 ppb.
5' , G. Metallic Impurities The fuel warranty limitation for metallic impurities (soluble and insol-uble) is 15 ppb (Fe, Cu, Ni, Cr) for operation above 50% power, with the usual interval o'f 14 days per 12-month period allowed for concentrations in excess of 15 ppb. Of this value, a separate limit was established for Cu (<2 ppb). For startuds, less stringent limits were provided. The plots for total metals are shown in Figures 19 and 20. There have been several excursions in excess of 15 ppb, but only 4 since 1978. Copper concentrations are shown in Figure 21 and 22. No excursions in excess of the 2 ppb warranty limit have been reported. f EVALUATION OF WATER CHEMISTRY TRANSIENTS In one study, Radiological and Chemical Technology, Inc. conducted an evaluation of conductivity and chloride transients at Nine Mile Point and 19 other BWR's (EPRI Report No. NP-1603). For each plant, attempts were made to define all the occasions where the conductivity and chloride concen-trations were out of specification during operating and shutdown conditions. The.NineJMile Point data encompass the period from startup through early 1978 Table 3 shows the maximum Nine Mile Point-1 chloride and conductivity transient values for each excursion in this study. Without extrapolating or interpolating beyond the actual reported data, attempts were made by Radiological and Chemical Technology, Inc., to estimate the integral number of days that each parameter was out of= specification for each event. The highest chloride concentration observed (683 ppb) occurred during a shutdown in 1977. 80% of the total inte-grated chlhride excursions have occurred during shutdown where the probability 6-6 r i ..
of stress corrosion damage to piping is extremely low, while 63% of the integrated conductivity transients have occurred during power operation. The highest at power chloride concentration and conductivity observed through early 1978 were 287 ppb and 30 uS/cm, respectively. The maximum at power chloride and conductivity transient values for the 20 plants in the study are shown in Table 4. As can be seen, when ranked from lowest to highest, NMP-1 is 7th for chloride and 16th for conductivity. It cannot be inferred from these maximum values alone that one plant is better or worse than another in terms of potential effect of transient Water Chemis-try on pipe crack propensity. In addition to these peak values, the integrated transient time history of conductivity and chloride values ab.ove the speci-fication limit have been evaluated and are listed in Table 5. For both of these time weighted parameters, Nine Mile Point-1 lies slightly below the median for all plants indicating its performance has been typical for those plants in the evaluation. In order to better reflect the effect of plant age on integral transient severity, Radiological and Chemical Technology attempted to normalize the previous data by dividing by plant age. The ranking of this age normalized parameter is shown in Table 6. Again, NMP-1 lies at the median value indi-cating typical plant performance. Further, if the absolute values of these transient chloride and conductivity parameters are examined, it is seen that the NMP-1 values are over one order of magnitude better than for the " worst" plants. Although the EPRI study treated transient chemistry behavior in an integrated comparative sense it is also useful to examine the specific transient behavior at NMP-1. There have been at least two significant water chemistry transients during the plant lifetime. One, which occurred in June 1971, resulted in conductivity values between 8 and 30 uS/cm cver 5 hours with the reactor at temperature. The high conductivity resulted from water pumped from the waste collection tanks through exhausted waste demineralizer beds to the condensate storage tank and then into the reactor. The major impurity was 6-7
i l' probably sulfate and the pH was not measured but could have been acid. The second known intrusion occurred in ic79 1, when chlorinated hydrocarbons were introduced into the primary coolant following valve maintenance. This tran- l sient was not included in the Radiological and Chemical Technology, Inc., EPRI study.I Little is known about the severity of this intrusion except that con-ductivity and chloride values went well above specification limits for several hours. It is possible but unlikely that these off-chemistry intrusicns con-tributed to early IGSCC initiation in the recirculation piping which then grew slowly over several years to the final March 1982 condition. In previous cases of severe off-chemistry intrusions at other reactors, there was evidence
- of stress corrosion damage indicated during or shortly following the intru--
sions. In these cases, damage was indicated by failure of in-core low power range monitors (LPRM). At NMP-1 there was no such damage at any time.
SUMMARY
4 With respect to the NMP-1 coolant environment, the steady state long term conductivity values have been equivalent to that of a typical BWR, while the weekly maximum chloride levels have been somewhat higher than average although still well within specification. In addition to these steady state values, the two most significant off , chemistry transients have also been evaluated. r As a result it is concluded that there is no obvious basis for believing that water chemistry played a significant role in accelerating IGSCC at NMP-1. 1 I EPR-09-101 Rev. 1 - Nutech Apt. 6-8
r- 1 l 4 l Table 1 I STEADY STATE CONDUCTIVITY - PLANT COMPARISONS WEEKLY MAX CONDUCTIVITY, uS/cm Plant Geom Geom Upper Rank Mean (1) Dispersion (2) 84% Value (3) 1 0.145 1.69 0.25 2 0.210 1.39 0.29 3 0.174 1.82 0.32 4 0.183 2.02 0.37 5 0.287 1.44 0.41 6 0.260 1.58 0.41 7 0.265 1.73 0.46 8 0.242 2.11 0.51 9 0.285 1.79 0.51 10 0.353 1.55 0.55 11 0.347 1.61 0.56 12 0.263 2.26 0.59 13 0.341 1.74 0.59 14 0.227 2.73 0.62 15 0.401 1.90 0.76 16 0.444 1.80 0.80 17 0.524 1.55 0.81 18 (NMP-1) 0.392 2.11 0.83 19 0.393 2.17 0.85 20 0.389 2.31 0.90 21 0.451 2.04 0.92 22 0.603 1.55 0.93 23 0.463 2.15 1.00 24 0.420 2.52 1.06 25 0.499 2.29 1.14 26 0.534 2.21 1.18 27 0.751 1.61 1.21 28 0.583 2.10 1.22 29 0.866 1.46 1.26 l 30 0.601 2.28 1.37 31 0.547 2.68 1.47 , 32 0.773 2.15 1.66 33 0.824 2.08 1.71 34 0.948 2.03 1.92 35 0.851 2.32 1.97 36 1.05 1.98 2.08 6-9
Table 2 AVERAGE STEADY STATE CHLORIDE LEVELS - PLANT COMPARISONS : I WEEKLY MAXIMUM CHLORIDE, ppb Plant Geom Geom Upper Rank Mean (1) Dispersion (2) 84% Value (3) 1 11.7 1.60 18.7 2 20.0 1.00 20.0 3 20.0 1.00 20.0 4- 20.0 1.00 20.0 5 20.3 1.10 22.3 6 -20.8 1.10 22.9 7 20.5 1.20 24.6 8 20.8 1.29 26.8 9 16.5 1.78 29.4 10 15.7 2.08 32.7 11 6.9 4.96 34.2 12 27.7 1.32 36.6 13 31.7 1.18 37.4
-14 27.6 1.36 37.5 15 14.0 2.09 40.6 16 20.6 2.09 43.1 17 34.2 1.49 51.0 18 50.2 1.06 53.2 19 26.8 -
2.06 55.2 20 50.4 1.15 58.0 21 50.9 1.18 60.1 22 47.7 1.32 63.0 23 30.5 2.26 68.9 24- 34.'3 - 2.03 69.6 25 38.1 1.88 71.6 26 50.9 1.43 72.8 27 '42.9 1.73 74.2 28 26.9 2.95 79.4 29 '50.7 1.58 80.1
- 30. 32.8 2.66 '87.2
-31 (NMP-1) 42.2 2.12 89.5 32 52.9 1.95 103.2 33' 102.0 1.21 122.4 34 73.0 1.74 127.0 35 67.3 2.03 136.6 36 77.7 1.80 139.9 7 10
NOTES FOR TABLES 1 AND 2 (1) Five to seven reactor water conductivity readings, and approximately three reactor water chloride readings are taken weekly, on different days. The weekly maximum is found, and its geometric mean is calculated. Geom mean is k (n xg)1/k; also, antilog of arithmetic mean of log x$ values, i (2) Geometric dispersion is antilog of standard deviation of log x j values. (3) Upper 84% conductivity and chloride values are based on log-normality of the weekly maximums. The Upper 84% value is (Geom means x Geom disp); also, antilog (arith mean of log xg values + 1 std dev of log x j) values. Logs on either base 10 or e can be used throughout. Example. For {x j l = 2, 3, 4, geometric mean of 2.884, geometric dispersion is 1.417, and upper 84% value is 4.085. 6-11
Table 3 NINE-MILE POINT CONDUCTIVITY AND CHLORIDE TRANSIENT VALUES (MAXIMUM AND TIME-INTEGRATED) Percent
" *** "'* " A' 1
Year Dates Power Cond. C1 Cond. C1 1%9 12/16 34 1.30 250 0.10 10.9 1970 1/5-1/9 38 1.56 0.82 1/13-1/16 62 2.6 4.34 2/3-2/7 80 1.52 0.59 12/22-12/26 80 2.20 2.22 1971 -2/24 82 1.42 0.14 2/4-5/8 0 3.2 1.06 5/18-5/22 :0 2.3 0.10 6/5-6/9 2 1.95 1.48 6/18-6/23 66 +30 232 4.79 3.3 7/8 70 1.9 0.52 10/26-10/30 55 2.4 2.25 12/21-12/25 95 1.52 0.59 -1972 7/31-8/4 83- 1.40 1.29 8/28' 47 2.0 _ 0.83 9/14 94 2.60 260 1.37 23.3 9/20 74- 2.0 0.86 9/25 64 1.5- 0.43 10/6 81 250 14.8 1973 10/27-10/31 85 2.5 287 2.81 74.5 11/30-12/4 82 1.12 0.34 1974 4/3-4/7 0 5.10 5.32 6/18-6/26 0 3.4 + 8.87 7/2-8/1 76 2.0 220 3.39 5.4
.11/26-11/28 93 1.2 0.32 1976 3/21-3/22 51 1.8 238 1.23 28.1 4/17 0 120 20.0 4/22 63 2.2 0.96 4/27-4/28_ 85 2.2 212 0.34 0.24 1977 3/9 0 7.2 683 + 5.36 327.7+
6/7-6/11 0 4.0 315 1,67 299.4 6/14-6/18 0 2.2 0.03 6/21-6/25 0 2.5 0.16 6/28-7/2 0 3.5 1.08
'7/5-7/9 0. 2.2 0.02 7/12-7/16. 0 (hot) 1.6 0.68 1978 1/17-1/21' 71 3.0 7.'35 TOTALS Reactor at power: 40.02 -160.5 Reactor shutdown: 23.67- 647.1
- Units Conductivity-(uS/cm)- * + Highest value
-Chloride-(ug/kg) ** Conductivity (uS/cm)-da 'above warranty limits Chloride (ug/kg)-da above warranty limits-6-12
Table 4 CONDUCTIVITY AND CHLORIDE MAXIMA FOR BWR CHEMISTRY TRANSIENTS Plant Rank- Plant Reactor at Power Plant Reactor at Power Chloride
- Conductivity **
1 7 # 19 2.0 2 8 # 18 2.7 3 12 # 17 5.0 4 15 # 11 5.8 5 16 #- 16 6.0 6 19 # 15 6.2 7 1 (NMP-1) 287 20 6.8 8 14 300 10 9.2 9 13 320 7 12. 10 18 355 14 12. 11 10 370 8 13.0 12 4 374 2 13.8 13 6 400 4 14. 14 2 495 12 23. 15 17- 495 3 25.6 16 11- 600 1 (NMP-1) 30, 17 20- 725 13 40.5 18 3 1200 9 72. 19 -9 3000 5 84. 20 5 14500 6 95. o units: ug/kg (ppb) M units: pS/cm
#_No data reported in excess of 200 pg/kg.
6-13
Table 5 CONDUCTIVITY AND CHLORIDE INTEGRALS FOR BWR CHEMISTRY TRANSIENTS Plant Rank Plant Reactor at Power Plant Reactor at Power Chloride
- Conductivity **
1 7 0 18 3.2 2 8 0 8 5.7 3 12 0 19 6.7 i 4 15 0 10 7.6 5 16 0 11 9.5 6 ~19' 0 7 13.7 $ -7 13 20 16 14.4 8 18' 32 5 19. 9 4 120 15 22. 10 14 120 17 22. i 11 10- 140 14 35. 12 17 140 1 (NMP-1) 40, 13 1 (NMP-1) 160 4 51. 14 11 300 9 69. 15 9 1250 12 84. 16 2 1260 20 96.5 17 6 1920 13 99. 18 -20 2770 2 112. 19 3 2810 3 206. 20 5 3280 6 530. f f i-t ' i
- units: .(ug/kg) - days above the specified limits.
** units: (pS/cm) - days above the specified limits.
6-14
Table 6 ) CONDUCTIVITY AND CHLORIDE INTEGRALS PER YEAR OF PLANT OPERATION FOR BWR CHEMISTRY TRANSIENTS Plant Rank Plant Reactor at Power Plant Reactor at Power Chloride
- Conductivity **
1 7 0 8 0.9 2 8 0 10 1.2 3 12 0 18 1.2 4 15 0 11 1.7 5 16 0 7 2.6 6 19 0 19 2.8 7 13 4.0 5 3.1 8 18 12.3 9 6.2 9 10 21.5 15 6.2 10 1 (NMP-1) 28.1 1 (NMP-1) 7.0 11 14 30.0 17 7.8 12 4 40.2 14 8.8 13 17 49.9 20 12.5 14 11 52.6 4 17.1 15 9 112.9 12 17.3 16 20 359.7 13 19.8 17 6 410.8 16 30.1 18 5 537.0 2 72.4 19 2 814.5 3 83.7 20 3 816.6 6 113.4 t o units: (pg/kg)-daysperyearofoperation. 00 units: (uS/cm) - days per year of operation. 6-15 l
O _O pyg 5
\ c.n ECCQH hg 2>
a JdW cSd n4_-% - p ze d . 2 x y*~ M.---.-- - wa 2 M m.
.m, ns.-
c r~.-q q n O .,
, 4 -
7 c O
.--- ,,_?"
,, 5 s
f Vo
< #p(__ mg g itr.E ~
%u-
& #116 w= - -
(: c-- +ct . f (K
-+W*
i 9-. - S 8%EN_ O C ~ oE
, s-/_ m 3 I A.;
3 4 O t WY o
"J K aM
~~ 4 b y Om g e > W r<
3 W _ a= o N> ail- - -2 N
.-,_c 4 _
C "' [p. ; x - 7
/
/ e
=
t' c e-t o
. Q,,
Wyg - ( k
> LaJ
-.h W
Z O O - Q O O e e.
- 8 d
(W3/0HWn) *0Ad AllA11300NO3 831tM 801303d FIGURE 1 6-16
O 3 O i l _e i
; a -e o
o i p QC* N
\ )N C G
----- a_
m
.= ?
A c3 Nh (
- sr- 1 W6-h e r%g O
C') g _ _ _ , er
/
gare > y ;fY D L4 r> ,g s cg Y w__-% i (rk 12 = ,,
* ' 3
>~ N 9
y (js N O d
a (- -+r ts dr
& d d.-<:-,
- _
v 9 W n . z d e M . ay a b 12.1 _---__ a w . .M[ i[-__, .e , z h E W _
.s<rf ,
~
L3 O- o o o ,5
- o. o. o. . o.
o o - O' (W3/0HWG) *0A0 A11AI1000NO3 831dM 2:010033 FIGURE 2 i 6-17 j
~
,J
,' ' i< " 7 7
)
;i Ji6% f nlJ 94W 4
%i '
f' [' [ _ i4 k . 4 LJ~ 6 1 0 7
- c. 4 2 1
,l;ilN
, L V~
3 7,14' .[ 8 J. ,I / f_ ,-' _ 7 n 0 E
/ M A 1 A
. 0 FR 3
,n3lt5h' l 1 5 f
,91A 7 L
'1 1 p4 d iI~
o 1
/f f
pillHl$** 0 slj i f.
,5 4$
4 7 1
- IS T
, i,'
k w cL6%9l t 3
$ I
. ,I' o
W 7 4 1
,lpN' -
T 7 r 0 S,Ui P JIE I l i %
- d. AR 7 E S Y E L
7 II4 ly C Y C fu
!t P
n ! L f" s L 1 A o ( ' isWll t % 1 1 7 o d." l
\
M
\ T i T N
_n l" / a . I O P
' 3 0 a yi 7 E L
I M
,ll E
N I ~ N 9 0 0 0 0 1 6 0 0 0 1 0 0 0 1 0 1 1
. g5 dE EhN9S Fo5 !UPu" s g@W
?5
!i
n' O
~
3@ c w N I . _. gM-t* g
'N, w
N
)u s x
- a wwg N* ;
a 21
- i. , , .
g - a o2Ag A > trm o p a~~(-__._._e So _, l
- N., , o w
_m @ f f, e
/
' / y -# _
5aF c
, Tk
<-=. -
c
-4%
3 =%_ o-- (; - -- O - ~
.r- .c,ito* .R H. m.
4 4 Iw 4 cv -4r " N
\ d a
~ ._=g e- -'" " ti c-3Ey= q._ AO l'
ar_ffy..
- o-cme c NR _
l t+A r a h i
- m> ts u
( 9<_._~~~~ y E y )b I (; -, c( r _ _. [ y c ; c* h E !. %c -e l - - .+ A*,b, _ N, . 7 O o o o o 9 o - 9 o J o o - e (W3/0HWR) *XUW A11AI1000N03 631bM 8010b38 FIGURE 4 l l l l 6-19 1 l
- _ _ . -- _q s f ts (W: o l O M---64 guw-FL --1 ,
C :( T i e
;4-_.
f %. % ,
+
- e-
. ::t c ) 5 Jm
<F-x -
>mqcF
- m m-#-g,- - q
+ s di 6, o u l' ' =. m 4
o
', E oew I >. ;
p+ amo 3 F m-B s P wpf *
<m s
Yd? " -r C <.-
~ .-.
.y 3b _
( W#7 1
- w .s a.._
%+
= ..
; R x
. , - .~. c3 %
,'. N C c)
#- , - w 0
l > A w a n gf a g x g o. I k% N h , $ i HC ' s'
# E il o
6 w s r w z z c)
. . . . . (D o o o o -
o o o - o o - o (8dd) *0AU 30IMOlH3 83'"M 6010038 FIGURE 5 6-20 l
r-$
# m'n d I.-
CJ D ' 3 cv)
%=
i ,A - +4 $ N
\ 5u N E N ,
~c 4 j rd ~
' w t
1 Q C ) c 4rg r b w
#J $
D 4 h .
/
*c 4 d
c$
- 8 d,S:"?
Mc,
~~
cr-R (*
%d*
C ._ i .
't o e-
'(
e4 U2 O O os >
)
O. O j a ~ c i
~ l "
O k ..$ (D:7 P= et?% ?
-+
ggg H-sau mg w C 2 45 ^ Ej 3 r * (*n g 1 C } - N (d( N S gg= -:n.e E } n. f - U b o 2 oq N4r-g . e
. . . , .s
- o. o o o -
o o o -
'o o -
o P1 (8dd) *0A0 30ld0 lH3 6310M 8010033 FIGURE 6 6-21
l b
.- 6 4
w tr~> M) g c ed(CT' o __h r a
.m y n .
w-wy- m*> o,, x "$ m w I?t Vf= 4 m a. m so u tM4 >
- s ra
'.t
-C==-..o ,o{
M& ' 0
,!yp
(*- pro
.- . d.~
M 4 P a i(E$--e - 5
^*
c >
..bh m .
-Wsh jh_dM Jt ;;_' : -: <f g-.-{-'v-=yg t b C .>
(Mr#Q , c I N e-" k; czc, m (m r pw
'~TA:a t, d c
.. __}'
i 1 i , U _a cm... n. [W \ ) c n ( s
-h" g
s s s z W -a----- o 0mr!is CL
' ' f o - - -w ,.
s u, r w i z z cn
. . . . O o o o o -
o o - o 8 o. (8dd) 'XUW 3GIb0lH3 b310M 8010d3d FIGURE 7 6-22
. c5 a
I a % ._
-. l i CJ %
& * :b -r i
i 2 $+"2 ~ c.c .[w rar I aN x N t N EY1 th
' ~
0,6db
- 0 62 a
Oh,- >
. $ a
~ll
+c (M :fA,
- u. f b
)& W (* ,
- l
%_ EM _c
/
/
l % > o 4*) N (Q * )
'+- ft
%s m-% --N5h
,,o m,-
rk a:- wW h) a,,t .a o w n m --o O
- p*T >-
O
\ L..
J k T p@@ rs
*> h gW
%Er: .
3td
- h*
g g Ar v
+~,'L 1-- w C 3 O-9 &
W " o e- ,,~ D j 1 R hw-g
]
~
r o .Y;E-.].. W se Z NN s f
<* 4. O
>t. "
ln o
.. * *N o O O ~ .o O O e O O m O m M
(8dd) 'XUH 30ldO1HO 6310M 0010033 FIGURE 8 6-23
--g Lbl r M2=
\" a
[~4._ R l
?Y l 5 '5 ;
\
c , s
- l i
,g 5
! [ _.
3 M
$~
< g IC
' .m
-X 29O d
G d C N r- b N z
) j!
n C . rs E f W s 5 d s' 6 g . m
.8 U 2 =r Hd b310n 3013y3y FIGURE 9 6-24
,1 1l l!l 3 _
U 1
' . 0 E 2 5 5 0 V J 3 1 3
0 R / 4 1 E
/ M 1 A 0 R F
;i 1 8
M
' I q '
\
Rp 0
' 0 g '
t n iaY
'J W 0 7
oT I
' . 4 R
L 0 DS f 7 Y EL C Y
- C L
L A 7 7
- T
- N I
- O P
0 E 7 L I M v E I N I 9 N 1 1 0 1 9 8 7 5 4 3 5 7 5 x S $bS" 1
,E s n
c l
G W,-C_?_:_. I
.-v b
- (-
<+Wl 9' Dr.- g c 1
?! h>
t t . uo
<--u__
o~ b ;j o H) r0
') C)
=- -
vn b oy L vm a o " o- m+- C;~Tlf - w h o _ -, k r~-~ - G ae p-" pmcC * (
%v a ,
'k N
Lee ow+y e O f;--'-' C O...
/~
.e Hp,_
e g y2 = h, _ I A b v T...-4"'- h PilII ~., . _ h r,-=_ .
=--
o-n= [4## a
= n = = O n;. y -
i
't t
w .33 o f
'.4>% '- K U
w ,
--_ ia -
' ,1 F1-r a es .e o ee
%se .
v o b o m -_
.--"~,.s W
y_r-m""- - O Z 4) 9 "O eAs-
- Q.
t-
% 0 GP b U d
! U t Z l' m Z b 9 e . . Q O O O O 9"9 O O O .-e e O.-. (Bdd) 03171S B31eM E013d33 k FIGURE 11 6-26
W a 3 l
%m .
~ c, p4 g .
o .s a-
% O g)" ku
- - ~ .J
\
- a b
cg -* a; - N a.se II p m r -__ . ..-- k u_-gg km n ,
~~-- ._.
g(#_ ,
~~~ - --.
a
..' +7 w .g o
+JIL- y;;
=- o 7*
===.o- ~. ."
,e --
.___2s q_'25 '~
Tr onL h l g a..,3aq
<> d 4
}' ~-o
+.-- au3 x (PW 3
i >- O N. J
\
4 9-o-.
-v, ,
+.
. o Wl C -
net
- 4' _
bb b ui o 1. (ma
]
- s u J
% "d> {
____u_ .., ' Mk &j
=M-L =
L vn =
'k I? c a~n a * . , * . Es O o O ~ .O o o_
o - eo O.-* (8dd) 03I7IS 8310M 801303d I FIGURE 12 i 6-27
4 l Il
!! ., i l
! 9 :
I < j : ,
; )
- i
\.j 1. .
1-g 1 ~; 1 j' l 7i
; i fll :-
a 1
,i .
ll
. ! l! !l i l &U l l
~%
a) l i
\
l T l 1 I l q P,# , I i c * .1, j i i i , K s i i o-d I
!ll n
,- !! l l l j i o I i l NI -l 6 l ill ill < l2l 1 i
! l l lP l
I- ! ! ll l
}hl .
'ly l~ l l l l il fl $
j l '!! a jjl!
- i!.I Li l l l .tii l i i W '
! ,ll I j' !!'
ii j ' i n s t l'lj,l! h ;
! ! !. l :
!. A I 3
[ +74 i
, i l l E i
t i l : E
- l I l i - !
'i . , 8 8 8 8 8
~
S 5 '~ R R R R R R 8 8
~
2 ~ S (W3/sn) *0M *CNO3 b21t'031 : FIGURE 13 6 .-
- . .~ ._ - _ - . _ - . . ,
a 4
'D j
r) N O
. . "> ~
M N e N i .; I N 34 a
'i; 12 2
Q t d
- .o j O
l U i l t t g ju a I
' l l
a a
- e. n I
~ ! .tj l,
su TWrl^
~
l1 N J > O I > LJ 4 ( J a o
) \ >~
\ u J
l d I e E
.J O
s-Z a .=* O
" , 1. $ .:Ns u y
i . r
.a s Z W
k - a O O O O O O tw O O O O O - O O O. O. ' O. . :. O. O. O O O ~ O O m O e e (WO/SA) *0A0 *0N03 2f310MG333 FIGURE 14 6-29
Inlm!!l! lU t ll
!!!i'
.;!,,;j lhl!I i i Y;l ct ,-
l r e ; ;, I f,i l ,; :!l:o
! ,i 'l, I ; ;i W:: ;i i.i-
' i i' l
>jj ;
, l hij i ! l l.l m
i, i
.I I %,
1 i l! il ll m!n l "Tin
#W i i i i " @ii H 'Hi '
l il! l l l l l' !l I'N
!, "l
'fR
- l.l ;u!!' l l'ii <I.
m' a li , j i ll l :
-i ei 3 ;l, 4 l .-
I! I-i i [jjj jm j
~
I i2 - i, i I" A..sllliTr ll' el ji 0 _/sd
' - Y l H
}i '
l - i liil' i l inlj
- il p ! , <Jr i l
N ji j , j t Q -- 8 i! I m. :n
}
l l l %,l 'j ?!lI
. 1. . l I !! I lib ll l ._
mp q1i, in
'il n!
nnI iv
!!: .I i Ic-I -I ,-e;I:i vill ! '
t, 'il' 1 lh l I :n ! c i i. ' on: in I ui., ,, ,, i Lj l :ll l' aml! lil :j "
!l ,
l, i,j I, p, [l i!, 4 tlull
,:. : 1
..s i c
- i , i, !
- r. , . .-d e a
' l
'!!!l;l l -.
ai: i f q: i I i 4 r:i, i i i d,. l i' 'll; ' !ii ! it *
..j, i . i
'!' j ' ty ' I li ' ' Q ,j ii i i;h '
Is . ' Nj ,i l : ni lin illl :. I.H ; . '
'l ll i; j , :. i i ij- ?; i i i
; j la p o i i i , ,
I l ll1 ril';l
;;. i
';l!
ii I; I I!'l i li, .' i ,
! n' l . ,l l,
, :t, ,
,t, ;
il I ll j!: ll i 1i'l l ! 'c : I ll ! !! lil:. ll l ! '!!'l I O O O O O O 8 8 8 8 2' 5' 5 2 5 8 ( W3/I,1)
- Xt.; 'CNO3 62n'-
FIGURE 15 6-30
u i g < Il
~ ^ Il r s O q
r
'a + l
- J
.,5 .
c ' o
\ :
% ;;)
' _., ^1 .
I
'. 's OV l
2' s I (M.. ll .
.-._ b
-An '. )
cq. 31 Yl- I d I' W ll lslr
~ ^
!E N. ..
-w , .
f ! *) A P
. .)
le .; q Y
~
-I s
b h _J ' I-f I a 6,4 r ;' c, I l O- j; 2
! 2
,l
[;li x
!I' D a JL =y s '
V:' -- a ll ~ I flr LJ
) Z E
i <. .
;3 O O O e O O 5 O O O O ~ O O O.-*
O.
- O. O. . O. O.
O O O m O O ~ l O ~ e (W3/SA) *XUW *0N03 8310M0333 FIGURE 16 6 _.
. - . ~ . . ~ . , .- . , , . -
t1
/ .\
!i '
\
f c.,. .- I ( l e O-n , 9-% k * .
>M .
l hl
~2 a , l 2 i
p;l
.a 1.
f , l J. As i , g[
- M i
h_ I cgT a
' l ,, Ih
' l l
I I ' l [ l ']
> 1 W -g .,
. e
]l, , i'!! i !* -;
.if I
il, hi t t llf.
;i .' ,
I . j f k
,hh3 h !
l ! I! i (i D 'l liii } ! l
!!l l Il 'll ,
I l li i4j ! i g llll I gli t_ , l i l,. g l 3 I. l :
. ml ie I
a l g
.=
2 I J o o ,? O O o o c o o o o o o c 9 . 9 9
~
8 8 S o.-.
- (Edd) 'OAd f139AX0. 2i31tMC3I -
FIGURE 17 6-32
. _ - - . . . . . _ _ _ _ _ _ . - - _ - . _ _ . - - _ . - . ~ . -...._n . ~ ~ ~ . . _ . - _ - . . _ - . - - - - , - . - . _ _ _ - -
1 () l 3 f l i . NVQ -
'o f cQ . ts lr O
4 s 2 g a t .._ OM k , ca l l 1N 4 1
'l is O
4 I . O 5 i b v (@ 5
\ x i s s '
4)
"~
gf- . < 3 r-o 4 I c > c %
< k l 0
L.J
$$ a u
O q
/ .c-h
-J
.s y-
%, , a L
O <- N s i-D Z aM .
\ u 7 %
) E LJ t ' Z C O
> n f .i a v? I
^
I L) O O O O O --s L' O O O O O O.-. O O. O. O. O. . .e o. O. O O ".* O O .C 1 O.-. (8dd) *0A0 N30AX0 d310MG33J FIGURE 18 ,7. 1 6-33
c" k-l
;'__ -:s t.
l c ,;e* ,!, j l , c ad pll e I]I % ,l lAn* i,ldl j; wr N- -.
~lif; I 2 T.wD mt l lll1 ,
i'
,j
~{
l } i cjL%:TY 'I -
- w -~ ~-.,% .
p E l . 2 4 .' , s ll l
- N l i +l o ag i I$'
g p-c .
+~ , - .
C
"~~ -' i
/
?@'
d JC fI l l; 4 i
% 3 , ) .;_ ___ T~~~ l l l ocq ) l la j l )i i,
bl ' l! 75, ."o , ll !! i jI' j~N I 1' i ' , s li,i ; ! i! l
;t c M . ,-
' II illll%f i d:!!![ j A
i ., I Fl ,- )' c~m i - ii
. U I , , ,1 ,
1il i l4[h7A
~
!! l I'!!!!,, .
ii-i 3 % l sl ! ! _j. I 4
!l ll
'Nf i 16 i
he l f I,!l ' ll l 1 I
. M%.
li pii- ..
' ~i tilli k%a ,
.I t ,i ,;
il s' I i : .;,,u .: l1 l p,, -
/ .
i ,, ;
.e.'
,- e
>I
,i.
,;l p l lI _
. e_-4 1 4l p;;; ll 3
@e
~ -
I ci l t I,i .,i,i
-r 1
ni .
~
l!l i
- l. l ::
t I bd ! ..', 3
'l ,! i i l'!!
~
'l ! l ,l l lll l O O o
8 o 8 8 0 5 9 o o o d ; - - i 8
~
2 - f .j (8dd) S1013W 10101 E:_ ::; FIGURE 19 6-34 l
-.G. -
t I') U
.%:? -
p: ,& K
* *% _ _=.2 i
s c.m ,
% M
% ~ , km N
N " CE m ( vim * ~ kN h g.~, ~,
,-,%. l mo UTi v K .. ~
- O
*'4= jy Jc, m NI
\
s
; h
- 0 C 3 c-u -- lc ~*(LI*
Mi I c-- !t % if?4'
- w ~ , ,Y, ,
__m-y_ g ul tn 4---: hi e w J O a 's a O S . = o.,,ae ~ _.m 'QMW)
}'
C i !-- '
]
(m+" 4 -* C n R c=
-- -p@ a 0
cz c. 3 N--% *4 ~ LJ J 1&=
@w -
(:) p - 18
=
2 a 'y% -
%~.T g o _
O O 8 o _
- e
- o. . * - R R t o o o 8 o.-.
(Odd) S7d13W 70101 831bM0333 FIGURE 20 6-35
,a- s -. ..a4 s._ m_ . . -m w.2 .;.__ _ _ - ..._.._m.+ -p ..c._aA._a_.__,,____A..____u.-,a.,,.2.A. y 4 _a -
)
e l lll m; CD . l
, i f i s 6 . . i..M .- l I t illi i i rQ l 1
- bl i 41111
- 40. . . . Ii I' # lli il ^
l I
' ~
Ngll 'i '
~
- TFai i .
I '~ ll llLW
.#.nll' ' i:j u
% l l.n
- . i 4 li ,
e .cw ;
% li . ,
~
' lYlll 4 NL c u
,In ll~ im s:p i
i l;l . lcF6 li i t s i I l r-, i l -':ri;l i . iloIi-e l l 6l i i
, I
, e I b l : l n .70 J l
'\\\\ll\$l illl\ \ \ '
! ,4l l!! I l f
! 'I l
onu ll e l l ' l #d s r...;i j' I; g) c- i f, 7
! l lil . p : . ! I, i I.
l
'w :+w
. . :i i i l
! '9 i .
i,i. I i j ; -
'N .
i i . h} ll Y' l!
"Nl ,
f ' I Ih_ . N :, i
\
[' ' : , . i . < f'I 5 Ni. - 2 p- -
,i.,
l j l Dih j
,! Jo i
. 4-i
!i 't O O o , '
O o - i f (8dd) 10 :01 n3 c2, - - - - FIGURE 21 1 I 6-36 i l
M l' D r I tf .
- ca a N o v>
C hW ~ M
'h..' O s
2 O ( a
..,c # O cx h I c)
= %<-e ll l WW
~-
. 3
~lUpF>
o s -QQ, , j;f eb o gm o co
%k4m C ._
N- , m
.~.,_ o l +
_C - _:_0 M b (, -
.i
,y i',.
<- M --_.+ ~
Q ,l
- )
,\ jh -
5 0-4 ._ ',- . _ ju ,
#t" i O hy
.y h4 5 N s4my g
- LaJ Cs, - ,~ . , g l' '
s s O N O p J d w,- 7o , C ,____- ;^* O N i
~ A--
0 - nh - U f: 24 { s LJ C. ;C [' <-1L .J z a# 2~ f LJ
< z G g. rll.
j q l
,wp--, --
Mt!T rl l . O o o o O o ,6 O O. O O o Q O - o
*
- o. o, "
O. O o o ,,,,
- o.
O O e O.e e (8dd) 7t1101 03 831eMG333 FIGURE 22 6-37/6-38
SECTION VII EFFECTS OF D'<0NTAMINATION l INTRODUCTION The IGSCC incident on large diameter Type 316 stainless steel recirculation piping at Nine Mile Point (Nf1P) Unit 1 in March,1982 provided the motivation for an in-depth analysis of several potential contributing factors to this deterioration in corrosion resistance. Since the NMP recirculation pumps were decontaminated by the London Nuclear Ltd. proprietary CAN-DECON process in April 1981, and the entire recirculation system in April-August 1982 period, it was considered prudent to determine if this process had any impact on the IGSCC propensities of the recirculation system either during the process application or sub-sequent operation. This section treats this concern. Prior to discussing the potential impact of the CAN-DECON deconta-mination on the stress corrosion cracking tendencies of BWR piping, a brief description of this proprietary process will be presented below. The CAN-DECON process utilizes a low temperature (85-90 C, occasional-ly 110 C) dilute decontamination solution (0.1 w/o) of acidic organic chelating agents (pH 3.5-4.5) plus excess hydrazine for oxygen scavenging that is circulated through the target system and through a purification
. system of filters and ion exchange columns. The loose crud is deposited on the filters while the ion exchange system cationic bed removes iron, cobalt, nickel, chromium, molydbenum and any other heavy metals that may be present. The process is terminated by removing the reagent on a mixed 7-1
bed column. Prior to returning to the nominal operating environment of the reactor, the target system may be re-passivated by the introductior. of ethyenediaminetetracetic acid (EDTA) or oxygen free water at 150 C. Specifically, in the case of the Nine Mile Point decontamination, two CAN-DECON decontaminations, separated by an oxidization step, were a performed on recirculation pump 15, the single pump scheduled for inspection. This oxidizing step was designed to alter the morphology of the oxide films remaining after the first CAN-DECON treatment so that they would be more amenable to dissolution during the second CAN-DECON treatnent. Only one decontamination was performed on the other four pumps, plus all 5 recirculation loops upto just below the vessel safe-end level. Decontamination of pump 15 lasted five days while the other four pump's j treatment lasted three days each, q DECONTAMINATION AND STRESS CORROSION CRACKING i i A. Laboratory Studies Takaku et al 3 investigated the ef fects of two undefined decontamination solutions on the IGSCC propensities of Type 304 stainless steel (0.07 w/o C) using test specimens fabricated from four inch diameter pipe weldments. The SCC tests were conducted in 4 pure water at 288 C containing 8 ppm oxygen utilizing three testing I techniques; constant extension rate test (CERT), cyclic tension i test by load controlled trapczoidal stress. waveform, and constant load test. Table 1 presents chemical cleaning conditions of the two reagents. ! 7-2
1 l It appears that Reagent A more closely approximates the CAN-DECON decontamination solution relative to concentration and temperature. The most striking disparity in this study, aside from not knowing the exact composition of the relative proprietary solutions, is the dissolved oxygen concentration. The results of the CERT SCC tests in 288*C pure water containing s8 ppm oxygen are summarized in Table 2. Various stechanical parameters are utilized in the table for evaluating the materials susceptibility to SCC. The data indicate that there were no significant differences in these mechanical parameters between the as-welded specimens and the chemically cleaned specimens (either Reagent A or B). It is also important to note that the IGSCC propensity was not increased in both reagents by either severe decontamination conditions including the con-tinuous long term cleaning of 72 hours, the extent of the decontamination cycle (three time cycles) or the presence of crevices and residual reagents (pre-cracked specimen). Table 3 presents the results of the cyclic tension tests which were characterized by a maximum holding stress of 175% of the 288 C yield stress for 30 minutes and a strain rate of 4.2 x 10 - 6 sec . A review of this data indicates that no notable differences in IGSCC susceptibility was observed between the decontaminated and non-decontaminated specimens. The final Takaku et al SCC study, constant load testing, was conducted at applied stresses of 150 and 175% of the 288 C yield stress. The data, as delineated in Table 4, indicates the same results as revealed in the two previous studies, that is, no sig-nificant difference was ooserved in IGSCC response whether the 7-3
l l specimens were decontaminated or not. Therefore, the Takaku et al data clearly indicate that their chemical cleaning solutions will not increase the IGSCC susceptibility of Type 304 even under ex-ceptional cleaning cycles. Two different types of SCC testing techniques were employed by 4 I Ishii et al in evaluating chemical decontamination cycles for BWR i primary systems component materials; pipe testing and creviced bent i beam (CCB) testing. The pipe tests were designed to evaluate the effects of decontamination on the propagation rate of pre-existing surface cracks in the heat affected zone of primary loop piping weld j joints. The objective of the CCB testing was to determine the effects of residual decontamination reagent on the SCC behavior of Type 304 i stainless steel (0.07 w/o C), Alloy 600 and X-750. Two types of pipe tests were performed. The first pipe test i procedure was characterized by a decontamination cycle with Reagent A or B (Table 1), followed with rinsing by operating the main loop with warm deion-ized water and followed further by pipe testing at 135% of the 288 C yield
, stress using a trapezodial wave form similar to other pipe tests as described elsewhere.5 The second experiment was characterize'd by pre-cracking in pure water for ten cycles followed by the decontamination procedure used in the first experiment, and then pipe testing.
The results of Ishii et al's pipe testing is summarized in I Table 5. No significant difference was observed in the time to failures (TTF) between the two types of pipe experiments (precracked and non-precracked), the type of reagent used or whether the pipe was subjected to the decontamination cycle or was exposed to only pure
-8 ppm oxygen water 7-4
The 500 hour CUB tests were performed on furnace sensitized Type 304 stainless steel, Alloy 600 and X-750 with compositions and heat treatments as presented in Tables 6 and 7, respectively. The assembly
- of the CBB rig with graphite wool as a crevice former is present in Figure 1. The test solutions initially contained %8 ppm oxygen and j
residual reagent at 0.1% of the specified concentration for process decontamination. Metallographic examination at test termination revealed that all three materials experience IGSCC in high temperature oxygenated water without l the presence of the decontamination reagents, Table 8 and 9. In fact, the presence of a trace level of the chemical reagent appears to be bene ficial . The cracking of Alloy 600 appears to be totally mitigated while the depth of cracking for Type 304 stainless steel is slightly reduced in the decontamination solution. The cracking propensities of Alloy X-750 is also dramatically reduced when the residual chemical reagents are present. Ishii et al suggest that since Reagent B acts as an oxygen scavenger, the mitigation result is not totally unex-pected. Reagent A does not reduce the dissolved oxygen significantly, but still appears to have some beneficial effect. Apparantly only limited SCC testing has been performed by London Nuclear on various reactor structural materials in their CAN-DECON solutions. Test specimen types included single and double U-bends and four-point loaded bent beams. The primary material evaluated was furnace sensitized Type 304 stainless steel and other austenitic materials. The standard London Nuclear test sequence consists of exposing the specimens for 28 days to 288 C oxygenated (0.2-0.3 ppm) deminera-7-5
lized water, decontaminating in a special loop, and then returning the specimens to the original autoclave for final exposure. , The results of this testing indicated that precracked sensitized l Type 304 specimens were characterized by similar crack extensions regardless of whether the specimens were decontaminated or only ex-posed to the simulated BWR environment. No crack initiation was detected in non-precracked Type 304 specimens. However, a slight surface etching was observed on all surfaces of this material exposed to the CAN-DECON solution. Since this morphology was observed on sur-faces which were in compression as well as in tension, the result is attributed to straight chemical attack on the metal oxide by the reagent, and not any stress related phenomenon. B. Operating Experience Prior to the April- August 1982 Nine Mile Point application, the CAN-DECON dilute chemical process had been applied to five BWR reactor water cleanup systems (Vermont Yankee - October 1979, Brunswick II - March 1980, Brunswick I - April 1981 and December 1981, Vermont Yankee - October 1981, Peach Bottom II - April 1982) and five primary recircualtion pumps (Nine Mile Point 1 - April 1981). This process had also been applied to systems of the NPD 25 MWe reactor in 1972 and 1973, the full reactor system including the core with fuel in place of the Douglas Point CANDU pressurized Heavy Water Reactor in August 1975 and several subsystems and isolated heat exchanger from various other CANDU reactors. To date, there has been no evidence of corrosion of any material or malperformance of any ~ reactor component which could be clearly associated with decontaminations. However, no data are avaialable to verify longer term materials performance. It appears that the only effect of the decontamination process on reactor components clearly identified to date, aside from what the solutions 6
i ( are designed to accomplish, is the effect on IGSCC detectibility. For example, dye penetrant examination of safe ends at KRB following cleaning in water by soft brushing did not reveal any indications. However, dye penetrant testing following chemical decontamination by an alkaline permanganate/ water rinse / oxalic acid / ammonium citrate solution at 90 C revealed crack indications. ' Prior to the ap-plication of this chemical decontamination procedure it was established by Kernkraftwerk Gundremmingen Betriebsgesellschaft that the chemical reagent did not affect the austenitic material and no new IGSCC was produced. ,10 DISCUSSION The study reported here was undertaken to evaluate the effects of the decontamination process per se as well as the effects on the subsequent stress corrosion cracking response on typical BWR structual materials. Although the volume of relevant available data is not extensive and principally short term . in nature, the data support the contention that decontamination does not adversely affect the stress corrosion resistance of austenitic stainless steels. The 3 investigation by Takaku et al and Ishii et al can be cited as evidence for this assertion since a wide variety of stress corrosion testing techniques including CERT, cyclic tension, constant load, full size pipe testing and creviced bent beam did not reveal that decontamination cycles have any acceler-ating deleterious effect on BW plant materials. However, longer term tests that might reveal more subtle effacts have not yet been performed. Operating experience in BWR's after decontamination appears to be 7-7 w - - -
l l E satisfactory with no indications that the decontamination process con- , 1 tributed to any subsequent deterioriation in corrosion performance. The only documented effect of decontamination appears to be in the area of inspection where IGSCC in decontaminated components (see Section 8) appears to be revealed more readily after cleaning. This factor could potentially be misleading in that one might assume decontamination has created the cracks rather than provided a means for easier detection. CONCLUSIONS The above discusision concerning the effects of decontamiantion pro-cesses on the recent identification of IGSCC at Nine Mile Point indicate the following conclusions:
- 1) Laboratory testing by the reviewing investigators, utilizing a variety of stress corrosion cracking techniques, indicate that decontamination solutions do not produce IGSCC on typical BWR structural materials. In fact, in some instances, some decontamina-tion solutions migitaged IGSCC.
j 2) Operating experience in various BWR and non-BWR plants with systems or subsystems decontaminated by the CAN-DECON process has been satisfactory to date, with no indications that chemical cleaning has resulted in malperformance of any structural material.
- 3) Although decontamination may result in more readily identifiable defects in reactor components upon inspection by ultrasonic or dye penetrant techniques, there is laboratory evidence that decontamina-tion in .itself does not contribute to or enhance the extent of cracking due to the processing per se. However, longer term tests that might reveal decontamination effects on subsequent crack initiation have not yet been performed.
7-8
- 4) It is unlikely that the decontamination of Nine Mile Point I recircu-lation pumps in April 1981 contributed to the subsequent identification of cracking of the recirculation system discovered in March of 1982.
Futher, the 1982 total recirculation system decontamination does not appear to have exacerbated the observed cracking. The only clear contribution decontamination appears to have had on this system is the increase in sensitivity for inspection of cracks which already existed in the system. , 1 REFERENCES
- 1) J. C. Cutt, " Laboratory Examination of Sections from Type 316 Stainless Steel Recirculation Piping - Nine Mile Point 1,1982" PMT Transmittal No. 82-178-41, November 17, 1982.
- 2) J. E. LeSuff letter to R. S. Tunder, " Decontamination and Stress Corrosion Cracking," January 29, 1980.
- 3) H. Takaku et al " Effects of Chemical Cleaning on the SCC Susceptibility of Type 304 Stainless Steel Pipe Weldment in 288 C High Purity," paper presented at " Decontamination of Nuclear Facilities" International Joint Meeting ANS-CNA, September 1982, Niagara ioH , Canada.
- 4) H. Ishii, M. Yajima and K. Hattori, "The Effect of Reactor Decontamina-tion Reagent on Stress Corrosion Cracking of Austenitic Materials Under High Temperature Water" paper presented at " Decontamination of Nuclear Facilities" International Joint Meeting ANS-CNA, September 1982, Niagara Falls, Canada.
- 5) R. S. Tunder, " Alternate Alloy for BWR Applications," Final Report EDRI Project T111-1, WE 8102-03, September 1981.
- 6) P. J. Pettit et al, " Decontamination of the Douglas Point Reactor by the CAN-DECON Process" paper presented at Corrosion 78, Houston, Texas March 6-10, 1978.
7-9
- 7) H. Flache, "KRB-A Experience with Ultrasonics for Identification of Intergranular Stress Corrosion Cracking in Heavy Austenitic Steel Components," paper presented at BWR Operating Plant Technical Conference III, Lugano, Italy, October 30, 1978.
- 8) N. Eickelpasch, H. Bertholdt and M. Lasch, "In-Situ Decontamination of Components of the Primary Loop in the Gundremigen Nuclear Power Plant," Kraftwerk Union Laboratory Report, 1978.
- 9) H. Meier and H. Peterreit, " Possibilities of Testing forged Austenitic Steel for Intergranular Stress Corrosion Cracking with the Ultrasonic Method," Kraftwerk Union Laboratory Report R 413 No. 42/78. February 3, 1978.
- 10) R. Riess and H. Bertholdt, " Chemical Decontamination of Reactor Com-ponents" Kraftwerk Union Aktiengesellshaft, Erlanger 1978.
7-10 , I
Table 1 l 3 TAKAKU et al DECONTAMINATION REAGENTS { Reagent A Reagent B Concentration 0.1 5 Temperature, C 125 90 Dissolved 0 ppm 8 8 2 Cleaning time, hrs. 24 and 72 24 and 72 7-11 I
Table 2 CERT SCC TEST RESULTS, TESTED AT CONSTANT EXTENSION RATE OF 4.2x10-6 SEC -I AND IN 288 C l HIGH PURITY WATER CONTAINING AIR-SATURATED DISSOLVED OXYGEN (4 ppm) Surface preparation Type of 6 max.III, U.E.(2) T.E.I3) . SCC area ratio I4) Mr of before SCC test reagent kg/mn2 % % % specimen As weld (no oxide film and 40.7 no chemical cleaning) 14.9 18.2 58 4 i t No oxide film : A 40.3 15.1 18.3 60 4 24 hrs. C.C.(5) B 41.4 14.4 18.9' 55 4 331 hrs. corrosion % A 40.6 14.5 18.5 61 4 24 hrs. C.C. B 39.6 15.8 18.9 65 4 953 hrs. corrosion % A 36.0 10.0 14.5 65 4 72 hrs. C.C. - B 38.9 12.4 16.5 55 4 A 38.4 11.3 15.5 Decontamination cycle I6I 64 3 2 B 40.3 14.9 19.1 57 2 Pre-cracked specimen (7) A 37.6 10.8 15.4 60 4
---> 2 4 hr s . C . C .
B 40.1 13.6 17.6 60 4 l l (1)6m'ax. : Maximum stress value, (2) U.E. : Uniform elongation. (3) T.E. : Total elongation. (4) SCC area ratio: SCC area ratio on the fractured surface, (5) C.C. : Chemical cleaning, " (6) Decontamination cycle : Three times procedure of"24 hrs. chemical cleaning after approx. 330 hrs. corrosion *,' (7) Pre-cracked *,pecimen : Prer.rnck introduced by CTRT SCC test in ?RR"C nure water
Table 3 RESULTS OF CVCLIC TENSION SCC TEST, CONDUCTED UNDER LOAD-CONTROLLED CONDITIOM IN 288 C HIGH PURITY WATER CONTAINING AIR-SATURATED DISSOLVED OXYGEN (4 ppm) Surface preparation before Type of Number of cycle SCC area ratioIII. Number of SCC test reagent to failure 1 specimen As weld (no oxide film and no 19 64 4 chemical cleaning) No oxide film A 25 58 4 24 hrs. chemical cleaning 31 B 61 4 331 hrs. corrosion % A 22 60 4
? 24 hrs. chemical cleaning 8 23 58 4 953 hrs corrosion
- A 18 59 2 72 hrs. chemical cleanirig B 26 64 . 2 l Decontamination cycle 8 27 62 2 (1).SCCarearatio: SCC area ratio on the fractured surface .
(2) Decontamination cycle : Three times procedure of 24 hrs. chemical cleaning after approx. 330 hrs. corrosion in 288"C pure water"
Table 4 RESULTS OF CONSTANT LOAD SCC TEST IN 288 C PURE WATER (18 ppm dissolved oxygen) Surface preparation before Type of A plied stress, Time to failure, SCC area Nuever of SCC test reagent dy(1) hr. ratio (2).1 specimen As weld (no oxide film and no 1.50 No failure to 2100 2 chemical cleaning) 1.75 40 59 2 1.50 No failure to 2100 2 231 hrs. corrosion -- A 1.75 38 62 2 24 hrs. chemical cleaning , 1.50 No failure to 2100 2 1.75 43 53 2
~
1.50 No failure to 1900 2 k 953 hrs. corrosion % A 72 hrs. chemical cleaning
- 1.50 No failure to 1900 2 B
4 1.75 41 52 2 1.50 No failure to 1900 -- 2 A 1.75 . 65 61 2 Decontamination cycle (3) ' 1.50 No failure to 1900 2
, B -
1.75 57 58 2 (1)d : 0.2% offset stress in 288*C pure water (2) SbC area ratio : SCC area ratio on the fractured surface (3) Decontamination cycle : Three times procedure of"24 hrs. chemical cleaning after approx. 330 hrs. corrosion in 288'C pure water"
Table 5 TIME TO FAILURE IN PIPE TESTING Type of Type of TTF in hour Decon. Reagent Experiment 1st Failure 2nd Failure 3rd Failure (1) 134 135 162 (1) 133 205 206 A (2)* 158 159 172 (2)* 182 193 222
~
(1) 203 208 237 (1) 173 178 181 B (2)* 128 148 150 (2)* 234 241 268 Experiment (1); DECON CYCLE + PIPE TEST Experiment (2); PIPE TEST (10 cycles + DECON EYCLE + PIPE TEST ) TTF (hrs) in Pipe Testing Type of Type Median TTF Standard Mean TTF Decon. Reagent Experiment in hour Deviation in hour (1) 245 1.5 268 A (2)* 240 1.4 252 (1) 225 1.3 26k 0 (2)* 320 1.8 380 NON DECON CYCLE 225 1.5 244
- includes pre-cracking time 7-15
Table 6 CHEMICAL COMPOSITION OF MATERIALS (wt. %) Material C Si Mn P S Ni Cr Fe Other El. Type 304 S.S. 0.06 0.59 1.62 0.032 0.010 E.68 18.17 bal. Inconel 600 .0.044 0.28 0.32 -- 0.010 74.28 15.32 8.50 Inconel X-750 0.04 0.09 0.05 0.004 0.005 73.60 15.57 6.62 hb+Ta 0.85 A1 0.49 Table 7 HEAT TREATMENT OF MATERIALS FOR CBB TEST Material Heat Treatment Condition Type 304 S.S. a) 650 C x 3h AC 4 b) 650 C x 10h AC Inconel 600 a) 650 C x 3h AC b) 620 C x 24h AC Inconel X-750 a) 985 C x 1h AC
+705 C x 20h AC A.C. = Air Cooling i
7-16
Table 8 SCC SUSCEPTIBILITY IN 500 HR CBB TEST IN HIGH TEMPERATURE WATER WITH RESIDUAL CHEMICAL REAG2NT Pure Reagent Reagent Material Water A B i Type 304 SS (a) 10/10 6/24 7/12 Type 304 SS (b) 10/10 3/7 2/7 Inconel 600 (a) 10/10 0/7 0/7 Inconel 600 (b) 8/10 0/7 0/7 Inconel X-750 5/8 - 1/2 Table 9 MAX. CRACK PENETRATION IN CBB TEST IN HIGH TEMPERATURE WATER WITH RESIDUAL CHEMICAL REAGENT (in um) Pure Reagent Reagent Material Water A B Type 304 SS (a) 991 900 760 Type 304 SS (b) 864 580 480 Inconel-600 (a) 1045 0 0 Inconel 600 (b) 324 0 0 Inconel X-750 591 - 84 1 i l 7-17
l l GRAPillTE FIBRE WOO' M*ECIM EN k50m10a21 r ,, , ,
,, , r,, -
~
O a t.m l I I I Wa = 4M l l
~ 8. ;
as llo Figure 1. Assembly of Creviced Bent Beam Test Jigs 7-18
- _- . . . -- - . - . - - . - - . _ - _-- . . _ _ . - . ~
I SECTION VIII ULTRASONIC EVALUATION
8.1 INTRODUCTION AND BACKGROUND
In March,1982, during a hydrostatic test at NMP-1 evidence of leakage was observed on two 28-inch diameter recirculation system safe-ends. Subsequently , ultrasonic (U.T.) examination was conducted and confirmed the presence of i intergranular stress corrosion cracking in the safe-ends. In addition a
; recirculation pump discharge nozzle to elbow weld had U.T. indications which were confirmed, by destructive examination, to be IGSCC. These events, and I those which occurred later are listed in chronological order:
- 8.1.1 Sequence of Events 3/23/82 - Visual discovery of leaking safe-ends - axial orientation.
3/26 to 3/31/82 - Informational examinations performed on leaking and non-leaking safe-ends. Indications reported. Safe-end activity
, stopped.*
4/1/82 - Commence informational examinations at pump end of recircu-i- lation system. Indications reported. 4-
'4/20/82 - Transfer measurements of cal blocks and riser elbow per-formed. 10 dB variance noted. (Resulted in 10 dB increase in examination sensitivity and changed recording criteria t for subsequent examinations.)
- 5/6/82 - Commence ultrasonic examination of balance of recirculation piping welds following confirmation of IGSCC existing at riser elbow.*
6/82 to 12/82 - Ultrasonic, visual and liquid penetrant verification of. cracking in removed material. CExaminations performed and results are presented in Section 8.1.2. . 1 8-1 .
,~. ..m_ ,. .,
(Appendix A of this section contains excerpts of the procedures used for the IGSCC oriented examinations and the examination of other austenitic S/S piping.) The word " informational" has been applied to examinations performed during this period for several reasons. In the case of the safe-ends, the examina-tions, while performed in accordance with procedure, did not generate care-fully plotted, detailed results. By direction and agreement, the exams were to confirm detectability of the discovered cracks and make a survey for other indications. Greater accumulation of man-rem for detailed results was con-sidered unnecessary given known failure and necessary replacement. Further, the informational quick-look approach continued into the next examinations on the recirculation system. Strict adherence to specified procedural techniques was not maintaired once additional indications had been detected. Different evaluation techniques were used as soon as the normal approach had provided evidence of cracking indications. , Transfer measurements from calibration blocks to components were made as the result of using alternative techniques for evaluation. A depth calibration approach employing small diameter side drilled holes rather than notches for calibration was desirable for aiding evaluation. It was determined that the depth calibration sensitivity was about 10 dB greater than that resulting from the notches. Subsequently, a transfer comparison was made which showed that for the recirculation pump elbow, there was a difference of 10 dB between the calibration blocks and the component, the component being the more attenuative. Primarily, this established direct application of the depth technique to examination of component and secondarily resulted in an increase of 10 dB being applied to the standard notch calibration sensitivity for the balance of the examinations performed on the recirculation system. Part of the transfer measurement activity was devoted to making the same com-parisons between the available flat and curved calibration blocks of the same material. There was approximately 2 dB difference noted. It is important to realize that the 10 dB factor did not result from a difference 1 8-2 ! 4
)
between flat and curved calibration blocks, but from attenuation differences between the calibration blocks and pump discharge elbow. A verification o'f the attenuation difference between the flat and curved calibration blocks was requested as a result of comments made by General Electric. The recheck confirmed the initial differential of 2 dB between calibration blocks. 8.1.2 Principle Aspects of the 1982 U._T. Examination at NMP_-1 i For the U.T. examination used to detect IGSCC in the NMP recirculation system, the procedure was modified from that used during previous in-service inspec-tions (ISI). Miniature 1/2 inch and 1/4 inch search units were used and scanning was conducted at 20 dB abave calibration gain. The effect of the l miniature search units upon scanning was that it allowed the ultrasound to reach the weld heat affected zone (HAZ) of welds with wide weld crowns. Pre-viously scanning was conducted at 6 dB above calibration gain and 1/2 by 1 inch search units were used. The latter restricted ability to examine the HAZ effectively. It was also discovered in the course of U.T. examinations conducted in 1982, that the attenuation of the piping, safe-end and elbow components measured from 6 to 10 dB higher than that measured on the flat 1.5 inch thick stainless steel calibration standard. This difference in attenuation resulted in insufficient gain being used during scanning to adequately detect and record low amplitude signals. The principal' aspects of the U.T. procedure employed in 1982 at NMP-1 required the following: (a) Calibration for examination using 10% of thickness notches in a flat 1.5 inch thick calibration standard. l (b) Correction for attenuation to accommodate for the higher attenuation ! measured on the safe-end, piping and elbow components. J 8-3
l l U (c) Use of miniature 1/4 inch 2.25 MHZ 45 angle beam and 1.5 MHZ 45 dual element angle beam search units. (d) Recording of all indications regardless of amplitude if considered to be caused by IGSCC. (e) Plotting of indications using full scale drawings of the weld configuration. . (f) Examination for axial and circumferential indications.- (g) Training of U.T. personnel on piping samples with actual IGSCC. Subsequent examination of recirculation piping system welds conducted at Monticello, Hatch, and TVA-2 (Browns Ferry 2) and others were performed to procedures similar to that employed at NMP-1 in 1982. Deviations if any, consisted of using strip chart recording, computers for data plotting, and calibration for examination on curved calibration standards of the same diameter and thickness as the component to be examined. The latter deviation would provide a minor increase in scanning gain, the other two would enable more exact plotting. , With respect to detection and sizing of axially. oriented 1SSCC in the weld HAZ, their detection is difficult. The majority of. the piping welds at operating plants are in the as-welded condition or are flat topped. As a result minipulation of the search units so as to transmit ultrasound per-pendicular to these axial cracks is not normally fully achieved. Instead examination is conducted from the nearest flat surface with the search unit ~ angled toward the anticipated area of cracking. 'Any IGSCC' detected would therefore have reduced amplitude and be difficult' to si;:e for depth. These difficulties were an impedance to the identification'of axially oriented cracks at NMP-1. Whenobserved,theindicationsof_axiallyorientedcradking were generally below the recording limit. The sanie situation would be true for any plant with extended weld crowns or unground welds. [ oO W 4 em N
.g
^
8-4 : N %' a
In order to improve the detection of axial IGSCC in weld HAZ's grinding of the piping system weld crowns is required. The degree of grinding should provide a smooth cylindrical surface without any obstructions to prohibit search unit movement. Of plants examined to date, all except plant Hatch, had unground welds. 8.2 RECIRCULATION SYSTEM ISI EXAMINATION RESULTS i For examination activities occurring prior to the 1979 refueling outage at NMP-1, no specific attention had been given to the potential for IGSCC : occurrences in the recirculation system. Neither the selection process for welds nor the examination techniques utilized reflected the requirements of NUREG 0313. Any welds that had been examined had been done in accordance with a procedure based on the requirements of Article 5 of Section V of the ASME Code. For the 1979 outage, an augmented program had been developed in accordance - wi n the provision of NUREG 0313, Rev. O. Although initial drafts of this program included the entire recirculation system, it was finally decided l that only the recirculation safe-ends techniques fell into the service sensitive category. The balance of the recirculation system piping welds remained in the regular ISI~ program. Also in accordance with NUREG 0313, the ultrasonic examination procedure for austenitic stainless steel piping welds was modified and augmented to address detection of IGSCC; specific transducers were called out, scan speed modified,
~
sensitivity increased, etc. However, these improvements only applied to these welds or items identified in the augmented program. In this case, the improved techniqu.es were Tapplied only to the safe-ends. For the 1981 outage year a new examination 4 procedure addressing only IGSCC 1 detection had been developed for use on'those items identified in the augmented program. While this effort further enhanced IGSCC detection, no changes were
~
made to the selection of items to be examined, and only the safe-ends 4 8-5
themselves were to be examined with the improved methods. The balance of the recirculation system piping remained part of the standard ISI program. This means that prior to 1982, less than 10% of the recirculation piping circum-ferential welds had been examined since the NMP-1 ISI plan was implemented and none of those examinations reflected IGSCC detection considerations. Following the safe-end leakage in 1982, indications were readily detectable in the safe-ends. It was at this point that the remainder o. the recircula-tionpipingwassubje,btedtoexaminationutilizing.IGSCCtechniquesforthe first time. Indications were detected in almost every weld examined once IGSCC techniques were applied. Table 1 provides graphic and tabular i...ormation about in-situ examinations performed op the recirculation piping in 1982. (Refer to Figures 1 through 5 of Section 2 for identification of weld numbers). It will be noted that the bulk of examinations were performed on the pump suction side of the loops. This was a result of access conditions. Other pertinent information is included where appropriato, eg., welds with particularly large cracks, welds examined in l981, etc. g Except for 1982, the safe-ends themselves have been examined'as part of the augmented program and are not part of this listing. 8.3 , ISI INSPECTION COMPARISONS - 1979-82 During;the primary system hydrotest in March 1982. leakage was visually detecte'd.at two of the ten furnace-sensitized, recirculation system safe-ends. Subsequent to the discovery of the leakage, U.T. examination of the two affected safe-ends and one other safe-end confirrred the presence of indica-tions of intermittent cracking around the inner surface. After the inspection of the safe-en'ds additional examinations,were carried out in selected areas of the balance of the recirculation piping system. These first added examinations revealed indications in the heat-affected zones of recirculation; pump discharge welds at the inner surface. Partial
~
v t
, i 0
8-6 Y l .. l
disassembly of some of the pumps allowed access to the inner surface of the subject welds with dye penetrant examination confirming the indications as cracks. The U.T. examinations were then extended to other welds in the five loops of the recirculation system. This resulted in the identification of a large number of indications considered to be cracks. In addressing the questions as to why certain defects in the total recircula-tion pipe system at,Nine Mile point were not detected in the 1981 inspection the subject must be divided into two categories. These categories are the (a) sensitized scfe ends and (b) the balance of the recirculation system. Different procedures' were applied to these two subdivisions of the piping system., The safe ends were part of the augmented inspection program according to NUREG 0313, while the balance of the system was not. 8.3.1 Safe Ends Tables 2-7 derived from the ISI records of 1979-82 compare the inspection parameters of the safe ends in loops 11-15. In reviewing these and other supporting records the following factors are observed:
- 1. In 1982 more gain was added to the calibration sensitivity for scanning than in 1981 (typically 10dB versus 6dB).
- 2. Less time was devoted to a typical inspection in 1981 than in 1982.
l
- 3. The 1982 inspection took place after the presence of IGSCC had been confirmed by leakage, creating a psychology of inspection contributing to more careful examination and more willingness to call cracks.
Elaborating on the time consideration, the scanning and recording times of 1981 appear to be too short for thorough evaluation. Even without any recordable indications, they are too short for sufficient scanning to find 8-7
IGSCC. For example note the case in Table 6 of discharge safe ends #11 and 12 in 1981. About 1.5 hours were spent on the examination from one side of two circumferential joints (nozzle to safe end and safe end to pipe) and one l whole safe end body. Rough estimates indicate 30 minutes for the safe end to pipe weld (safe end side only). This means one axial scan and two cir-cumferential scans (CW and CCW) of a 28" diameter pipe took place in 30 minutes. Experience shows that rapid rates of manual scans tend to miss indi-cations in the areas where IGSCC occurs. It is difficult to say how slow the scan should be. This must be learned by UT personnel through hands-on train-ing on octual IGSCC specimens. In summary, the following conclusions are reached relative to a comparison of the 1981 and 1982 NMP-1 safe-end inspection.
- l. A higher sensitivity was used for the scanning in 1982.
- 2. The time devoted to some of the safe end examinations in 1981 appear to be too short for optimum detection of IGSCC.
- 3. All UT inspection in 1982 was done after the presence of IGSCC had been confirmed by leakage, probably resulting in a somewhat more careful inspection.
i 8.3.2 Balance of Recirculation Pipe System Table 8 was derived from the ISI records of 1981 and 1982 for the common joints inspected in both years. In reviewing these and other supporting records the following factors are observed:
- 1. Prior to 1982 sampling (examination frequency) and techniques were based on Code requirements with no IGSCC consideration.
- a. Only two joints were inspected during the 1981 ISI, namely i 32-FW-10-W and 32-FW-36-W. Comparison of ISI results in 1981 and 1982 is hence limited to these two welds.
8-8
- b. The 0.5" x 1.0" 2.25 MHZ transducer used in 1981 will have a lower sensitivity to small defects (due to its large size) than the transducers used in 1982. It is expected that these indications would have been on the order of 20% DAC or less in 1981. The 1981 procedure required a 50% DAC reporting level .
- c. In 1982 more gain was added to the calibration sensitivity for scanning than 1981 (both 10 dB and 20 dB versus 6 dB).
- d. Indications were found in both joints in 1982 with amplitudes less than 50% DAC (10% notch) using 1/2" diameter 1.5 MHz transducers,
- e. More time was typically devoted to scanning and recording of data in 1982 than in 1981.
- 2. PT of 32-FW-36-W revealed a substantial axial crack. This crack is not detectable ultrasonically due to iriterference from unground weld crown.
Relative to the time devoted to inspection in 1981 versus 1982 note from Table 8 that the examination from one side of joint P32-FW-10-W plus the j examination of two other circumferential joints from one side and 48 inches of axial weld from both sides took one hour and 49 minutes in 1981. It is estimated that the subject examinetion of joint P32-FW-10-W took about 30 minutes. Note that 3 hours and 20 minutes were devoted to the same examination in 1982. In summary the following conclusions are considered to relate to the differ-ence between the 1981 and 1982 inspection results on the balance of the recirculation system: 8-9
, -e
i
- 1. The procedure used in 1981 (a procedure acceptable to Section XI Appendix III) is ineffective for detection of IGSCC because of the 50% DAC reporting level.
- 2. Unground crowns may interfere (often do) with detection of axial cracks.
- 3. The time spent on scanning and recording is considerably lower for 1981 than 1982, and may be too short for optimum inspection for IGSCC.
- 4. IGSCC experience of inspection personnel was higher in 1982 than 1981 (availability of IGSCC samples and participation in EPRI NDE Center workshops).
- 5. The transducers used in 1982 resulted in effectively a more sensitive examination compared to 1981.
- 6. A higher sensitivity was used for scanning in 1982.
- 7. The same psychology of inspection after confirmation of IGSCC was present in 1982 as in the case of the safe end inspection.
8.4 EVALUATION OF ULTRASONIC EXAMINATION PERFORMANCE FOLLOWING PIPE REMOVAL In addition to examinations performed in-situ on the recirculation system piping, a number of welds were examined subsequent to their removal. Fig-ure 1 is a summary of examinations performed, noting procedure used and whether the in-situ examination occurred before or after initial system decontamination. Due to a number of practical limitations, a complete in-situ and post-removal examination comparison was precluded. Six welds were examined both in-situ and after removal. A limited comparison of examination data from before and 8-10
~
and after in-situ chemical decontamination was also made. The following sections summarize the results of these analyses. 8.4.1 Correlation of UT and PT Data To date, U.T. examinations on the specimens have been performed without benefit of ID PT information. This was generally followed from efforts to proximate the in-situ examination condition. Comparison of several ID PT examinations with the U.T. data suggest a relatively good correlation overall with a number of indications detected by only one of the examination methods. Where there is a lack of correlation, several possible examination conditions may be responsible. Where a crack is identified by PT but not identified by U.T., two conditions are most likely:
- 1. The surface geometry precludes search unit positioning favorable to detection. This is likely to be the most frequent limitation.
- 2. The grain boundaries in the area of IGSCC may have deteriorated enough that local attenuation precludes U.T. response above the detection threshold.
Where an indication is identified by U.T. but not PT. the indication may result from reflection from significantly degraded grain boundaries which have not yet been opened by corrosion. Another possibility exists. The cracks may have been tight and filled with oxides, presenting the entrance of l dye penetrant material into the crack. I 8.4.2 Search Unit and Procedure Correlation
- In order to evaluate the relative detection capability of the various U.T. l techniques applied to piping, a combination of available search units were tested with three examination procedures on a flawed specimen (Figure 2).
4 8-11
Indications were found with all three procedures using the 45 ,1.5 MHz dual 3/8 x 3/4 inch search unit, and the 60 ,1.5 MHz 1/2 inch round search unit. Other search unit / procedure combinations were less sensitive. All but the 45 , 2.25 MHz 1/2 x 1 inch search unit detected the indications using the IGSCC oriented procedure (80A2818) which was used for 1982 Nine Mile Point recirculation system piping examinations. As illustrated in Figure 3, indications from these tests were relatively repeatable. The high sensitivity of the 60 ,1.5 MHz .5 inch round or 45 , 1.5 MHz 3/8 x 3/4 inch dual search units are considered an acceptably reliable detector of IGSCC when used with the 1GSCC-oriented procedure (80A2818) for piping similar to the Nine Mile Point recirculation system piping. 8.4.3 Effects of Chemical Decontamination on Detection e Field Comparison Eleven of the thirty-six welds examined in-situ were examined prior to system chemical contamination. Of these, four were identifNd to have 360 con-j tinuous intermittent indications, with six varying from few discrete indica-tion to partial areas of continuous intermittent indication and one with no recordable indications. Of twenty-six welds examinations in-situ after chemical decontamination, four had from few discrete indications to partial areas of continuous intermittent indications. These include three pipe-to-recirculation suction safe-end welds which were not directly subjected to decontamination, and will therefore be excluded from the tabulation. The fourth was limited to a one sided examina-tion by accessibility. One weld was examined in-situ before and after chemical decontamir.ation and had 360 intermittent indications in both examinations. All other in-situ, post-decontamination weld examinations found 360 intermit-tent indications. 8-12
Before Decon Of 11 welds examined: U 4 (36 percent of welds) had 360 intermittent indications 6 (54 percent of welds) had "few" discrete indications 1 (s10 percent of welds) had no recordable indications After Decon Of 22 1/2* welds examined after exposure to chemical decontamination: 22 (98 percent of welds) had 360 intermittent indications 1/2*(2 percent) had "few" discrete indications 0 had no recordable indications. The effect of chemical decontamination on detection by UT is tabulated below The results suggest that chemical decontamination may have increased the detectibility of the cracks by ultrasonic examination. Because of this potential enhanced U.T. visibility associated with decontami-nation, the U.T. procedures and results were examined more closely. Evaluation of the U.T. procedures employed revealed that they were identical for both the pre and post decontamination examinations. The on,1y known difference was a somewhat longer examination time for the decontaminated weld. However, this difference is not considered significant. In comparing the reported U.T. results before and after decontamination, it is important to establish that there are no significant differences, i.e, that both categories of welds are from the same population and have similar actual crack patterns. In comparing welds, it is found that except for two discharge safe-end = to pipe spool welds, the weld population examined by U.T. before decontamination consisted entirely of elbow welds associated with the pumps in each loop. Since there is some uncertainty associated with whether-CNote: 1/2 weld represents a one sided examination by accessibility. 8-13 v v * -
the two safe-end welds actually saw decontamination solution, the comparison was made excluding these two welds. The welds examined after decontamination I were more randomly distributed. The actual welds evaluated are listed in Table 9 along with the U.T. results (i.e. , whether termed continuous [360 intermittent] or discrete). Also listed is the stress rule index value for each weld as well as the post-decontamination P.T. indication length for each heat affected zone. Post-decontamination P.T. gives the best available measure of total indication length and provides a basis for comparison between pre- and post-decontamina-tion U.T. results. Examination and analysis of the results presented in Table 9 leads to several interesting conclusions: The weld HAZ indication patterns and stress levels appear to represent a common population for both pre- and post-decontamination welds. This is based on; a ) The average P.T. indication length (13.6% of circum-ference) is the same for pre- and post-decontamination. b) The cumu-lative distributions of HAZ P.T. indication lengths (Figure 4) is very similar for pre- and post-decontamination welds. c) The average stress rule index values for the pre- and post-decontamination welds are also very similar, 1.26 and 1.25 respectively. Based on the above, it can be concluded that the pre- and post-decontamination welds are from a common population with similar cracking patterns and there-fore the apparent reported differences in U.T. response on welds examined before versus after decontamination appear to be related to some aspect of the decontamination process. Perhaps the removal of oxide from the crack mouth by decontamination enhances U.T. visibility leading to the reporting of longer indication lengths. Additional evidence supporting enhanced U.T. visibility after decontamination is available from further analysis of the results in Table 9.- It is assumed that the true indication length and circumferential distribution is best represented by the post-decontamination P.T. pattern. In Table 9, the available P.T. patterns for each weld HAZ are classified as 8-14
continuous (i.e., 360 intermittent indications), discrete, or none (no apparentindications). When only those HAZ's with continuous P.T. patterns are c.ompared with the respective HAZ U.T. patterns, it is found that if U.T. was performed before decontamination only 33% of the continuous P.T. patterns also exhibited continuous U.T. patterns. However, if U.T. performed after decontamination, 89% of the continuous U.T. patterns also exhibited continuous P.T. patterns, indicating again that decontamination apparently significantly enhances U.T. visibility. In some other cases of continuous U.T. patterns, listed in Table 9, the corresponding P.T. patterns were non-continuous (discrete or non-existent). In these cases, it is likely that the U.T.
, " indications" were associated with geometric reflectors rather than cracks.
Since crack depth sizing was not attempted during the bulk of the U.T. exams, the available data are not adequate to determine whether decontamination also enhances crack depth determinations. 8.4.4 In-Situ vs. Laboratory U.T. Examination A total of six welds were subjected to similar U.T. examinations both in-situ and after removal from the system. Due to repair and sample decontamination schedules, the balance of U.T. examinations were performed uniquely either before or after removal . All samples had been exposed to chemical decontamination. Of the six data comparisons:
- four have numerous discrete indications in the laboratory compared to continuous intermittent indications in-situ. Of these four one weld has 100% correlation between discrete indi-cations in-situ and laboratory.
- one has fewer discrete indications in laboratory than in-situ examination (about half of the indications in the examination correlate approximately).
9 8-15 l
- one has no laboratory indications compared to continuous intermittent indications in-situ (possible weld identification problem). l I
A controlled test would be required to more accurately determine the effects of cutting out the piping on detectability of existing IGSCC cracks. It does not appear, based on the limited sample, that there is any substantial difference in IGSCC crack detectability between in-situ post chemical 1 decontamination examinations and post removal examinations using the same U.T. examination procedure. In the reduced radiation exposure environment of the laboratory, sizing and discrimination of small cracks is improved as expected. 8.4.5 Crack Depth Information Due tc the small through-wall dimension of typical IGSCC cracking relative to see ch unit beam width, it is probable that typical sizing parameters ("W" measurements and metal path change) are characterizations of the search unit beam spread, not necessarily the crack depth. (Figure 5.) This is substantiated by review of the basic trigonomatric parameters. These demonstrate that neither the plotted ID reflection of the crack tip (Figure 6) nor 6 dB down beam angle from the crack corner (Figure 5) satisfy the empirical metal path data for the 6 dB down "W 2
" p sition.
It is most likely that the "W "2 (6 dB down) position represents the point at which the crack corner reflects the spread beam from the forward search unit edge. Thus, the metal path may result from the angle of beam divergence rather than from an extended " vee" path (Figure 8). It is concluded that the distance from maximum signal "Wm" to the 6 dB down point "W " suggests an upper limit on IGSCC crack depth but that the actual 2 crack depth is considerably smaller than this limit. Additional studies have been initiated to improve sizing techniques; however, these techniques have not been applied to the Nine Mile Point welds. 8-16
TABLE 1 ISI EXAMINATIONS PERFORMED ON NMP-1 RECIRCULATION PIPING WELDS IN 1982 RECIRCULATION LOOP 11 WELDS EXAMINED DESCRIPTION '82 EXAM DATE RESULTS COMMENTS FW-2 Pipe. Valve 5/29 Circ. indications IGSCC Techniques FW-3 Valve / Pipe 5/29 Circ. indications IGSCC Techniques SW-3 Pipe / Elbow 6/1 Circ. indications IGSCC Techniques FW-4 Elbow / Elbow 5/4 Circ. indications IGSCC Techniques FW-26 Pump / Elbow 5/6 Circ. indications IGSCC Techniques SW-16 Pipe / Pipe 8/6 Circ. indications IGSCC Techniques NOTE: Visual detection of leakage on this loop - safe end on pump discharge leg. RECIRCULATION LOOP 12 WELDS EXAMINED DESCRIPTION '82 EXAM DATE RESULTS COMMENTS FW-6 Pipe / Valve 6/2 Circ indications IGSCC Techniques FW-7 Valve / Pipe 5/31 Circ. indications IGSCC Techniques SW-5 Pipe / Elbow 5/31 Circ. indications IGSCC Techniques FW-8 Elbow / Elbow 5/8 Circ. indications IGSCC Techniques FW-3) Pump / Elbow 5/6 & 6/12 Circ. indications IGSCC Techniques SW-20 Pipe / Elbow 8/6 Circ. indications IGSCC Techniques RECIRCULATION LOOP 13
- WELDS EXAMINED DESCRIPTION '82 EXAM DATE RESULTS COMMENTS FW-32 Pipe / Safe End 3/31 Circ. indications IGSCC Techniques FW-10 (Note 1) Pipe / Valve 6/4 Circ. indications IGSCC Techniques FW-11 Valve / Pipe 6/4 Circ. indications IGSCC Techniques SW-8 Pipe / Elbow 6/4 Circ. indications IGSCC Techniques FW-12 Elbow / Elbow 5/8 Circ. indications IGSCC Techniques FW-36 (Note 1) Pump / Elbow 4/30 Circ. indications First evidence of cracking detected here using non- l Note 1
- Weld nos. FW-10 and FW-36 were examined IGSCC procedure in 1981 - no indications reported. 4/2/82, at 20% DAC. 1 l
Note 2: Weld no. SW-6 contained .5" deep crack. Data obtained after removal. 8-17
Table 1 (Continued)
}
WELDS EXAMINED DESCRIPTION '82 EXAM DATE RESULTS COMMENTS FW-14 Pipe / Valve 6/3 Circ. indications IGSCC Techniques FW-15 Valve / Pipe 6/3 Circ. indications IGSCC Techniques FW-11 Pipe / Elbow 6/3 Circ. indications IGSCC Techniques FW-16 Elbow / Elbow 5/8 Circ. indications IGSCC 'iechniques FW-41 Pump / Elbow 5/6 Circ. indications IGSCC Techniques RECIRCULATION LOOP 15 WELDS EXAMINED DESCRIPTION '82 EXAM DATE RESULTS COMMENTS SW-13 Pipe / Tee 5/26 Circ. indications IGSCC Techniques FW-18 Tee / Pipe 6/2 Circ. indications IGSCC Techniques FW-19 Pipe / Valve 6/2 Circ. indications IGSCC Techniques FW-20 Valve / Pipe 6/2 Circ. indications IGSCC Techniques FW-14 Pipe / Elbow 6/4 Circ. indications IGSCC Techniques FW-21 Elbow / Elbow 6/3 Circ. indications IGSCC Techniques FW-46 Pump / Elbow 4/2 Circ. indications First UT evidence (75% DAC) of cracking detected here using IGSCC procedure. (75%DAC).PTand film metallurgical verification. NOTE 1: Visual detection of leakage this loop - Safe end on pump suction leg. NOTE 2: Weld No. SW-12 contained .55" deep crack. Data obtained after removal. 8-18
TABLE 2 INDICATIONS SUCTION DISCHARGE 11 12 13 14 15 11 12 13 14 15 1979 NO N0 GE0 GE0 ? NO N0 NO N0 N0
- 1980 -- --
GE0 -- -- -- -- -- -- -- i ~ 1981 GE0 GE0 GE0 GE0 N0 N0 N0 NO N0 NO 1982 -- -- -- -- IGSCC I IGSCC -- IGSCC -- -- DP/ LEAK *2 *3 *3 *3 LEAK LEAK *3 *3 *3 *3 i NOTES: i
- 1. Leaking crack investigated by UT, no examination performed.
I
- 2. No cracking found in DP examinations by NDE Center at Battelle Columbus Laboratories (BCL)
- 3. DP investigation at plant or BCL not completed and documented NO - Inspection performed, no reportable indications No inspection GE0 -
Inspection performed, geometry indications reported DP/ LEAK - Determination of flaw by leakage or dye penetrant indication i TABLE 3 j -SEARCH UNIT (NAME/MHz) i ,_ . . . . _. _ . . _ _ _ _ . _ _ . _ _ . . . . _ . . - _ _ . _ _ - - 2 t SUCTION DISCHARGE I 15 12 l 13 l 14 15 l I 11 12 13 14 l 11 _.j . j i ._ _ _ q i i l 5051-10/: ; SUSI-10/' i 1979AER0/1.6lAER0/1.6;AER0/1.6AER0/1.6AER0/1.61 1.5 'AERO/1.6,AER0/1.6: 1.5 DUEL /1.5 , ;1980. -- l -- AER0/1.6, -- -- -- --
.SUSI-10/;5USI-10/!SUSI-10/!SUSI-li/;SUSI-39/.SUSI-39/,5USI-10/{SUSI-10/SUSI-10/ ;
' 1.5 1.5 1.5 1.5 1981lAER0/1.5 1.5 1.5 1.5 l 1.5 j 1.5 i l
iSUSI-10/ SUSI-10/ 1982 -- -- -- -- -- l.5 -- l.5 -- --
..{_. ._ .l.
AER0/N Aerotech transducer of frequency N, MHz SUSI-X/N Search Units Systems, Inc. transducer. Model X, frequency of N. MH2 i 8-19
TABLE 4 SENSITIVITY * (CAL. SENSITIVITY WITH P8F-1.5-1), CAL / SCAN SUCTION DISCHARGE 12 13 14 15 11 12 13 14 15 11
~
1979 43/? 49/? 41/? 38/? 43/? 77/? 45/? 47/? 39/? 62/? 40/46 -- -- 1980 -- -- -- -- 1981 47/53 55/61 55/61 55/61 68/74 40/46 40/46 98/104 102/108 102/108 1982 -- -- -- -- -- 44/54 -- 55/65 -- --
- All except Suction #13,1980 were with Sonic Mk-I. (USL-31 for Suction #13).
TABLE 5 U TEMPERATURE (CAL. BLOCK / COMPONENT F) SUCTION DISCHARGE 11 12 13 14 15 11 12 13 14 15 1979 63/84 64/86 64/81 67/80 63/84 72/96 80/90 65/85 65/86 73/90 1980 -- -- 110/132 -- -- -- 1981 65/86 65/86 65/86 65/86 74/99 75/95 75/95 76/96 65/86 65/85 1982 -- -- -- -- -- 88/100 -- 72/88 -- -- 8-20
TABLE 6 SCANNING AND RECORDING TIME (HR. MIN.) SUCTION DISCHARGE s II 12 13 14 15 11 12 13 14 1 15 1979 2.50 2.05 0.55 3.00 2.50 3.00 3.00 1.30 2.00 2.48 1980 -- -- 1.50 -- -- -- -- -- -- -- 1981 1.25 2.13 2.12 2.13 2.20 1.30 1.30 1.40 1.40 1.40 6.201 1982 -- -- -- -- 1.45 -- 2.15 -- -- NOTE: The listed times are the time between calibration time and final check. This typically includes inspection for scanning of r,ozzle/ safe end weld (safe end side), safe end body and safe end/ pipe weld (safe end side). I I Elapse time includes activities other than examination. Scanning and
' recording time consists of two 1-2 hour periods.
TABLE 7 UT PERSONNEL (LEVEL) SUCTION DISCHARGE 11 12 13 14 15 11 12 13 14 15 1979' II,I III,II II,II III,II II,I II,I II,I II,I II,I II,I + 1980 -- -- III,II -- -- -- -- -- -- -- 1981 II,1 III,III II,I III,III II,I III,III III,III II,I II,II II,I III,II : 1982 -- -- -- -- -- III,III -- III,III -- -- l l 1 l 21 nw
TABLE 8 OVERALL COMPARISON COMMON JOINTS 1981, 1982 l 151 122 l
- P3231Gf P32 3 W P32-Al 10f P32 3 m l
Ineicaese a e s. n Orc asOc ses Oc (15E OAC at +103) l st +10 s l lJT Instnment 51 El 12 5 , 3 E-1 l I I Search thit AE E CH AEREH AE M Di AERMIN AEMTEDi 1/2"d" RECT. 1/2"d" IECT. 1/2" e 1/2" e l 1/T e 2.25 Mtr 2.25 sett 1.5 % 2.25 Mtr 1.5 Mtz - I cal. M ock Pen 1.050 1 PEE 1.0SO-1 PEbl.0501 P 5 1.050 1 I Sensitivity l Cal.(e)/sem 72/78 72B8 42/62 31/41 38/58 I . Ta meret a (T) l Cal. Mk/Cagmrn G/72 GR2 W76 62/80 62/70 l _ Scan & IIncord 9.15/11.04 9.15/11.04 8.40/11.60 9.45/15.15 l 9.00/10.30 11e (W. Min.) 1.495 1.49' 3.208 5.308 1.E' I tJT persrael !! ! II.! III.II !!!.!! l !!.I
- u. m
- 1his total tie includes scaming both joint P32310s and P3233Si from me sie only plus 1 other cirtuurerential sids from one side only and 4 - 1r sectims of longitudinal meld fra both sides.
. 8 line for scaming P32310i fra one side only.
8 llan for scaming P32335d from one side only plus three other circumferential elds frun one side only.
- T1e fw scanning P3233Si from one side only.
8-R2
TABLE 9
SUMMARY
OF COMPARATIVE U.T. AND P.T. RESULTS Welds U.T.'d Before Decon, P.T.'d After Decon Stress Rule Weld Index % Circumference With HAZ Classification P.T. Indication P.T. Indications UT PT FW-4 Elbow 1.3 D D Circ. Ind. 8 FW-26 Elbow 1.2 C D Circ. Ind, FW-22 Pipe 5 1.2 N D Circ. + Axial 5 Safe-End 1.2 D C Circ. + Axial + Leak 10 FW-8 Elbow 1.3 D C Branched Circ. 40 FW-31 Elbow 1.2 C C Branched Circ. 26 FW-12 Elbow 1.3 C C Branched Circ. 44 FW-36 Elbow 1.2 D D Circ. + Axial 1 FW-32 Pipe 1.4 Safe-End D C Circ. + Axial 6 1.4 D C Circ. + Axial 3 FW-41 Elbow 1.2 C D Circ. + Axial 3 FW-21 Elbow 1.3 D D Cire. 5 FW-46 Elbow 1.2 D C Circ. 21 1.26 Avg. 13.6 Avg. Welds U.T.'d_After Decon, P.T.'d After Decon SW-2 Tee 2.1(1.7)* N N No Ind. O Pipe 2.1(1.7)* D N No Ind. 0 FW-2 Pipe 1.2 C - - FW-3 Pipe 1.2 C - - SW-16 Pipe - C N No Ind. O Pipe - C N No Ind. O SW-3 Pipe 1.2 C N No Ind. 0 Elbow 1.2 C D Cire. 2 FW-5 Safe-End 1.2 D N No Ind. 0 Elbow 1.2 D N No Ind. 0 FW-6 Pipe 1.2 C - - FW-7 Pipe 1.2 C - - SW-17 Elbow 1.2 D N No Ind. O Pipe 1.2 N N No Ind. O SW-5 Pipe 1.2 C D Axials 1 Elbow 1.2 N D Circ. 2 FW-31 Elbow - C C Branched Circ. 26 SW-20 Elbow 1.2 C D Circ. 2 Pipe 1.2 C C Circ. 13 SW-19** Pipe 1.2 C C Branched Circ. 55-60 Pipe 1.2 C C Branched Circ. 25-30
- SRI = 1.7 if treated as branch connection; 2.1 if treated as Tee
- U.T. performed after pipe removal.
C = 360 Intermittant Indications , N = No Indications D = Discrete Indications 8-23
TAdLE 9 (Continued)
SUMMARY
OF COMPARATIVE U.T. AND P.T. RESULTS Welds Y.T. 'd, P.T. 'd After Decon (Cont'd) Stress Rule Post Decon % Circumference With Weld HAZ Index Classification P.T. Indication P.T. Indication UT PT FW-9 Safe-End 1.2 D C Circ. + Axial >50% Elbow 1.2 0 N No Ind. O SW-7 Pipe 1.2 D - - - FW-10 Pipe 1.2 C' - - - FW-11 Pipe 1.2 C - - - SW-8 Pipe 1.2 C C Circ. + Axial 42 Elbow 1.2 C C Circ. + Axial 16 FW-14 Pipe 1.2 C - - - FW-15 Pipe 1.2 C - - - SW-11 Pipe 1.2 C C Circ. + Axial 14 Elbow 1.2 C C Circ. + Axial 15 FW-16 Elbow 1.3 C C Branched Circ. 18 FW-17 Safe-End 1.2 D D Circ. + Leak 22 Elbow 1.2 N N No Ind. O SW-12 Elbow 1.2 C D Circ. 11 Pipe 1.2 C C Circ. 17 SW-13 Pipe 2.1(1.7)* C - - - Tee 2.1(1.7)* N - - - FW-18 Tee 2.1(1.7)* N - - - Pipe 2.1(1.7)* C - - - FW-19 Pipe 1.2 C - - - FW-20 Pipe 1.2 C - - - SW-14 Pipe 1.2 C C Circ. Ind. 8 Elbow 1.2 C N No Ind. O SW-6** Pipe 1.2 C C Branched Circ. 45-50 Elbow 1.2 C C Branched Circ. 35-40 SW-23** Pipe 1.2 N C Branched Circ. 30-35 Elbow 1.2 C C Branched Circ. 20-25 SW-10** Tee 2.1(1.7)* N - - - Pipe 2.1(1.7)* N - - - SW-26** Elbow 1.2 C C Branched Circ. 20-25 Pipe 1.2 C- C Circ. 5-10 SW-25** Pipe 1.2 C 0 Circ. Ind. 0-5 Pipe 1.2 D Branched Ind. 0-5 1.25 Avg., C 13.6 Avg. Average based only on welds with PT results. *
- SRI = 1.7 if treated as branch connection; 2.1 if treated as Tee.
**U.T. Perfonned after pipe removal .
C = 360 Intermittent Indications N = No Indications D = Discrete Indications 8-24 , i
Loop IN SITU UT Before/After Other NDE (Data Sheet a) No. Weld No. (Procedure * - Data Sheet *) Initial Decon 80A2818 (UT) 80A2819 (PT) 80A4022 (UT) 1I FW-1-W SP-03 11 SW-1 -W - SP-02 11 SW-2-W 2818-19 After EP-UT-10 11 FW-2-W 2818-17 After 11 FW-3-W 2818-17 After 11 FW-4-W 2818-6,7 Before 11 FW-26-W 2818-8 Before 11 SW-16-W 2818-49 After EP-UT-1 11 FW-22-W 2818-1 Before 11 SW-3-W 2818-24 After EP-UT-7 EP-PT-6 EP-DC-3 11 SW-17-W EP-UT-8 EP-PT-7 EP-DC-2 12 FW-5-W 2818-SP-01 After SP-05 12 FW-6-W 2818-30 After to 12 FW-7-W 2818-22 After b
- 12 SW-5-W 2818-23 After EP-UT-4 EP-PT-3 EP-DC-4 12 F W-8-W 2818-10 Before 12 FW-31-W 2818-9/47 Before/After 12 SW-20-W 2818-51 After EP-UT-5 EP-PT-4 EP-DC-5 12 SW-19-W EP-UT-12 12 SW-4-W SP-04 13 SW-6-W SP-04,05,06 SP-06 13 FW-9-W 2818-SP-01 After SP-02,03 SP-07 13 SW-7-W Sp-og 13 FW-10-W 2818-36 After 13 FW-il-W 2818-35 After 13 SW-8-W 2818-37 After EP-UT-09 13 FW-12-W 2818-11 Before 13 FW-36-W 2818-4, SP-07 Before 13 SW-23-W EP-UT-06 EP-PT-05 EP-DC-01 13 SW-22-W SP-09 Figure 1. Summary of Examinations Performed
l l Loop IN SITU UT Before/After Other NDE (Data Sheet *) No. Weld No. (Procedure * - Data Sheet *) Initial Decon 80A2818 (UT) 80A2819 (PT) . 80A4022 (UT) 13 FW-32-W 2818-3 Before 14 FW-14-W 2818-31 After 14 FW-15-W 2818-32 After 14 SW-! !-W 2818-33, SP-08 After EP-UT-11 14 FW-16-W 2818-12 After 14 FW-14-W 2818-9 Before 14 SW-26-W EP-UT-02 EP-PT-01 EP-DC-07 14 SW-25-W EP-UT-03 EP-PT-02 EP-DC-06 15 FW-17-W 2818-SP01 After 15 SW-12-W 2818-46 After 15 SW-13-W 2818-14 After
, 15 FW-18-W 2818-28 After 4 15 FW-19-W 2818-27 After cn 15 FW-20-W 2818-26 After 15 SW-14-W 2818-38,39 After 15 FW-21-W 2818-5,7 Before 15 FW-46-W 2309-4 Before Figure 1. Summary of Examinations Performed (Continued)
NUMBERS ARE MAX % OF DAC AS RECORDED BY EXAMihER AS DETERMINED DURING AXIAL SCAN OF WELD WELD P32-SW4Mf PROCEDURE TRANSDUCER CAL BLOCK SCAN SENS (dB) REMARKS 80A2309 A CURVED 36 100+ 100+ 100 80 REV1 8 P8R 1.050-1 76 100 80 90 100 100 100 100 100 100 100 C 50 c NO INDICATIONS = D 42 E 44 . F II 42 lI 80A2818 A CURVED 36 100 100+ 30 25 100 20 REV1 8 P8R 1.050-1 70 100 80 90 100 100 100 100 40 100 $ 100 C 50 15 15 to 5 D 42 20 20 15 E 44 20 25 20 5 F Jf 42 = NO INDICATIONS C-80A0835 A FLAT 60 100 100+ 15 20 10 100 y REV O/FC-3 B P8F 1.5-1 73 100+ 100 100 100 80 75 100 15 50 10070 ro C 63 15 10 15 5
" D 50 8 15 5 E 51 25 5 50 5 20 F 60 C NO INDICATIONS O 80A2818 E CURVED 62 FC 1,2,3, AND 5 90 90 80 55 I % 00 90 REV1 PB R-1.0501 DATA AT BATTELLE 100 95 100 100 55 90 E 62 SCAN 20 dB > REF 90 RECORD 10 dB > REF DATA AT NMP 1 I I i i i i i I I e I t i I I e i NOTES: 1) TR ANSDUCER INFORMATION 2) ALL SCANNING WAS 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 ANGLE PERFORMED AT INCHES LTR MFG P) SIZE FREQ TYPE BATTELLE ON WELD A AERO 45 3/8 x 3/4 in. 1.5 M DUAL RING P-32-SWB-W r CLOCKWISE LOOKING IN DIRECTION OF FLOW B AERO 60 0.5 in. 1.5 M y EXCEPT EXAM
,C SUSl 45 3/8 x 3/8 in. 1.5 M DUAL IN DICATED IN l
D AERO 45 0.5 in. 2.25 M 7 R EMAR KS ("D ATA E OF PIPE LONG SEAM E AERO 45 0.5 in. 1.5 M 7 AT NMP 1") ! F AERO 45 1/2 x 1 in. 2.25 M y l Figure 2. Summary of Recordings - Using Various Transducers and Nine Mile Point Unit 1 Procedures
ai CLOC X WISE : WELD NO. P32-SWB-W DIRECTION E LDOW SIDE OF FLOW
~ -~
B-100 B 0 B-100 8-90 B-100 0 0 B100 0 ioo 00 O 860 1 l 8100 B-90 550 00 O 100 0 0 .iOO B g A@ B B-20 ' 10F 100g 68100+ g g100 g ~ 909 ACA A CIRC 80 100 6 O C 15 B 80 6 8 756 B 1002.1 90G OE-50 B 50A o 90 Q 95 9 6 O8 55h DE6 B-100 0 Q OC 15 g"A 90Q gB-10 0 A 1000 h 25 0 jod g A-100+ A ISA ADB B-1000 g0 B-100 B-1000 ,$o B.i00 0 A B-5 %y 0 , AC-10 E O A 100 O O O A-100+ B iOO
, ,,3 O A-100+ A 30 O O D-20 895 20 8 00 A-100 0 815A O O E.5 Q E-20 OC.35 A-25O OD-20 b20 C-i Q O
B400 8 m e.2SO q OF q OF PIPE SIDE PIPE SIDE LONG BEAM PlPE SIDE LONG SE AM I I I f 0 i 180 270 O* 90 9 UT IN SITU (NUMBER INDICATES PERCENT OF DAC) - ALSO INDICATIONS 360o INTERMITTENT h UT AFTER REMOVAL (NUMBER INDICATES PERCENT OF DAC O UT AFTER REMOV AL FOR TR ANSDUCER COMPARISON. NUMBER INDICATES PROCEDURE NO. 80A2309. REV 1 ANGLE PERCENT OF DAC. FREO TYPE O UT AFTER REMOVAL FOR TRANSDUCER COMPARISON. MFG 101 SIZE
" LETTER INDICATES 1.6 M DUAL PROCEDURE NO.80A2818. REV 1 TR ANSDUCER TYPE: A AEHO 45 3/8 x 3/4 in.
d UT AFTER REMOVAL FOR TR ANSDUCER COMPARISON. B A US 8 x 3/8 in. DUAL PROCEDURE NO. 80A0835, REV 0/F L-3 y D AERO 45 0.5 in. 2.25 M 45 0.5 in. 1.5 M y E AERO y F AERO 45 1/2 x 1 in. 2.25 M I I Figure 3. Graphical Display of Recorded Indications
i l l 1.0 f 0.9 -
/
38 WELD HAZs UTd AFTER DECON 0.8 - n**** 0.7 - p/ 13 WELD HAZ UTd BEFORE DECON c j 2 0.6 - a / I ' 0.5 -
/
- D P
5 g 0.4 - I 0.3 ( f 02 -
/
0.1 - I I I I I I 0
)
O 10 20 30 40 g g PERCENT OF AN HAZ FOUND CR ACKED BY PT AFTER DECON I Figure 4. Cumulative Probability of Percent of an HAZ Found Cracked by P.T. After Decon 8-29
l l W W MAX W2 MAX W2
- + e 0.2 in. 4 0.2 in.
ll 9 1.6 in. yp 1.6 in. 45 / 1.05 in. 1.05 in. / 1.75 in. MP 1.7 in. ,
/
U U Figure 5. Typical Measured Data Figure 6. Minimum Theoretical Metal Path W MAX W2
+ h 0.2 in.
1 II e 0.2 in. MEASURED MP 1.6 in. MP MEASURED 1.6 in. MP 1.7 in. MP / THEORETICAL MP 1.75 in. 1.7 in. / PLOTTED _J,_f / MP 1.7 in. Fugyre 7. Typical Dimensions Figure 8. Possible Condition , 8-30
APPENDIX A Procedure Excerpts The following paragraphs are taken from the procedures used for (1) IGSCC oriented examinations and (2) other austenitic piping, respectively. This material illustrates the salient differences in examination methods.
- 1. 80A2818, Rev. 1 - Applied to Safe Ends, '81 and '82; applied to Recirculation Piping '82 (following IGSCC confirmation), " Ultrasonic Examination of Stainless Steel Piping for Intergranular Stress Corrosion Cracking (IGSCC)."
5.2.1 The rate of search unit movement shall not exceed two (2) inches per second. The search unit shall be swiveled a 0 full 90 0during axial scanning (45 left and 45 right). 6.1 Examination Contractor's Equipment The following test equipment or its equivalent shall be pro-vided by the Examination Contractor (as a minimum) for examina-tion of welds when specified in this procedure:
- 1. Pulse echo ultrasonic instrument
- 2. Search Unit: dual 1.6 MHz or 1.5 MHz for T > 0.375" 6
- 3. Search Units: 0 ; 1/4" - 1/2" dia. ; 5.011Hz
- 4. Search Unit: 1.6 Mhz, 1/2" dia.; single element for T < 0.375"
- 5. Wedges: 45 shear and longitudinal, 1/2", 60 shear for wall thickness 0.200" and less 8-31
p , 8.2.2 Angle Beam Calibration: 1-1/2 Vee Technique One and one-half vee path calibration'shall be ,the preferred method of calibration for the examinations de' scribed in this procedure and shall be accomplished as follows:
- 1. Obtain maximized signal response from the notenes and
~
mark the signal response positions on the instrument's CRT screen.
- 2. Maximize the signal from the notch producing the highest j response and set its amplitude to 80% FSH._, ;
9.2 Additional Straight Beam Examination - (When indications are detected) . 9.2.1 In order to accurately plot indications, the ID.and OD weld contour must be established. The OD contour shall be -, recorded with the use of a contour gauge.. The IDfcontour'if j f,
..~ . . - ' ,'
established by performing a thickness check of the weld and, . adjacent base material to establislSthe. location,$ depthand
..}
slope of any existing counterbore.~
- .- x 9.2.2 The thickness and contour data is. then plotted on'asfull scale
~ . o,.% . . - -
weld profile. Angle beam recordings are trahsferted to,thi plot to determine the true nature'of all indications,. - N, c* .
-s u[ .
9.3 Angle Beam Examination' . 1 .-
-/ ,
;N 4 9.3.1 The scan sensitivity shall 'be-a pininum of"2b(6 dB) greater '
but no more than 10 dB greater., than'the calibration reference V
< , s , ..
sensitivity level. e
~
'Q - {t
, [,\ Q ,
.s .\ ,
' ~\ y y A. r i
e q .. , Q
. c s e* A 8-32' / hD.)
~
w .
*% - _ _I
9.3.2 The search unit shall be swiveled (45 each way) as it is moved along a rectilinear scan pattern to ensure a minimum of 25% overlap of the transducer width. This is required to detect cracks which may be oriented at odd angles. 9.3.3 The examiner shall pay particular attention to indications which originate adjacent to the weld root and up to 0.5" out from the root. These typicalz locations of IGSCC are of ID origin. 9.3.4 Any such indications which may be considered as counterbore mun be verified by 0 thickness check.
- 10. 'EvaluationCriteria l: ~,, <'10il.1 Indications from the weld crown and from other OD geometric Ih [ m
*U origins may be ignored g The-liquid penetrant examination will cover this area.
(3 \ (s s s ._ All other indications r.:gardless,of amplitude shall be
-, 10.1.2 V
A. . e recorded ... 6 < 1
, /
\
- 2. 80A2309, Rev.1 - Applied to Recirculation Piping, '81, " Ultrasonic ExaminationProcedurisforAusteniticPiping."
4 9' 6.1 * ' Examination Contractor's Equipm'ent ( w
\ v 7 The following test equipment or its equivalent shall be pro- '/.,
(
- Vided by the Examination Contractor (as a minimum) for
.1 /* =
. +
exbmination of welds spec,ified in this pr'ocedure. y . f'.- ., g~ , v s y y , i 1. Pulse echo ultrasonic instruments
%,+- .
% ;7 ,+ ,
o
% s% : 2. Search Units: 1/4" through 1-1/8" dia., 2.25 MHz, 0
%v x s
,+ /
u
.. p
.) '
l \, '
;, ' l \
Qf ! Nm'.i 4y l 4 e O.,* 8-33 l g wt NC
- l. H F &@.. ni !
l}
, l 0 , 3. Search Units: 1/2" through 1-1/8" dia. , 5.0 MHz, 0 }
- 4. Search Units: 2.25 MHz (all sizes) for angle beam wedges.
8.3.2 Sensitivity Calibration; 1-1/2 Full Vee Technique One and one-half vee path calibration shall be the preferred method of calibration for the examinations described in this procedure and shall be accomplished as follows: 1
- 1. Obtain maximized signal responses from the notches and mark tne signal response positions on the instrument's CRT screen.
- 2. Maximize the signal from the notch producing the highest
,; response and set its amplitude to 80% FSH.
( o
- 3. .Without changing sensitivity settings, maximize successive notch indications and mark their peak amplitudes on the
- CRT screen, and on the Calibration Data Sheet.
9.2 Stra'ight and Angle Beam Examination of WRV 9.2.1 All straight and angle beam examinations of the WRV shall be performed at a scanning sensitivity level, a minimum of 2X (6 dB), but no'more than 10 dB greater than the calibrated reference sensitivity level. 9.2.2 The search unit shall be swiveled as it is moved along at a recilinear scan pattern to ensure a minimum of 25% overlap of the transducer width. , 9.2.3 For the locations and the numbers of the welds, refer to the Program Plan. Examinations shall not be considered complete until all recordable indications have been evaluated. 8-34
- 10. Evaluation Criteria 10.1 All indications showing a signal amplitude > 50% of DAC, at the strongest location, shall be plotted full scale. If the
\ indication is nongeometric, the examiner shall continue recording until all necessary data is obtained. .r 8-35
i l APPENDIX B 1 Transducer Usage for Safe End Examination The following tabulation illustrates the various types of transducers used for safe end examinations in 1981. The inconsistency in transducer size could cause some variation in examination results.
Safe End 1981 1982 #11 Discharge (leak) SUSI 39 SUSI 10 #12 Discharge $USI 39 N/A #13 Discharge SUSI 10 SUSI 10 #14 Discharge SUSI 10 N/A #15 Discharge SUSI 10 N/A fil Suction Aerotech 3/8 x 7/8 N/A Dual. 1.5 MHz #12 Suction SUSI 10 N/A #13 Suction SUSI 10 N/A #14 Suction SUSI 10 N/A #15 Suction (leak) SUSI 11 SUSI 10 (incomplete)
SUSI 10 = 3/8" x 3/8" Dual 1.5 MHz SUSI 11 = 1/4" x 1/4" Dual 1.5 MHz SUSI 39 = 1/2" x 1/2" Dual 1.5 MHz SUSI = Search Unit Systems, Inc. 8-36 l
SECTION IX APPARENT CRACK GROWTH RATES - GENERIC IMPLICATIONS The leaking axial through-wall cracks in two furnace sensistized safe ends dis-covered in March 1982, were not detected during ISI nine months earlier (1981). Through-wall crack growth in a 28-inch safe end in nine months or less is completely unexpected based on previous crack growth assessments of furnace sensitized safe ends. Therefore, it was important to establish the rate at which IGSCC would propagate from the true U.T. detectability threshold to through-wall leakage. To accomplish this, it was necessary to:
- 1. Measure actual crack growth rate of a NMP-1 Type 316 furnace sensitized safe end.
- 2. Evaluate the through-wall residual stresses for the NMP-1 safe end geometry in the region of cracktrg.
- 3. Assess for that geometry the U.T. detectability threshold to establish '
the size of crack tnat might have gone undetected during the 1981 ISI. A. MATERIALS DATA BASE i Previous tests have been conducted to measure crack growth rates of furnace sensitized 304 stainless steel under constant loads.1 The objective was to support a methodology used to predict crack growth behavior in large diameter pipes exposed to service environments. Tests were conducted in 0.2 ppm20 and 8 ppm2 0 high purity, high temperature (5500 F) water. The 0.2 ppm low oxygen environment and sensitized material condition is representative of service conditions. Figure 1 presents the data obtained from these tests along with two evaluation curves. The upper bound curve represents a high degree of 9-1
sensitization due to post weld heat treatment. The lower bound curve represents data for moderate degrees of sensitization. The shape of the lower curve was selected to display characteristic stress corrosion cracking behavior. These data results are the basis for predictions used to explain field experience. l l B. CRACK GROWTH RATES OF NINE MILE POINT MATERIAL l A test program was initiated to characterize furnace sensitized safe end material from the Nine fille Point plant that had exhibited field cracking. Constant load crack growth tests were performed using a .9T-WOL (Wedge-Open Loaded) specimen, shown in Figure 2. This specimen was fabricated from NMP 316 stainless steel safe end material, Heat #E-5349. Testing was conducted in the General Electric Small Environmental Fatigue Test (SEFT) machine which is a 100 kip capacity hydraulic test machine. A computerized data acquisition system (Figure 3) was utilized to measure in-situ compliance data. This data is obtained from a Linear Variable Differential Transformer (LVDT) mounted on the specimen face, as shown in Figure 4 The SEFT test vessel was supplied with high pressure / temperature water environment provided by a high flow test loop (Environmental Fatigue Loop II) in General Electric's Experimental Mechanics Laboratory. The i loop included a canned rotor pump which provided sufficient flow to in-sure that the specimen was subj?cted to a refreshed environment. Dissolved oxygen level was controlled by a gas control system that continually 9-2
purged a gas mixture through the makeup tank supplying water to the loop. A schematic of the test loop is shown in Figure 5. The level of dissolved oxygen and conductivity of the loop water was continuously monitored during testing. Table 1 lists the environment specifications. The .9T-WOL specimen was fatigue precracked in room temperature air prior to test to produce a s.100" long precrack. A test load of 5000 pounds was chosen to produce a stress intensity level (K) of N37 ksiG. The loading was divided into two phases, shown in Figure 6. Phase I (cyclic loading) was to insure that an active crack existed. Phase II (constant loading) was the primary test phase wherein all crack growth data is obtained. The total time on test for the 316 stainless steel specimen was 1351 hours, with the initial 248 hours under cyclic loading (Phase I). At the test completion, the specimen was broken apart and the total crack growth measured using an x-y traversing bed optical microscope. The fracture surface morphology indicated an average of s0.020" of intergranular stress corrosion crack growth (IGSCC) under the constant load phase. The average crack growth rate at a K level of 37 ksiG was determined to be 1.6x10-5 ; in/hr, with an upper bound of 7.5x10-5 in/hr for the heavily attacked 1 regio ns . The results from the constant load tests performed on 316 SS Nine Mile l oint safe end material are plotted with the data base from previous tests in Figure 7. The data lie among other data that is from heavily sensitized l material, typical of post weld heat treated material . The crack growth behavior, therefore, is consistent with previous test results on highly sensitized material and supports the GE crack growth prediction cethodology. 9-3
C. MODEL QUALIFICATIONS AND PREDICTIONS 1 Predictive methods developed as part of the Large Pipe Program were used to evaluate crack growth behavior in the safe end heat affected zone of the Nine Mile Point large diameter recirculation system. The predictive methodology developed assumes the existence of a fully cir-cumferential/or axial flaw. These geometries are consistent with those observed for Intergranular Stress Corrosion Cracking (IGSCC) in butt- 1 welded pipes. In addition to determining the rate of growth of these flaws, it is necessary to determine the critical flaw size for net section collapse to demonstrate that adequate margin still existed at NMP so that the plant could have operated for at least one additional 12-18 month cycle without exceeding Code type structural margins. The methodology for calculating this size and for adding a factor of r.afety to calculate a smaller " acceptance flaw size" are discussed in Reference 1. The time required for a crack of some initial known size to grow to this acceptance flaw size is the subject of this section. The calculated crack growth under operating conditions, aa, can be used to establish the appropriate time increment between inspections. The methodology for detennining the growth of the flaw size uses LEFM to determine the stress intensity factor at the crack tip and includes residual stress as well as other applied loads. The assumption is made that the stress intensity
. factor, K, is the principal factor controlling the rate of crack growth.
It is a function of the flaw geometry, component geometry, and the stresses. Crack growth rates as a function of stress intensity were deter-mined from laboratory data and are used for the crack growth predictions. 9-4
The crack growth rates depend on material condition, loading history, and environment. To predict the behavior of heavily sensitized material, upper bound crack growth rates are generally used to assure the most con-servative prediction crack deepening. These upper bound rates have been derived from tests perfonned in severely sensitized material, in a condition expected to be worse than that in as-welded pipe. For as-welded pipes, the behavior would be expected to be similar to or slower than that predicted by expected (or average) IGSCC crack growth data and much slower than the upper bound data that can be used to bound field experience accurately. Circumferential Flaw The methodology required input operating stresses and input throughwall residual stresses. These stresses were used to define the stress in-tensity as a function of crack depth for the pipe configuration. It is assumed that the crack is fully circumferential. This output was used with the crack growth data to develop crack depth as a function of time using a time-step integration to arrive at the crack length. This methodology is described in Reference 1. The input stresses for the Nine Mile Point large pipe are given in \ Table 2. For the crack growth calculations, the total operating stress was used to derive the stress intensity, K, as a function of depth listed in Table 3. The throughwall residual distribution was picked to be representative. It is shown on a plot of measured residual stresses in 9-5
Figure 8. Through-wall residual stresses were obtained from a comparable pipe-to-safe-end weld from i large diameter recirculation segment from the KRB plant after operating for approximately the same number of years as NMP-1. l The heavy line superimposed on the other data of Figure 8 is the experimentally measured through-wall residual stress distribution from the furnace sensitized safe end. These residual stresses were obtained by d;.ect strain gage measure-ments and were determined with a high confidence of accuracy. (Re f. 2. ) The measured results for the KRB safe end, which is similar in design to the NMP-1 safe end, are typical of the band of the experimentally determined data for large diameter pipes. Figure 8 illustrates the comparison, and lends confi-dence to the use of the existing through-wall residual stress data base. Residual axial stresses were also calculated analytically using finite element methods (Figure 9). The magnitude of the predicted stresses agreed very well with the experimentally determined residual stresses and those used for the crack growth analysis. The listing of stress intensity as a function of crack depth is listed in Table 4. Because the method uses linear elastic fracture mechanics, these two K-solutions for operating stresses and residual stresses are superimposed to arrive at the driving force for crack growth. The crack growth data used is displayed in Figure 7. For the evaluations performed, two crack growth evaluation curves were used: the upper bound, shown as the solid line and expected IGSCC crack growth shown as a < dashed line. The expected curve is comprised of a crack growth rate as a function of K that is one-third (1/3) that of the upper bound. This rate curve fits the data well and therefore represents a good estimate of growth for highly sensitized stainless steel. 9-6
,w
The crack size as a function of time assuming upper bound rates is given in Table 5 and displayed in Figure 10. The table lists the stress intensity and crack growth rate used as well as the crack size and time. Table 6 lists the crack size as a function of time using expected IGSCC crack growth rates. Figure 11 displays these results. Using the results of the predictions for the Nine Mile Point safe end, it is possible to summarize the time for a starting crack, a ,gto grow to a maximum allowable size, a . Table 7 shows the time to grow to 50% of the thickness c as a function of initial size. The table display!: times for both crack growth evaluation curves. Growth is more rapid initially (up to 20%) than at greater depth due to the compressive nature of the residual stresses. Beyond 50%, the time for additional growth is still significant due to the nature of the _ residual stresses. For a degree of sensitization expected in the safe end, an initial circumfer-ential crack of depth 5% of wall is predicted to grow to 20% of wall in about one year and from 20% to 50% of wall in an additional six years. By way of' contrast, a prediction using the upper bound growth rates shows the initial 5% flaw would grow to 20% of wall in a few thousand hours and from 20% to 50% ' in an additional two years. The decrease in growth rates as the crack depth l increases is again due to the compressive nature of the residual stress field in the mid-thickness of the safe end. The allowable circumferential flaw size curve for a 28-inch safe end is shown in Figure 12. The acceptance line represents a safety factor of 2.773, while the 9-7
cross-hatched failure region indicates net section collapse. Average and maximum crack depths representative of those found at Nine Mile Point are plotted, as well as their calculated crack growth in a period of 18 months (using the upper bound crack growth rates). It can be seen that even the worst case cracks still remain below the acceptance line after the 18 month period. Even in the event that a growing crack were to cross the acceptance case on the failure diagram, it is likely that a portion of the crack would penetrate the wall well before entering the net section collapse region. Typical cracks have considerable variability in depth. The crack depth profile of Loop 15, SW-12 is shown on Figure 13. Because the deepest parts of the crack are typically the fastest growing (Figures 10,11), it is expected that a small fast growing crack front would penetrate the wall in advance of the remainder of the cracks, producing a leak-before-break condition. Axial Crack The fracture mechanics methodology for an axial crack analysis requires hoop pressure and residual stress levels to define the stress intensity as a func-tion of crack depth. A pressure stress of 15.5 ksi and residual hoop stress of 38.5 ksi were used as representative stresses. The residual hoop stress was calculated analytically using finite element methods similar to those ( used for the axial through-wall residual stress (Figure 14). The magnitude of the residual stress was assumed to be constant through wall for the crack growth calculations. The intitial flaw size was estimated to be 10% of 9-8
{ the thickness. The stress intensity (K) for this crack length was 27.8 ksi G . It can be determined from the crack growth data curves (Figure 10) that a K level of s28 ksi M. places the crack growth rates at a plateau of 6x10-5 in/hr for 0.2 ppm oxygen environment. Assuming this rate, it would be predicted that the crack would grow to a length of *0.50 in. in 9 months of operation. It would require an additional 12 months to propagate the crack through-wall at these plateau rates. Even for a limiting case, using the highest 0.2 ppm 02, 288 C data for furnace sensitized Type 304 SS, one can calculate that the maximum growth that could have occurred in the last 10 months of operation; 5220 hours of actual opera-tion, would still not be through wall. From Figure 10 this rate is 1.2 x 10-4 in/hr. The total crack extension determined from this upper bound rate is 0.626 inches. If one presumes that the crack was 15% through wall prior to this last period of operation, 0.158 in. in the 1.05 in. thick safe end, the flaw would have been predicted to be 0.784 in. deep at the end of the ten month period. This is approximately 75% of the total thickness. ihe allowable axial flaw size was determined for a 28-inch safe end (stress ratio = 0.92) and is shown in Figure 15. The acceptance curve contains a safety factor of 3 and is indicated as a solid line. The dashed line is the present extension of the code acceptance limits; however it does not accur-ately model the crack as it approaches through wall. The cross-hatched area represents the failure region. A crack growth curve is shown for an initial flaw size of 0.105 in. It can be seen that the crack growth falls within the acceptance limits and well below the failure region (supporting leak before break). 9-9 l l
D. U. T. DETECTABILITY vs GE0 METRY vs CRACK GROWTH RATE l The results of Section 8 indicate that the 1982 examination was more sensi-tive than the 1981 examination, such that axially oriented cracks with a depth less than a 20% wall thickness might not have been detected in 1981. However, assuming a 20% wall crack was present in 1981, a higher than expected residual stress and an abnormally high crack growth rate would be required to drive the crack through-wall in the time between the 1981 and 1982 exam dates. A crack growth rate for a sensitized material subjected to a more normal residual stress field would require an axial crack pre-existing in 1981 to be 40 to 50 perccnt wall thickness. To help resolve this disparity, a study was per-1. formed to develop a U.T. detectability vs. weld joint geometry vs. growth rate correlation: a) An anlaytical residual stress analysis was developed and results were compared with experimentally determined thru-wall stress profiles to establish a residual stress estimate for axial cracks in the furnace sensitized NMP safe-end welds (this subject is discussed earlier in this section). b) A section of a cracked NMP safe-end was characterized by U.T. examination, and select cracks were removed for a three dimen-sional profiling. A 20 inch circumferential segment of loop 11 pipe-to-safe-end weld FW-22 was selected to develop this U.T. detectability vs. weld joint geometry vs. growth rate correlation. 9-10
In 1981 U.T. examinations were performed at a gain of approximately 6 dB over the reference gain used on the calibration block. As a result, the safe-end scan was effectively perfomed at reference gain. Prior to removal from the recirculation system of the plant in April 1982, NES had examined this weld, and, using improved techniques with a scan gain
- of 10 to 14 dB over reference gain, they had reported the finding of axial and circumferential indications adjacent to a through-wall leaking axial crack. For the 1982 U.T. examinations, transfer measurements of cal blocks and safe-end components were performed. A 6 dB variance was noted. As a result, the 1982 examinations were performed at a scan gain of 20 dB over reference, and the recording criteria was changed accordingly. The effective scan gain of approximately 14 dB over reference produced an 1982 examination significantly more sensitive than the 1981 examination.
1 Figure 16 is a composite photograph of the PT indications on the inner surface of the pipe segment. The 7 cracks are identified by number. After the sec-tion was removed from the plant, personnel of the J. A. Jones Center, and General Electric perfomed a U.T. examination. At the J. A. Jones Center, the indications were examined with a 45 shear wave crystal at 2.25 MHZ. General Electric used a 45 crystal at a frequency of 1.5 MHZ. In addition to duplicating the improved techniques applicable to detection of IGSCC as used by NES in 1982, GE personnel also performed a U.T. examination with the less sensitive techniques used in the 1981 ISI program. (See Section 8 for , a full discussion.) On the basis of these examinations, four cracks (cracks
#1, 4', 6, and 7) were selected for sectioning and determination of the 9-11 !
l l
. _-. _ l
crack profile. Crack 1 could be detected only with the improved techniques (35% FSH at a gain of 16 dB above reference). Using the 1981 techniques this crack would not be a reportable indication. Crack 4 was a stronger reflector. (42% FSH at a gain of 16 dB above reference.) This crack would be only margi-nally detectable with the 1981 technique of using a scan gain of 6 dB above reference. Crack 6 would have been called a reportable indication using the ) 1981 NES techniques. Crack 7, on the pipe side of the weld, was nearly invis-ible, even at the 20 dB gain, unless the examination was made from the surface of the weld crown. And even then, the detectability was poor (5% FSH at reference gain). During a normal ISI, using the improved IGSCC detection methods, crack 7 would likely have gone undetected. The Table below summar-izes the detectability of the cracks selected for sectioning. Detectable
- by 1982 Improved Crack 1981 ISI Methods Techniques
- 1 No Yes 4 Marginally Yes 6 Yes Yes 7 No Marginally Sectioning of the cracks for detennining profile involved first mounting the sample for metallographic examination to view the crack on the ID pipe sur-face. This surface was polished, etched, and photographed for each sample. f Subsequent to the original polish and etch, the metallographic sample was ground to expose a plane 0.050 ic. below the original surface. This plane
- With unground weld crown.
9-12
k was polished, etched and photographed. Then, the samples were again ground to a depth of approximately 0.100 in, below the original surface and again polished, etched, and photographed. This sequential grinding, polishing and etching was repeated at 0.050 inch intervals for each specimen until the crack disappeared. Figures 17 and 18 are sketches of crack profiles result-ing from the sequential sectioning and photographing. Similar results are obtained for each crack. Detectable
- by Detectable
- by Actual
_ Crack 1981 ISI Methods 1982 Improved Techniques Crack Depth 6 Yes Yes 0.052 irl. (50% wall) and 0.20 in. deep secondary crack) 4 Marginally Yes 0.610 in. d (58% wall)
, 1 No Yes 0,620 in.
I (59% wall) t 7 No Marginally 0.400 in. (38% wall) ) The suninary table suggests the limit of detectability for axial cracks during
, the 1981 examination without grinding the weld crowns first was approximately 45% to 60% thru-wall. With the improved technique cracks with depths less than 38% were detectable. (Further study would be required to establish the actual lower limit of detectability of an axial crack in the configurations observed at NMP-1.) Axial crack 7 on the pipe side of the wall has poor detectability primarily due to the interference of the extended weld crown i with the placement of the U.T. crystal, a *With unground weld crown.
9-13
. = _ _ ----. . - - .
It can be concluued from this study that because of the U.T. methods employed and the presence of unground extended weld crowns at the safe end to pipe / elbow welds, axial cracks up to 45 percent thru-wall were probably present at the time of the 1981 ISI examination, but not detected. Wi:h the improved methods used in the 1981 ISI examination, the axial cracking of this depth i would have been detected. 1 l Based on the previously discussed crack growth analysis, cracks in 1981 of 45 percent of wall could be expected to grow through wall and leak in the 9 month period from 1981 to March 1982. E. REFERENCES
- 1. "The Growth and Stabiliy of Stress Corrosion Cracks in large Diameter BWR Piping," Final Report, EPRI NP-2471, July 1982.
?
- 2. " Metallurgical Examination and Destructive Assay of Piping Samples from KRB Nuclear Reactor," Prepared by Southwest Research Institute,
, San Antonio, Texas, EPRI NP-2382-SY-LD, Project T114-2, May 1982. e 9-14
l Table 1 ENVIRONMENTAL CONDITIONS FOR TESTING IN i SIMULATED BWR SERVICE CONDITIONS Tenpera ture 550 F 10 F Pressure 1230 psi ! 20 psi Oxygen 0.2 ppm 0.1 ppm Conductivity 0.5 pmho 0.2 unho pH 6.5 0.5 at 25 C Table 2 ASSUMED OPERATING STRESSES USED IN CRACK GROWTH EVALUATION
- 28" Pipe - Nine Mile Point
- 1. Pressure stress -
7.75 ksi
- 2. Dead weight -
1.0 ksi
- 3. Thermal stress - 2.0 ksi
- 4. Pressure on crack- 1.0 ksi d'
- Based on discussions with H. S. Mehta 9-15
Table 3 STRESS INTENSITY VS. DEPTH, FOR NINE MILE POINT PIPE, DUE TO OPERATING STRESSES CPPCK KI 9TFE9S IEPTH INTENSITY l 0.0100 2.??593 0.0?10 4.06272 0.0520 5.3173' O.07?O 6.??5?5 l 0.0940 7.?299? O.1150 9.07495 0.1260 9.***1* 9.1570 9.70*92 0.1780 10.45.521 0.19CO 11.?4126 0.*200 12.02106 0.2410 12.*37?? 0.2620 19.65144 0.2830 14.46499 0.?C40 15.27*1? 0.?250 16.177d? 0.3460 17.17701 0.?670 18. Ice 22 0.3880 19.?!71? GCRAK01 Assumptions: O.40*O 20.25776 0.4300 MFtX. CFFCK IEPTH OF 0.999 TTH: O.4510 21.N465.
- 22. .
O.4720 29. 2 959
*0 24.76767 1.050 A.4*i40 P5.c4g77 Ft. RTE THICVW99 CF ITCH 6.5 0.5350 27.17*1?
0.5560 29.49??9-29.69780
- 0.5770 6.59?O 31.15A64 CIPOMEPENTIFL CRA0< TN Cn.INDER YR/T1=18 0.6190 ?2.51215 0.6400 39. 4177 0.6610 35.4467' O.6?20 36.97???
0.70?O ??.52035 0.7240 40.09?4? 0.7450 41.74920 0.7660 42.5$964 0.7870 45.89518 Input and calculated throughwall operating stresses POItE X-8 8LUE Y4 TM.f.E Y-(ALC Y-MT 1 0. 0.1175CE 0? A.117"cE 02 0.11921E-06 2 0.20CCCE 00 0.1175CE 02 o.11750E 02 0.
? 0.4000CE 00 0.1175CE n2 0.11750E C? -A.11*?1E-06 4 0.dOOOCE 00 0.117"CE 02 G.11750E 02 -0.11921E-06 5 .O.90CCCE 00 0.1175C{ 02 0.11750E a2 0.11921E-06 9-16
Table 4 STRESS INTENSITY VS. DEPTH DUE TO THROUGHWALL RESIDUAL l STRESS FOR NINE MILE POINT PIPE CFBCK VT STcE?? IEPTH INTEN?ITY 0.0100 5.9?.721-0.0?10 9.62935 0.0520 11.42919 0.0730 12.?1070 0.0a40 12.61??4 0.1150 12.54?O3 0.1360 12.1M46 0.1570 11.56701 0.1780 10.69747 0.19c0 0.62164 0.2200 0.40967 0.2410 7.07731 0.2620 5.50004 0.28?O ?.*00?1 0.3040 ?.?7471 0.?250 0. fM?.9 0.?460 -0.974Q5 0.3670 -?. W 99 GCRAK01 Assumptions-* 0.3880 -4.271?O O.4000 -5.o*080 0.4?00 -7.ft9?? MAX. CFACK EPTH OF- 0.90s THCH: 0.4510 -9.2f70? 0.4720 -10. M335 0.4930 -1?.447?9 PLATE THTCKNES9 CF 1.0"0 INCH 0.5140 -19.o?o?? G.5?50 -15.414?4 0.5560 -16.?7f2? 0.5770 -19.21725 CIPCUMFEFENTT9_ CRACK IN CY1.INDER 'c "=" 0.5090 -19.4111? 0.6150 -20.4?019 0.6400 -20.02f6? . 0.6610 -20.70440 0.6820 -20.01514 - 0.7030 -19.51?71 0.7240 -18.2*?72 0.7450 -17.2014~ 0.7660 -16." 65f 0.7870 -15.?o785 i Input and calculated throughwall residual stresses: FCINT X 8ALUE Y-8 HU.E y-rR.C Y-PIFF 1 A. 0.2000M A2 0.?!4 ACE 02 A.14Q04E 01 2 6.1050CE 00 0.1400M 0? 0.11091E n2 0.2*096E n.5?O14E 01 00 3 0.1710CE 00 0. -0.5?O14E 00 4 0. 2100CE 00 -0.5M00E 01 -0.4*'?M 01 *.?62M-01 5 0. 3150CE 00 -0.1900CE 02 -0.1595?E 02 -0.?147?E 01 6 0. 42COCE 00 -0.21000E 0? -0.207"1E 02 -0.24691E 00 7 0.5250CE 00 -0.le00 M e2 -a.18843E 02 0.84B9BE 00 g o.6?CCCE 00 -0.7000M 01 -0.8B889E 01 0.9999BE 01 9 6.70COCE 00 0. O. t709mi 0I -0 17099Ei al 9-17
Table 5 CRACK DEPTH VS. TIME FOR NINE MILE POINT PIPE MADE USING UPPER B0UND da/dt HINEMILE-11.75KSI, AVG.RS.FS da/dt Wall Thickness = 1.05 (in) , Init ial Crack Depth = .6525 (in) a a/t K(load) K(resid) KCsotal) da/dt Time (in) (in) ksiftn ksifin kalfin (in/he) (hours)
.053 .050 5.34 11.43 16.77 2.98E-95 0 .076 .872 6.43 12.89 19.49 3.84E-05 709 .997 .892 7.33 12.59 19.92 4.67E-95 1200 .116 .118 8.10 12.48 29.58 5.83E-05 1689 .137 .138 8.93 12.18 21.11 5.43E-95 2000 .153 .146 9.54 11.59 21.14 5.46E-85 2300 .175 .166 10.36 10.81 21.18 5.48E-95 2700 .196 .187 11.16 9.70 20.05 5.26E-95 3100 .217 .207 11.91 8.57 28.40 5.02E-85 3500 .237 .225 12.67 7.29 19.96 4.69E-85 3900 .255 .243 13.38 6.07 19.45 4.38E-95 4300 .276 .263 14.21 4.49 18.70 3.96E-85 4000 .295 .281 14.97 2.98 17.96 3.56E-95 5300 .315 .300 15.79 1.37 17.16 3.16E-85 5990 .333 .318 16.59 .05 16.54 2.87E-85 6508 .353 . 3'J6 17.52 -1.54 15.97 2.62E-85 7200 .373 .355 18.48 -3.18 15.39 2.37E-05 0800 .393 .374 19.49 -4.68 14.81 2.14E-85 8988 414 .394 28.56 -6.31 14.25 1.94E-85 9988 434 413 21.62 -7.92 13.78 1.75E-85 11890 .454 .432 22.67 -9.49 13.18 1.58E-05 12200 .474 451 23.72 -19.94 12.78- 1.46E-85 13500 494 478 24.87 -12.39 12.47 1.37E-85 14908 .514 489 26.83 -13.86 12.17 8.29E-85 16400 .533 .507 27.12 -15.23 11.89 1.21E-95 L7998 .553 .526 28.29 -16.54 11.75 1.18E-05 19600 .573 .546 29.57 -17.81 11.76 1.18E-05 21300 .593 .565 30.88 -19.09 11.79 1.19E-05 23808 .613 .584 32.12 -19.74 12.44 1.36E-05 24600 .634 .603 33.53 -20.19 13.33 1.63E-05 26888 .653 .622 34.89 -20.61 14.2e 1.95E-05 2718e .675 .643 36.51 -20.22 16.28 2.75E-05 28100 .694 .661 37.89 -19.47 18.42 3.81E-95 28700 .716 .682 39.48 -18.60 28.80 5.28E-05 29288 .733 .698 48.82 -17.91 22.91 6.80E-85 29580 .756 .728 42.78 -16.99 25.71 9.63E-95 29800 . .773 .737 44.17 -16.58 27.59 1.22E-84 29970 .793 .756 45.30 -16.50 28.72 1.41E-04 30129 9-18
Table 6 CRACK DEPTH VS. TIME FOR NINE MILE POINT PIPE (EXPECTED IGSCC da/dt ASSUMED) l H1HE MILE-11.75KSI, AVG.RS,FS/3 da/dt Wall Thickness = 1.05 (in) Initial Crack Depth = .0525 (in) J a a/t KCload) K(resid) K(t ot al ) da/ot Time (in) Cin) kstiin ksitin kalfin Cin/hr) (hours]
.853 .950 5.34 11.43 16.77 9.93E-06 0 .873 .869 6.27 12.91 18.28 1.24E-05 1999 .893 .889 7.21 12.68 19.88 1.53E-95 3498 .113 .197 7.97 12.43 20.48 1.66E-95 4689 .133 .127 8.80 12.22 21.02 1.79E-95 5808 .153 .146 9.55 11.59 21.14 1.82E-85 6900 .173 .165 19.38 10.07 21.17 1.83E-85 8088 .193 .184 11.83 9.88 28.91 1.77E-85 9108 .214 .294 11.79 8.75 29.54 1.69E-65 19398 .234 .222 12.54 7.50 20.05 1.58E-95 11580 .253 .241 13.32 6.18 19.49 1.47E-05 12880 .273 .268 14.18 4.71 18.81 1.34E-95 14288 .293 .289 14.91 3.10 18.02 1.20E-05 15000 .313 .298 15.69 1.57 17.26 1.87E-95 17598 .334 .318 16.62 .11 16.52 9.53E-06 19600 .353 .336 17.52 -1. 55 15.97 0.73E-06 21600 .378 .368 18.73 -3.49 15.24 7.71E-86 24690 .393 .374 19.48 -4.67 14.81 7.15E-06 26608 414 .394 29.57 -6.32 14.25 6.46E-96 29688 .433 412 21.55 -7.81 13.74 5.88E-06 32698 455 433 22.72 -9.57 13.16 5.25E-86 36600 475 453 23.81 -11.05 12.75 4.84E-06 40600 .494 471 24.91 -12.44 12.46 4.57E-96 44600 .517 492 26.19 -14.07 12.13 4.26E-06 49608 '
I
.533 .588 27.16 -15.29 11.87 4.83E-06 53600 .553 .527 28.32 -16.57 11.75 3.93E-06 58600 .573 .546 29.57 -17.81 11.76 3.94E-96 63688 .593 .564 38.86 -19.97 11.79 3.96E-06 68608 .613 .584 32.29 -19.74 12.45 4.56E-06 73600 .633 .692 33.46 -28.17 13.29 5.39E-86 77600 .656 .625 35.19 -20.68 14.42 6.67E-96 81600 .678 .646 36.72 -20.11 16.61 9.67E-86 84600 .694 .661 37.86 -19.48 18.38 1.26E-85 86180 .714 .600 39.35 -18.66 20.69 1.72E-05 87509 .733 .698 40.86 -17.88 22.98 2.29E-85 88500 .754 .718 42.68 -17.84 25.55 3.15E-05 89300 776 .739 44.37 -16.58 27.79 4.10E-35 89968 .794 .756 45.30 -16.58 28.72 4.79E-85 90300 1
9-19 l
Table 7 TIME REQUIRED FOR STARTING FLAW TO GROW TO 50% AND 75% FLAW - NINE MILE POINT PIPE * (Circumferential Crack Geometry Assumed) A'. Upper Bound da/dt a (initial) a (final) Time 5% 50% 2.45 yr 1 10% 50% 2.25 yr 20% 50% 1.95 yr 30% 50% 1.60 yr 40% 50% 1.0 yr 50% 75% 1.85 yr
- 8. Expected IGSCC da/dt a (initial) a (final) Time 5% 50% 7.35 yr 10% 50% 6.75 yr 20% 50% 5.85 yr 30% 50% 4.80 yr 40% 50% 3.0 yr 50% 75% 5.45 yr
*80% usage assumed 9-20
10~3
/
/
10 -* - s O _ V
'A
_ G V SENSITIZED AT 1150*F. 2 h.0.2 ppm
, 02 (HE AT 04904) (GE - T115-11
- / A SENSITIZED AT 1150*F,2 h;0.2 ppm g / O2 (M E AT 03580) (GE - 7118-1)
! 10-5 _ [g y Q SENSITIZED AT 1150*F. 24 h;
, 0.2 wm O2 (GE - T1181) 2 y h SENSITIZED SEV ERELY. Q.2 ppm O 2 V g (GE - RP1332 2. REF H-36) 0 4
Y O SENSITIZED AT 11MF,24 h;8 ppm 02
$ V V
@ GE - T118-1
@ W ANG. CLARKE - GENEL L @ SOLOMON - GECRO 10-6 - @ M ASAOKI - HITACHI RE-SE ARCH LAS h PARK - ARGONNE N AT LAB IREF H J6)
~
d SENSIT62ED BY WELOING. LTS AT 932'F. 24 h; 8 po n 02 (SR(RE F H-373 10*I I I ' ' l 1 0 10 20 30 40 50 80 70 STRESS INT ENSITY. K Ikst6) l Figure 1. Summary of Constant Load Crack Growth Data (Curves are evaluation curves.) Data collected in 0.2 ppm 02 and 8 ppm 02 water. Different levels of sensitization examined. (Reference 1) l 9-21
.,,,,, 8 32 UNC -_
_m 1.187 l { 2 HOLES
-a>.g g,30*
1r _
} ,
D - 10 24 UNC 1/4 De 2I 7 d ' l a _ _ _ J T 2 HOLES 106 SEE NOTE 1 i--i y o.626 - e. *- DETAIL S -"*"" N e- SEE NOTE 2 1.26b _ DETAIL J 1.2bo e DI A E RE F --e. 2 HOLES S , a e,}}