ML20247A046
| ML20247A046 | |
| Person / Time | |
|---|---|
| Issue date: | 04/30/1989 |
| From: | Wu P Office of Nuclear Reactor Regulation |
| To: | |
| References | |
| NUREG-1344, NUDOCS 8905230018 | |
| Download: ML20247A046 (48) | |
Text
j NUREG-1344 R
i I
Erosion / Corrosion-Induced i
Pipe Wall Thinning in U.S. Nuclear Power Plants U.S. Nuclear Regulatory Commission Office of Nuclear Reactor Regulation P.C. Wu ps**""uq, e
HP2388a S*
a 1344 R PDR y
I
AVAILABILITY NOTICE Availability of Reference Materials Cited in NRC Publications Most documents cited in NRC publications will be available from one of the following sources:
1.
The NRC Public Document Room, 2120 L Street, NW, Lower Level, Washington, DC 20555 2.
The Superintendent of Documents, U.S. Government Printing Office, P.O. Box 37082, Washington, DC 20013-7082 3.
The National Technical information Service, Springfield, VA 22161 Although the listing that follows represents the majority of documents cited in NRC publica-tions, it is not intended to be exhaustive.
Referenced documents available for inspection and copying for a fee from the NRC Public Document Room include NRC correspondence and Internal NRC memoranda; NRC Office of Inspection and Enforcement bulletins, circulars, information notices, inspection and investi-gation notices Licensee Event Reports; vendor reports and correspondence; Commission papers; and applicant and licensee documents and correspondence.
The following documents in the NUREG series are available for purchase from the GPO Sales Program; formal NRC staff and contractor reports, NRC-sponsored conference proceed-ings, and NRC booklets and brochures. Also available are Regulatory Guides, NRC regula-tions in the Code of Federal Regulations, and Nuclear Regulatory Commission issuances.
Documents available from the National Technical Information Service include NUREG series reports and technical reports prepared by other federal agencies and reports prepared by the Atomic Energy Commission, forerunner agency to the Nuclear Regulatory Commission.
Documents available from public and special technical libraries include all open literature items, such as books, journal and periodical articles, and transactions. Federal Register notices, federal and state legislation, and congressional reports can usually be obtained from these libraries.
Documents such as theses, dissertations, foreign reports and translations, and non-NRC conference proceedings are available for purchase from the organization sponsoring the publication cited.
Single copies of NRC draft reports are available free, to the extent of supply, upon written request to the Office of Information Resources Management, Distribution Section, U.S.
Nuclear Regulatory Commission, Washington, DC 20555.
Copies of industry codes and standards used in a substantive manner in the NRC regulatory process are maintained at the NRC Library, 7920 Norfolk Avenue, Bethesda, Maryland, and j
are available there for reference use by the public. Codes and standards are usually copy-righted and may be purchased from the originating organization or, if they are American National Standards, from the American National Standards institute,1430 Broadway, New York, NY 10018.
,s' J
)
Erosion / Corrosion-Induced Pipe Wall Thinning in U.S. Nuclear Power Plants Manuscript Completed: March 1989 Date Published: April 1989 P.C. Wu Division of Engineering and Systems Technology Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Weshington, DC 20555 y au
/
ABSTRACT Erosion / corrosion in single-phase piping systems was not rences in nuclear power plants. In addition, efforts by the clearly recognized as a potential safety issue before the NRC, the industry, and the ASME Section XI Committee
. pipe rupture incident at the Surry Power Station in to address this issue are described. Finally, results of the December 1986. This incident reminded the nuclear in-survey and plant audits conducted by the NRC to assess dustry and the regulators that neither the U.S. Nuclear the extent of erosion / corrosion-induced piping degrada-Regulatory Commission (NRC) nor Section XI of the '
tion and the status of program implementation regarding American Society of Mechanical-Engineers (ASME) crosion/ corrosion monitoring are discussed. This report Boller andPressure Vessel Code require utilities to manitor will support a staff recommendation for an additional crosion/ corrosion in the secondary systems of nuclear regulatory requirement concerning crosion/ corrosion powerplants. This report provides a brief reviewof the monitoring.
crosion/ corrosion phenomenon and its major occur-l l
l l
l l
i I
1 l
iii NUREG-1344 l
j J
i Contents 1
Abstract......................................................................................... m Abb re via t ion s.................................................................................. vii 1 I n t rod u c t io n................................................................................... 1 2 Major Incidents Of Pipe Wall Thinning And Rupture In Feedwater Systems............................. 3 2.1 Erosion and Rupture of Heater Drain Piping................................................. 3 2.2 Fe ed wa t er Lin e R u pt u re................................................................... 3 2.3 Catastrophic Rupture of Feedwater Line.................................................... 4
]
2.4 Severe Pipe Wall Thinning of Feedwater Lines............................................... 4 J
' 2.5 Accelerated ";r e Wall Thinning.............................................................. 5
)
2.6 Ot h er Incidents of Pipe Wall 'Ihinning....................................................... 5 3 Codes, Standards, And Regulatory Requirements Of Carbon Steel Piping............................... 7 4 Industry Effort To Address The Erosion / Corrosion Issue........................................... 11 5 NRC Efforts To Address The Erosion / Corrosion Issue........................................... 13 5.1 Issuing Information Notices................................................................ 13 5.2 Organizing Technical Panel Discussion.................................................... 13 5.3 Requesting ASME to Address the Erosion / Corrosion Issue...................................... 14 5.4 Issuing NRC B ulletins.............................
........................ 14 6 Summary Of Licensee's Responses To NRC Bulletin 87-01............
........................... 17 6.1 Design and Fabrication Code or Standard................................................ 17 6.2 Pipe Wall-Thinning Monitoring Program................................................... 17 6.3 Frequency and Method of Inspection................................................... 17 6.4 Affected Systems and Com pon ents......................................................... 17 7 O nsit e In spection.......................................................................... 19 7.1 Inspection Crit eria................................................................. 19 7.2 Summary of Inspection Results.......
................................................. 19 7.2.1 Erosion / Corrosion Monitoring Program................................................. 19 7.2.2 Corrective Actions and Repair / Replacement Criteria................................... 20 7.2.3 Inspector Qualification and Training Program........................................... 20 7.2.4 Overall Program Assessment....................................................... 20 7.3 Plant-Specific Inspection Findings........................................................ 20 8 Co n cl u sio n................................................................................. 27 9 R e fe re n ce s.............................................................................. 29
. Appendix A Tables Table 3.1 Design codes standards for typical BWR piping systems (high.cnergy systems).................. 7 Tab!c 3.2 Design codes / standards for typical PWR piping systems................................. 8 Table 6.1 Plants experiencing wall thinning in the feedwater condensate system........................ 18 1
I v
NUREG-1344 i.
l l
ABBREVIATIONS l
l ANSI American National Standards Institute ASME American Society of Mechanical Engineers ASNT American Society of Nondestructive Testing ASTM American Society for Testing of Material AVT all-volatile treatment l
BNL Brookhaven National laboratory BOP balance of plant Blu British thermal unit l
BWR boiling-water reactor CFR Code of FederalRegulations l
CRT cathode-ray tube EPRI Electric Power Research Institute FAC flow-assisted corrosion GTAW gas tungsten arc weld IITG R high temperature gas reactor IN information notice INPO Institute of Nuclear Power Operations ISI inservice inspection LER licensee event report MSR moisture separator reheater MWe megawatt electric NDE nondestructive examination NRC Nuclear Regulatory Commission NRR Office of Nuclear Reactor Regulation NUM. ARC Nuclear Utility Management and Resource Council PORV power-operated atmospheric relief valve ppb parts per billion psig pounds per square-inch gauge PWR pressurized-water reactor RCS reactor coolant system RFO refueling outage RG regulatory guide RT radiographic testing RWCU reactor water cleanup SER significant event report SMAW shielded metal arc weld SRP Standard Review Plan l
USAS United States of America Standards l
UT ultrasonic testing vii NUREG-1344 l
1 INTRODUCTION Erosion / corrosion, or flow assisted corrosion (FAC), is a Water Chemistn form of material degradation that can affect metallic ma-Chemistry parameters affecting crosion/ corrosion are terials that are normally resistant to corrosion because oxygen concentration and pH. The erosion / corrosion they are protected by an oxide film that forms on the sur.
wear rate of carbon steel increases significantly in the pH face. However, turbulent and fast-flowing water or wet range of 7 to 9.The rate drops sharply at pH levels above steam wears a,way the protective film and leads to contm' 9.2. Highly dissolved oxygen in water reduces the rate of ued dissolution of the underlymg metal. Erosion /
crosion/ corrosion by keeping the steel surface passive. It corrosion, or FAC,,s clearly different from crosion that is has been reported (Huijbregts,1984) that iron release i
caused by mechanical processes such as abrasion (caused rates from carbon steel in pure water (neutral pH) de-by particles in water), impingement (caused by water creased by up to 2 orders of mr.gnitude for the tempera.
droplets m steam), and cavitation (caused by collapsing ture range of 38 to 204 *C with increasing oxygen concen-gas bubbles).
tration from 1 to 200 parts per billion (ppb). The precise oxygen level required to prevent erosion / corrosion de-Erosion / corrosion can occur in both single-phase and pends on other factors such as pH level and the presence two-phase carbon steel systems. It is basically a material of contaminants. However, operating experience indi-transport process. Carbon steel piping that has been af.
cates that little erosion / corrosion has occurred in the fected by erosion / corrosion under single-phase condi.
condensate-feedwater system of boiling-water reactors tions shows evidence of uniform wall thinning similar to (BWRs) where the dissolved oxygen level is recom.
~ that caused by general corrosion. In the case of two-phase mended to be around 30 ppb.
flow, the damaged surface has the appearance of " tiger striping."
Temperature Erosion / corrosion-induced piping degradation has been Substantial research has been performed to establish the reported (Bignold,1980) in single-phase carbon steel sys-main factors that control erosion / corrosion. Those factors tems within the temperature range of 80 to 230*C, that control the crosion/ corrosion of carbon steel in water whereas for two-phase lines the temperatures ranges are discussed below, from 140 to 260'C.The temperature at which maximum crosion/ corrosion occurs changes depending on water chemistry; however, it is about 130 to 150*C under single-Alloy Composition phase conditions.
Alloy composition significantly affects the resistance of Pipmg Des.ign and Hydrodyna m. Conditm.
c ns carbon steel to crosion/ corrosion. By increasing the alloy content (e.g., chromium, molybdenum. copper), the resis-Piping configuration and flow rate also strongly affect tance of carbon steel to erosion / corrosion improves sig-erosion / corrosion rates because geometry and flow rate nificantly. Field experience has shown that carbon steel control the mass-transfer rate of oxide dissolutive prod-piping with a chromium content of 0.02 percent haslittle ucts. laboratory results have confirmed that the mass-or no resistance to crosion/ corrosion in the secondary sys-transfer coefficient is the controlling parameter. Unfor-tem of pressurized-water reactors (PWRs) while using tunately, local mass-transfer coefficients are dependent 2-1/4 percent chromium,1 percent molybdenum (2-1/4 on local geometrical discontinuities (such as a backing Cr-1 Mo) steel improves piping resistance to erosion /
ring for welding) and, at this time, they can only be de-corrosion by a factor of 4. Furthermore, austenitic stain-rived empirically or by estimation. Eliminating the local less steels are practically immune to erosion / corrosion geometrical discontinuities is an important step toward
- attack, reducing crosion/ corrosion damage.
l 1
2 MAJOR INCIDENTS OF PIPE WALL THINNING AND RUPTURE IN FEEDWATER SYSTEMS 2.1 Erosion and Rupture of Heater rate at this location was 20 to 24 feet per second at normal P" lin8 C nditi nS f approximately 450 psig and Drain Piping 350*F. Apparently, the pipe had been mstalled as a modification in about 1977 to aid in maintaining heater The'1,roj.an Nuclear Power Plant is a 1080 megawatt elec-drain tank levels during plant startup, but was not in-tric(MWe) pressurized waterreactor(PWR)designedby tended to carry full flow during normal full-power opera-Westinghouse. It is located 32 miles north of Portland, tion. However, as the result of operational problems, the Oregon, and is operated by Portland General Electnc.
pipe did become the normal flow path. Subsequent in-On the evening of March 9,1985, the Trojan reactor was SPections showed that the only other section of piping operating at 100 percent power. Average coolant tem.
where significant erosion / corrosion had occurred was at a I
perature was 585*F and reactor coolant system (RCS) 10-to-14-inch expander section downstream of a 10-inch pressure was 2235 psig. At 9:50 p.m., a reactor trip oc-control va:ve. Although minimum wall thickness was still curred from automatic actuation of the reactor protection met at that location, repairs were made during the 1985 system following a main turbine trip.The turbine trip was outage. The damaged section pipe was replaced, and caused by a spurious main turbine bearing high-vibration power operation resumed on March 15,1985.
signal.The reactor protection system and plant safety sys-l tems functioned as designed during the transient. Follow-2.2 Feedwater Line Rupture ing the turbine trip, the resulting automatic main feed-The Haddam Neck Nuclear Power Plant is a 569 MWe water isolation produced a pressure pulse to approxi-PWR designed by Westinghouse. It is located 13 miles mately 875 psig in the heater drain and feedwater sys-cast of Meriden, Connecticut, and is operated by Con-tems,as expected. However, the pressure surge caused an necticut Yankee Atomic Power, eroded section of the 14-inch-diameter heater drain pump discharge piping to rupture, resulting in the release On March 16,1985, with Haddam Neck operating at 100 of a steam-water mixture of approximately) 350 *F into the percent power, the operators in the control room heard a 45-foot (ground-level) elevation of the turbine building.
" pop" from the turbine building at 8:05 p.m. Security and In addition to the fire suppression (deluge) system actua-health physics personnel notified the control room of a j
tion by heat sensors in the turbine building and damaged steam leak in the northeast, lower-level area of the tur-secondary plant equipment, one member of the plant op-bine building. The control room operator notified the crating staff received first and second degree burns on 50 shift supervisor who was making a tour for plant status.
percent of his body from the high temperature fluid.
The main control board indications appeared normal.
Because of the ruptured piping and steam and water The shift supervisor and secondary side control operator buildup in the turbine building, condenser vacuum was investigated the steam leak. Steam and water appeared to lost approximately 4 minutes after the reactor trip. Loss be coming from the area of the IB feedwater heater nor-of vacuum rendered the steam dump system inoperable, mal level control valve. Steam was blowing toward the so the steam line power-operated atmospheric relief steam generator feed pumps. The shift supervisor or-valves (PORVs) were used to control steam pressure and dered a manual trip of the reactor and turbine because of plant temperature. Makeup water for the steam genera.
the possibility of grounding the steam generator feed tors was supplied from the condensate storage tank using pump motor (s) and/or heater drain pump mot or(s). In ad-the auxiliary feedwater system.The plant was maintained dition, the exact k) cation of the pipe rupture had not been determmed.
j in hot standby until 3:50 a.m. on March 10,1985, when a forced cooldown, using PORVs and auxiliary feedwater, The reactor and turbine were manually tripped. Steam was initiated. The plant entered hot shutdown at 10:20 generators 1 and 3 were noted as having lower secondary a.m. and the residual heat removal system was placed in side levels than steara generators 2 and 4.The steam gen-service 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> later-erator low levels were attributed to the tripping of reactor j
coolant pumps 1 and 3 during the event, which caused The ruptured section of the carbon steel heater drain sys-shrinkage followm, g idimg of the loop. These pumps are tem piping (American Society for Testing Material r quired to trip by design during four-loop operation fol-
[ ASTM] A-106 Grade B) had been severely damaged by I wmg a reactor scram. These pumps were restarted by erosion / corrosion. Because the feedwater flow out of a 8:30 p.m.
normally open 14-inch manual globe valve was directed against the pipe wall, the pipe wall at the rupture location After manually tripping the reactor, the A steam genera-had eroded from a nominal thickness of 0.375 inch to a tor feed pump was shut down. The A and B auxiliary i
thickness of approximately 0.098 inch. The system flow steam generator feed pumps were started to ensure l
l 3
2 MajorIncidents availability in the event that both feed trains required iso-water temperature, pressure, and enthalpy are 370'F, lation.
450 psig, and 346 Btu /lb, respectively. During these con-ditions the fluid is in the single-phase liquid-only regime.
Condenser vacuum was lost because of air leaking through the turbine seals. The gland steam supply had The event was initiated by the main steam isolation valve previously been isolated for reasons unrelated to this on steam generator C failing closed. Because of the in-event. The high-pressure steam dump was lost when the creased pressure in steam generator C that collapsed the condenser vacuum reached 20 inches mercury. This was voids in the water, the reactor tripped on low-low level in recognized after the primary side control operator noted that steam generator. A 2-by-4-foot section of the wall of an increase in average primary coolant temperature. In the suction line to the main feedwater pump A was blown response to the increase in primary temperature, the sec-out and came to rest in an overhead cable tray. ne break ondary side control operator manually opened the atmos-was located in an elbow in the 18-inch line about 1 foot pheric steam dump at about 8:15 p.m. Primary side tem-from thc 24-inch header.The lateral reactive force gener-perature and pressure were controlled by the atmos-ated by scraping feedwater completely severed the suc-pheric steam dump until restart.
tion line.The free end whipped and came to rest against the discharge line for the other pump.
Automatic initiation of auxiliary feedwater flow occurred at 8:13 p.m.as a result of the low levelin steam generators Steam flashing from the break and condensing in control 2 and 4,44 percent and 45 percent wide-range level, re-cabinets and in open conduit piping apparently caused the spectively. The levels of steam generators 2 and 4 re.
fire suppression system to actuate, resulting in release of mained below 45 percent for 20 seconds.The levels were halon and carbon dioxide in the emergency switchgear low because of increased boil-off, which was caused by room and in various cable tunnels and vaults and in the loops 1 and 3 being idled when their reactor coolant cable spreading room.
pumps were tripped. After the reactor was tripped, the Investigation of the accident and examination of data by pipe rupture was located. The pipe had ruptured down-the licensee, NRC, and others led to the conclusion that stream of the IB feedwater heater normal level control failure of the piping was caused by erosion / corrosion of valve, which is a Masonellan Camflex valve. The actual the carbon steel pipe wall. Although crosion/ corrosion rupture was approximately 1/2 inch by 2-1/4 inches.
pipe failures have occurred in other carbon steel systems, The pipe rupture occurred because the flow exiting the particularly in small-diameter piping in two-phase sys-1B feedwater heater normal level control valve impinged tems and in water systems containing suspended solids, directly on the pipe surface and severely eroded the pipe there have been few previously reported failures in large-in that area.The eroded section of pipe was replaced. In diameter systems containing high-purity water. Consis-addition, the corresponding pipe on the A feedwater train tent with general industry practice, the licensee did not was checked for crosion. The licensee already had a pro-have in place an inspection program for examining the gram for monitoring pipe elbows for crosion in the main thickness of the walls of feedwater and condensate piping.
steam and condensate and feedwater systems. The licen-see plans to include sections of pipe adjacent to flow con-2.4 Severe Pipe Wall Thinning of trol valve configurations, similar to the pipe that run-Feedwater Lines tured, m the plant s reliability engmeermg program for monitoring erosion of secondary system pipe elbows.
During theJune 1987 outage at the Trojan Nuclear Power Plant, it was discovered that at least two areas of the 2.3 Catastrophic Rupture of Feedwater straight sections of the main feedwater piping system ex-Line perienced wall thinning to an extent that the pipe wall thickness would have reached the minimum thickness re-The Surry Nuclear Power Station.. located on the James quired by the design code (American National Standard River approximately 12 miles from Newport News, Vir.
Jnstitute (ANSI) Standard B31.7, " Nuclear Power Pip-ginia, is operated by the Virginia Power Company, ing") during the ne.xt refueling cycle. These areas are m safety-related portions of the American Society of Me-On Tuesday, December 9,1986, at 2:20 p.m., both units at chanical Engineers' Boiler and Pressure Ves3e/ Code the Surry Power Station were operating at full power (ASME Code),Section III, Class 2 piping inside contain-when the 18-irch suction line to the main feedwater ment.
pump A for Unit 2 failed catastrophically.
The pipe wall had thinned in both horizontal and vertical Units 1 and 2 are identical. In each unit, feedwater flows runs that were at least seven pipe diameters downstream from a 24-inch header to two 18-inch suction lines that of elbows or other devices that can cause flow distur-each supply one of two main feedwater pumps. At maxi-bance. Criteria developed by the Electric Power Research mum load under normal conditions, feedwater flow Institute (EPRI Users Manual NSAC-112, "CHEC" through each pump is 5 million pounds per hour. Feed-
[Chexal Horowitz Erosion-Corrosion), June 1987)would NUREG-1344 4
l
2 MajorIncidents not have required the pipe wall in these straight sections feedwater piping and in other non-safety-related conden-to be examined.
sate piping.
In addition, the licensee discovered approximately 30 ad-On the basis of partial inspection results, the licensee in-ditional areas of the main feedwater piping system where dicated that the broad area thinning rate for the replace-the pipe wall had thinned so that the thickness of the pipe ment piping, installed during the last refueling outage, is wall was either less than the minimum thickness required roughly 60 mils / year. The maximum localized thinning by the design code or would have eroded to the minimum rate is 90 mils / year. These rates were higher than the required thickness during the next operating cycle. Of 20-to-30 mils / year rate estimated previously. The esti-these areas,10 were in the safety-related portions of the mated rate of 20 to 30 mils / year was based on a single system, while the rest were in non-safety-related por-measurement and an assumption that wall thinning had '
tions. All of these 30 additional areas were in regions that been progressing linearly since initial full-power opera-the EPRI criteria would have indicated as needing exami-tion was achieved. This new rate of wall thinning, which is nation.
based on a second data point, indicates that significant wall thinning may have coincided with a reduction in feed-Pipe wall thinning of the condensate and feedwater sys-water dissolved-oxygen concentration following replace-tem was discovered when the piping inspection program ment of the steam generator. The lower rate of wall thin-at Trojan was expanded to include single-phase piping, in-ning associated with a higher feedwater dissolved-oxygen cluding all safety-related high-energy carbon steel piping concentration is consistent with the low rates of erosion /
inside containment.
corrosion reported in NRC Information Notice (IN) 88-17, " Summary of Response to NRC Bulletin 87-01, Results of the licensee's failure analysis and the staff's in-
' Thinning of Pipe Walls in Nuclear Power Plants,'" for dependent verification indicate that erosion / corrosion boiling-water reactors (BWRs), which typically operate at coupled with cavitation caused by severe flow conditions a feedwater dissolved-oxygen concentration of approxi-at the pump discharge elbows are the primary mechanism mately 30 ppb.The licensee is continuing its failure analy-that caused pipe wall thinning of the feedwater line at sis to determine the cause(s) of the increase in the esti-Trojan.
mated pipe wall-thinning rate.
2.5 Accelerated Pipe Wall Thinning 2.6 Other Incidents of Pipe Wall During the September 1988 outage at the Surry Nuclear Power Station, the ticensee discovered that pipe wall thin-In addition to the major incidents of pipe wall thinning ning had occurred more rapidly than expected. On the and rupture of feedwater piping systems, numerous pip-l suction side of one of the main feedwater pumps, an el-ing degradation resulting from erosion-or erosion /
bow that was installed during the 1987 refueling outage corrosion-induced wall thinning has occurred in the sec-lost 20 percent of its 0.500-inch wall in 1.2 years, in addi-ondary systems of many operating nuclear power plants.
tion, wall thinning is continuing in safety-related main A brief summary of these events are described below.
Plant Year Description Reference
]
Oconee 3 1976 Extraction line pinhole leak NRC IN 82-22 1980 Replace erosion / corrosion thinned elbow NRC IN 82-22 Browns Ferry 1 1982 Failure of 8-inch discharge line on the MSR drain pump INPO Significant Event Report (SER) 41-82 Oconee 2 1983 Failure of a 3-to 10-inch expander down stream INPO SER 23-85 of a reheater drain tank
)
Calvert Cliffs 1 1984 Rupture of a 16-inch cibow in a branch line from a cold INPO LER 88-84 reheat steam line Haddam Neck 1985 Pipe rupture downstream of a feedwater heater INPO Licensee Event Report (LER) 305-85006 i
Kewaunee 1985 Rupture of a 2-inch excess steam vent line from a MSR INPO IIR 305-85017 Hatch 2 1986 Rupture of a 20- to 16-inch reducer in an extraction INPO LER 366-86010 steam line Ginna 1986 Failure of a 6-inch cibow of a moisture separator INPO LER 244-86004 reheater drain line 5
i J
- --------s.-_--
- ~ - - - - ----.,, -. - - _ _
L 4
L.
n s
.i i
n
_--m a-
r__u----
_._-s_.--
3 CODES, STANDARDS, AND REGULATORY REQUIREMENTS OF CARBON STEEL PIPING
. The requirements for the construction and inservice in-related and non-safety related components is that the spection of safety-related systems differ from those of sys-safety-related systems are required to meet seismic crite-tems that are not safety related because safety-related ria and requirements for design quality assurance that systems are relied on to provide the capability to prevent complies with Appendix B toTitle 10 of the Code offed-or mitigate the consequences of accidents, remove heat cral Regulations Part 50 (10 CFR 50). Safety-related por-from the reactor, and maintain it in a safe shutdown con-tions of these lines also are required to receive inservice dition. The construction requirements of safety-related inspection and testing under 10 CFR 50.55a(q), which in-systems differ from non-safety-related systems in the ar-vokesSection XI of the ASME Code.Non-safety-related cas of materials inspection and nondestructive examina-systems are not required by any standard or regulatory re-tion of piping system weldments, overpressure protec-quirement to receive inservice inspection. Design codes /
tion, and quality assurance, including third-party standards for typical BWR and PWR systems are listed in inspection. For the main steam and feedwater system, the Tables 3.1 and 3.2, respectively, principal difference between the design of the safety-Table 3.1 Design codes standards for typical BWR piping systems (high. energy systems)
System Code / Standard
- Nuclear boiler ASME Sec. III, ANSI B31.1 Reactor recirculating ASME Sec. III, ANSI B31.1 Control rod drive hydraulic ASME Sec. III, ANSI B31.1 Standby liquid control ASME Sec. III, ANSI B31.1 Residual heat removal ASME Sec. III, ANSI B31.1 IAw-pressure core spray ASME Sec. III, ANSI B31.1 Ifigh pressure core spray ASME Sec. III, ANSI B31.1 Reactor core isolation cooling ASME Sec. III, ANSI B31.
ASME Sec. III,. ANSI B31.1 Filter /demineralyzer RWCU ASME Sec. III ANSI B31.1" Main and reheat steam ASME Sec.111, ANSI B31.1 Auxiliary steam ASME Sec. III, ANSI B31.1 l
Condensate ANSI B31.1 Feedwater ASME Sec. III, ANSI B31.1 Heater vents and drains ANSI B31.1 Main and reactor feed pump turbine seal ANSI B31.1 Moisture separator-reheater ANSI B31.1 Extraction steam ANSI B31.1 l
- 'Ihe safety-related portion of the i ingisdesigned to ASMECode Section!!!. The non-safety-related portion of the piping is ANSI Standard ID.
" Piping is designed to ANSI Standard 1131.1 and fabricated to ASME Code Section III.
I i
3 Codes and Standards Table 3.2 Design codes / standards for typical PWR piping systems System Code / Standard
- Main steam ASME Code Section III, Class 2 ANSI Standard B31.1 Extraction steam ANSI Standard B31.1 Auxiliary feedwater ASME Code Section Ill, Class 2, ANSI Standard B31.1 Main feedwater (outside containment)
ANSI Standard B31.1 Main feedwater ASME Code Section III, Class 2 Steam generator blowdown ANSI Standard B31.1 Heater drains ANSI Standard B31.1 The term non nuclear is not well defined, but as used by sections of the Standard Review Plan (SRP),
many and in the response below it describes piping not NUREG-0800, Revision 2 (July 1981).
constructed to Section III of the ASME Code. Power plants built before the adoption of Section III of the SRP Section Iitk ASME Code were constructed to other standards such as 10.4.1 Main Condensers American National Standards Institute (ANSI)/ASME 10.4.7 Condensate and Feedwater System Standard B31.1, " Power Pipmg.
10.4.9 Auxiliary Feedwater System The condensate and feedwater systems of PWRs provide The Filowing regulatory guides also provide guidance feedwater at the required temperature, pressure, and with ry,ard to quality group classification (applicable flow rate to the secondary side of the steam generators.
codes and standards), scismic design requirements, and Condensate is pumped from the mam condenser hotwell quality assurance requirements for components of nu-by the condensate pumps, passes through the low-clear power plants.
pressure feedwater heaters to the feedwater pumps, and then is pumped through the high pressure feedwater Reculatory Guide Tjlk heaters to the secondary side of the steam generators.
That portion of the condensate and feedwater system 10-1.26, Revision 3 Quality Group Classifications cated within the turbine building and the portion of the (February 1976) and Standards for Water,
feedwater lines between turbine building up to the con-Steam, and Radioactive-tainment isolation valves located outside the reactor con-Waste-Containing Compo-tainment building are not classified as safety related.The nents of Nuclear Power Plants portion of the feedwater system from the containment 1.29, Revision 3 Seismic Design Classification isolation valves located outside the reactor contamment (September 1978) buildmg up to and mcluding the secondary side of the steam generators are within the nuclear portion of the The construction codes and standards applicable to the power plant and are classified as safety related.
auxiliary feedwater system and the safety-related portion of the main feedwater system at Surry Units 1 and 2 are as An auxiliary feedwater system is connected to the main follows:
feedwater system and normally operates during startup, hot standby, and shutdown to provide feedwater to the Portions of main feedwater piping-ANSI Standard e
steam generators. This system also functions as an emer.
B31.1-1967 supplemented by ASME Code Case gency system for the removal of heat from the primary N-7. Auxiliary feedwater piping-ANSI Standard system when the main feedwater system is not available B31.1-1967 and for emergency conditions including small loss-of-coolant accidents.The entire auxiliary feedwater system Pumps-such as auxiliary feedwater pumps-manu-e is classified as a safety-related system.
facturer's standards Regulatory guidance with regard to the auxiliary feed-e Valves-manufacturer's standards and ANSI Stan-water system, the main feedwater system, main condens-dard B31.1-1967 and related standards such as Stan-ers, and condensate system is provided in the following dard B16.5 NUREG-1344 8
3 Codes and Standards The construction codes and standards applicable to the Section XI of the ASME Code currently does not contain non safety-related portions of the condensate and feed-a requirement to explicitly measure wall thickness to de- -
water system at Surry Units 1 and 2 are as follows:
tect thinning. Weldments are inspected by nondestructive examinations to determine if indications are within allow-Condensate and feedwater piping-ANSI Standard '
able limits.
e 1131.1-1967, " Power Piping Code" Classification of the Piping System e
Pressure vessels such as feedwater heaters-ASME Boiler and Pressure Vessel Code,Section VIII, Regulatory Guide (RO) 1.26, SRP Section 3.2.2, and 10
" Pressure Vessels" CFR 50.55 provide the staff's criteria for classifying the main steam line and the feedwater line from the reactor
' Pumps, such as condensate and feedwater pumps, -
up to and including the outermost isolation valve as Qual-e and steam turbines-manufacturer's standards ity Group A (ASME Code Section III, Class 1). RG 1.26 also classifies the main steam line up to but not including e.
Valves-manufacturer's standards and ANSI Stan.
. the turbine stop valve and bypass valves as Quality Group dard B31.1-1967 and related standards such as B (ASME Code Section III, Class 2)(see Table A.1, SRP ANSI Standard B16.5 Section 3.2.2). Alternatively, for BWRs containing a shut-
-off valve (in addition to the two containment isolation Section XI of the ASME Code, " Rules for Inservice In-
. valves)in the main steam line and in the main feedwater spection of Nuclear Power Plant Components,"is used by line, Quality Group B standards should be applied to the licensee to provide guidance during plant operation those portions of the steam and feedwater systems ex-on inservice inspection of components and inservice test.
tending from the outer most containment isolation valve
- ing of pumps and. valves that are safety related because up to and including the shutoff valve (see SRP Section Surry was constructed before the development of ASME 3.2.2). Weldments in Quality Groups A and B are subject
. Code Section III, which is applicable to safety-related sys-to periodie inservice inspection in accordance with Sec-tems today.
tion XI of the ASME Code per 10 CFR 50.55(a)(g).
l 9
"- NUREG-134 c-,
4 INDUSTRY EFFORT TO ADDRESS THE EROSION / CORROSION ISSUE Shortly after the December 1986 accident at the Surry corrosion damage in a single phase secondary coolant sys-Nuclear Power Station, the nuclear industry took the in-tem. To assist utilities in identifying areas of carbon steel itiative to address the single-phase crosion/ corrosion is-
. piping that might be undergoing crosion/ corrosion dam-sue by ensuring that initial inspections would be con-age under single-phase conditions, NUMARC and EPRI ducted at all U.S. domestic power plants.
developed a recommended inspection plan to monitor pipe wall-thinning problems. The major elements of this In March 1987 the Nuclear Utility Management and Re-plan consist of (1) where to look for locations that are sus-source Council (NUM ARC) established a working group ceptible to crosion/ corrosion attack, (2) how to look for on erosion / corrosion; that group developed a recom-these potentially vulnerable locations, (3) when to look mended industry program (Appendix A) to address the is-for them, and (d) what are the repair / replacement crite-sue. The program identifies potential evidence of a ria.The NUMARC program is designed to provide the in-
- single-phase crosion/ corrosion problem and provides dustry with a predictive capability with regard to the pipe guidelines for utilities to follow. In general, the recom-wall-thinning rate as a function of operating time for a mended NUMARC guidelines are threefold: (1) to con-given component and to provide a cost-benefit analysis duct appropriate analysis and a limited but thorough with regard to repair / replacement options.The staff re-baseline inspection program, (2) to determine the extent viewed NUMARC's program on crosion/ corrosion in of thinning,if any, and to repair / replace components as single-phase lines in June 1987 and found it to be accept-l necessary, and (3) to perform followup inspections (to able (Appendix A).
confirm or quantify thinning) and longer term corrective actions (i.e., adjust chemistry, operating parameters, or others) as appropriate. The NUM ARC program specifies On March 17,1988, NUMARC representatives provided an initial inspection requirement of 15 fittings, with maxi-a briefing on industry efforts to implement a program to mum reliance for their selection based on engineering address crosion' corrosion. On the basis of NUMARC judgment and a computer code (such as the CHEC code data,112 of 113 plants have committed to implement an developed by EPRI). This program is developed to opti-erosion / corrosion program for single-phase piping by Oc-mize nondestructive examination (NDE) resources tober 1988. Subsequent to the meeting, the remaining needed forinspection, plant, a new BWR, had an unplanned shutdown and con-ducted an examination of single-phase piping. As of early Before the Surry accident, EPRI research programs had 1988,54 plants had completed inspections. Components helped the industry by identifying twe-phase crosion/
were replaced in 19 of these plants as a result of single-corrosion as a flow-accelerated corrosion process that phase erosion / corrosion. Only one BWR replaced com-leads to wall thinning (metalloss) of carbon steel compo-ponents, demonstrating that the problem is more exten-nents exposed to flowing wet steam. An inspection guide-sive for PWRs. For the 19 plants replacing components, line (Kastner et at D86) was issued to help utilities in de-10 replaced more than five components and 9 replaced veloping their crosion/ corrosion monitoring program for fewer than five components.
the two-phase lines.Ti:e guideline was developed to meet utilities needs to address the erosion / corrosion problem and is based on combining the resultE of the root-cause Although the staff accepted the NUMARC working research work and the firsthand site inspection with prac-group's program, it indicated that assurances are needed tical experience. A methodology for developing a plant-that all plants have systematically addressed the issue of specific, cost-effective, and reliable inspection program two-phase erosion / corrosion and have procedurally im-l was provided, which included rating and prioritizing sys-piemented long term programs for single-phase and two-ems and subsystems and identifying components most phase piping systems. To address these concerns, l
susceptible to severe crosion/ corrosion degradation.
NUMARC is supporting EPRI in the development of a computer-based program for two-phase erosion /
i The Surry accident focused attention on the potential de-corrosion. This program is similar to the single-phase terioration of carbon steel piping because of erosion /
program.
l 11 NUREG-1344 L______________
5 NRC EFFORTS TO ADDRESS TIIE EROSION / CORROSION ISSUE 5.1 Issuing Information Notices tion to determine the long-term operability of the Trojan feedwater systems.
Shortly after the catastrophic failure of the main feed-water pipe at the Surry Nuclear Power Station in Decem.
By letter dated June 22,1988, the licensee provided re-ber 1986, the NRC dispatched an investigation team to suits of its failure analysis regarding pipe wall thinning of determine the cause of the incident and its genericimpli.
the feedwater and condensate system at the Trojan plant.
cation.The team determined that the failure was caused The staff determined that the piping material was consis-by wall thinning because of crosion/ corrosion. On De.
tent with design specification, the damage mechanism cember 16,1986, the NRC issued Information Notice was identified, and corrective actions are adequate to pro-86-106 "Feedwater Line Break," to all nuclear power vide reasonable assurance for continued operation of the plant licensees describing the event and its generic impli.
feedwater piping system at Trojan, including seismic cation so that utilities could review the problem for appli-Category 1 pipmg.
cability to their facilities. Subsequently, extensive investi-gations at Surry revealed that crosion/ corrosion also had 5.2 OrEanizinE Technical Panel occurred in the other pipmg systems and locations. Re-pairs were made by replacing the piping. During the in.
DISct1SSlon vestigation, the NRC issued supplemental information noticer, to all nuclear power plant licensees on Febru-On Jamnary y,19p,,the staff m.ytted experts from sev-ary 13 and March 18,1987.Those notices provided the in-eral engmeermg disciphnes (pipmg design, metallurgy, n destructive examination, water chemistry, corrosion, formation developed through the investigation, including and fluid mechames) to participate in a techmcal panel potentially generic system interaction problems that were discussion on the parameters beheved to have had an im-caused by release of large quantities of feedwater, portant role m the pipe break at Surry and on tbc means to predict and mitigate the effects of crosion/ corrosion in After the Troj.an mcident, the NRC issued another Infor-piping systems.The cause of the Surry failure, which oc-mation Notice 87-36 on August 4,1987, to alert licensees curred in the feedwater piping near the suction side of to a potentially generic prob!cm pertammg to sigmficant one of the feedwater pumps, has been identified as pipe unexpected erosion that resulted m pipe wall thmmng in thinning from crosion/corrosiont however, at least one the safety-related portions of feeJwaterlines. In addition, panel member believes that cavitation crosion cannot be the staff established a task force on July 9,1987.De task corrpletely excluded. The actual failure of the thinned force is made up of staff from different offices of the NRC pipe wall resulted from a system pressure transient, not a as well as consultants, many of whom were mvolved in the classical water hammer event.
resolution of the Surry feedwater line break meident.nis interdisciplinary team has the needed technical expertisc The complex interactions of the individual variables that to investigate the Trojan pipe wall-thinning event and to affect the crosion/ corrosion phenomenon and influence address general issues related to crosion/ corrosion and the rate at which it proceeds, have not been thoroughly pipe wall thmmng.
established by available research activity in this country or abroad. The panel discussion elaborated on the role of On July 22,1987, the task force visited theTrojan plant to those parameters that could have potentially contributed review the event and to evaluate the licensee's corrective to crosion/ corrosion in the feedwater piping system. The actions and other related activities. The task force at-panel made several observations and recommendations, tended briefings, interviewed licensee staff, and made vis-which are summarized as below.
ual examinations of selected piping sections and compo-nents removed from the feedwater lines. In addition, the Observatiorts task force selected piping samples to dependently verify the licensee's analysis.
The phenomena associated with pipe wear as a re-e sult of crosion, crosion/ corrosion, or cavitation have The task force completed its preliminary review of the not been recognized as significant problems by pip-Trojan event on August 13,1987, and determined that ing designers. Consequently, designers do little to the damage mechanism at Trojan was similar to that of accommodate piping wear resulting from these phe-the Surry accident. Although the licensee has provided nomena in their design.
rcasonable assurance for safe operation of the feedwater The NRC initiative on omitting the dynamic effects systems through the end of the 1988 operating cycle, the e
task force will review the results of the licensee's final of postulated pipe ruptures, whenever technically failure analysis and those from the independent verifica-justified, should not be affected by the Surry failure.
13 NUREG-1344 C____________
5 NRC Addresses Issue l
The amount of crosion/ corrosion damage and the quest to the ASME Section XI Committee asking that the e
rate at which it proceeds is a complex phenomenon issue concerning erosion / corrosion-induced pipe wall dcpending on a number of variables.
thinning be addressed.
In the case of Surry and other U.S. plants, the three in response to the NRC requests, the ASME Section XI most important variables influencing the crosion/
Committee established a special working group on pipe corrosion process are material, local fluid velocity /
walf thinning in April 1988 and started to assess the need turbulence, and water chemistry /pH.
of a possible monitoring requirement.The current status concerning the progress made on this issue is summarized Although crosion/ corrosion is not a new or unknown
- below, e
phenomenon, it has received relatively li:lle study in the United States because incidence of recorded With regard to balance-of-plar "OP) piping, the Com-failures has been low and the relationship of the mittee believes the historic ar m.ch used by many utili-variables influencing the proce. sing is complex.
ties is appropriate; namely that utilitics develop their own nondestructive examination (NDE) program to deter-min if there h a problem with BOP piping. With regard Recommendations to safety-relat:d pipmg such as Class 1 and 2 pipmg sys-tems, the Committee agrees with the staff recommenda-The panel beheves that adjustments to pil or oxygen tion; thus, the Section XI special working group on pipe e
content, from sevels now in use to protect steam gen, wall thinning is developing requirements and procedures crators, should not be made without a therough to detect and prevent pipe wal! thmmng resultmg from study of the possiule global effects of such changes erosion /c rr sion in nuclear power plants. It is expected on the entire system.
that the final version, standard code language, vill be an The panel believes that the factors influencing the equitable balance between the existing ASME Section XI e
requirements for Class 1 and 2 piping and the added re-rate at which the crosion/ corrosion phenomenon proceeds in singe-phase systems cannot be ranked quirements resulting from examinations fcr wall thinning.
because the a~i Ale current quantitative data are Currently, Class 1 examinations are predominantly volu-metric while Class 2 examinations are a mixture of volu-insufficient.
metric, surface, and visual. In systems such as the feed-Data resulting from baseline wall thickness meas.
water line,7.59 percent of the welds are required to be e
urements taken as a result of recommendation 2 examined volumetrically.The special working group is re-should be correlated with flow vehicity, turbulence, viewing this requirement to see if the percentage can be water chemistry, temperature, actual material lowered, provided that the welds with highest stress in-chemical composition, and installed original w/t tensity factors are examined.
thickness, to the extent that such data are avai!Ls Codifying pipe wall-thinnmg examinations is not an easy task. Because of the nature of the damage mechanism, ASME should consider the need for providing ap-e several interactive parameters, such as temperature, alloy propriate guidance to system designers on the sub.
composition, pH and dissolved oxygen content of the ject of erosion and crosiodcorrosion in its conven-water, and piping geometry are all important considera-tional pressure piping and sclear piping cojes and tions in selecting the inspection points. Furthermore, the standards.
damage may occur at a variety of k> cations with different 5.3 Requesting ASME to Address the Erosion / Corrosion Issue 5.4 Issuing NRC Bulletins As mentionea above, following the Surry feedwater line An inf, m ' NRC staff survey, conducted during the first rupture, then, wasa technicalmeeting heldat the request wee?,.f 11 ruary 1987, demonstrated (1) that the wall-of thestaff to onsiderthegene.icimplicationsof thisfail-thinto o " ablem is widespread in two-phase lines at nu-ure. Onc of the recommendations made by participants of c! car power plants and (2) that most licensees either did the meeting was that ASME should consider the need for not have a monitoring program for pipe wall thinning or providmg appropriate guidance to system designers on had an inadequate program.
pressure piping and nuclear piping codes and standards.
Main feedwater systems, as well as other power conver-To implement the above recommendation, the staff re-sion systems, are important to safe operation. Failures of quested that the ASME Sectior XI Committee consider active components in these systems, such as valves or this subject during its meeting of March 19,1987. After pumps, or of passive components, such as piping, can re-the Trojan event in Juh - 1988, the staff sent a second re-sult in undesirable challenges to plant safety system NUREG-1344 14
5 NRC AddressesIssue required for safe shutdown and accident mitigation. Fail-pil of water in the system (e.g., pH less than 10) ute of high-energy piping, such as feedwater system pip-ing, can result in complex challenges to thc operating staff system temperature (e.g., between 190 and and the plant because of potential systems interactions of 500'F) high-energy steam and water with other systems, such as electrical distribution, fire protaction, and security sys-fluid bulk velocity (e.g., greater than 10 fels) tems. All licensees have either explicitly or implicitly committed :o maintain the functional capability of high-energy piping systems that are a part of the licensing basis oxygen content in the system (e.g., oxygen con-for the fac!!ity. An important part of this commitment is tent less than 50 ppb) that piping will be maintained within allowable thickness Chronologically list and summarize the results of all values.
inspections that have been performed, for the pur.
As a result of the survey findings, NRC Bulletin 87-01 was pose of identifying pipe wall thinning, whether or issued on July 9,1987.This bulletin required alllicen-not pipe wall thinning was discovered, and any other sees to provide information to the NRC on their crosion/
inspec Lons where pipe wall thinning was discovered I
corrosion experience and monito-ing programs for single-even though that was not purpose of that inspection, phase and two-phase high-energ/ carbon steel piping sys-tems. Specifically, the licensees were requested to pro-Briefly describe the inspection program and in-vide the following information:
dicate whether it was specifically intended to measm waH tMness m wheecr waH te Identify the codes or standards to which the piping n ss nw mnwnts were an Mental deten e
was desi ned and fabricated.
mination.
E Describe the scope and extent of programs for en-Describe what piping was examined and how e
suring that pipe wall thicknesses are not reduced be.
low the minimum allowable thickness. Include in the (e.g., describe the inspection instrument (s),
description the criteria established for; test method, reference thickness, locations ex-l amined, and means for locating measurement l
selecting points at which to taake thickness point (s)in subsequent inspections).
measurements Describe plans for revising existing pipe wall-e determining how frequently to make thickness thinning monitoring procedures or developing new measurements or additional inspection programs.
selecting the methods used to make thickness All licensecs respondad to the bulletin and the staff com-measurements picted its review of the responses in December 1987 Fur-thermore, at the end of September 1988, the staff com-making replacement / repair decisions picted inspection of 10 plants to assess the licensees' ef-forts toward impicmenting their crosion/ corrosion moni-For liquid-phase systems, state specifically whether toring program.He purpose of this report is to summa-e the followmg factors have been cormdered m estab-rize the results of the staff review of the licensecs' re-lishing criteria for selecting point:. which to moni-sponses to the bulletin and the inspection results of the 10 tor piping thickness (second item under 2 above):
plants. On the basis of these results and the efforts of NUMARC and the ASME Section XI(Inservice Inspec-piping material (e.g., chromium content) tion) Committee in terms of addressing the pipe wall-thinning issue, recommendations will be made with re-piping configuration (e.g., fittings less than 10 spect to the need of regulatory requirements for pipe pipe diameters apart) wall-thinning monitoring.
15 NUREG-1344
6
SUMMARY
OF LICENSEE'S RESPONSES TO NRC BULLETIN 87-01 6.1 Design and Fabrication Code or Responses to the bulletin indicated that limited inspec-ti ns f the single-phase feedwater and condensate sys-Standard tem were conducted in the majority of plants after the Surry Unit 2 incident. Most licensees developed their The staff review of licensee responses to the bulletin wall thinning monitoring programs for single-phase pip-showed that for domestic nuclear power plants the recon-ing because of the events at the Surry and Trojan plants.
dary system piping and components are all made of car-Some plants apparently developed programs after NRC bon steel. The material specification for straight runs of Bulletin 87-01 was issued. Out of a total of 110 units,23 pipmg is Amencan Society forTesting Material (ASTM) units still have not established an inspection program for A-106 Grade B steel, and the specification for cibows is monitoring wall thinning in single-phase lines. Of these ASTM A-234 Grade WPB carbon steel.
units,17 are operating and 6 are under construction.
The review indicated that before the ASME Code Section III rules far piping, pumps, and valves were revised in 6.3 Fre9ucncY and Method of 1971, the secondary coolant systems in nuclear power plants were designed and fabricated in accordance with Inspection the ANSI requirements of ANSI Standard B31.1, which includes 57 of alllicensed units. After 1971, safety-related The inspection frequency for pipe wall-thickness meas-portions of the secondary coolant systems were designed urementsand replacement orrepairdecisionsis basedon and fabricated in accordance with ASME Code Section a combination of predicted and measured erosion /
111 rules while non-safety-related portions of the secon-corrosion rates. In general, the pipe wall-thickness accep-dary coolant systems continued to be designed and fabri-tance criteria use measured wall thicknesses and an cro-cated in accordance with ANSI Standard B31.1, v hich in-sion/ corrosion damage rate to predict when the pipe wall cludes 43 percent of alllicensed units.
thickness will approach its code-allowable minimum wall thickness. The acceptance criteria provide guidance for determining if a piping component must be replaced or 6.2 Pipe Wall-Thinning Monitoring repaired immediately or for projecting when a piping component should be replaced at some future time.
Program For two-phase, high-energy, carbon steel piping systems, The primary method of inspection reported was ultra-the responses to the bulletin indicated that programs exist some testmg (UT), supplemented by visual examination at all plants for inspecting pipe wall thinning. Inspection and, in a few cases, by radiography. Pipe wall thickness locations are generally established in accordance with the measurements were either made by or verified by Level 11 1985 guidelines described in the Electric Power Re.
or Level W inspectors certified to the American Society search Institute (EPRI) document NP-3944, " Erosion /
of Nondestructive Testing (ASNT)TC-1 A Standard.The Corrosion in Nuclear Plant Steam Piping: Causes and In.
NRC staff considers this to be an adequate inspection spection Program Guidelines." However, because imple.
technique.
mentation of these guidelines is not required, the scope and the extent of the program vary significantly from pl nuo plant.
6.4 Affected Systems and Components In early June 1987, NUM ARC, in conjunction with EPRI, developed guidelines for inspection and repair of single.
The staff's review of licensees' responses to NRC Bulle-phase pipmg. These guidelines utilize a computer code tin 87-01 showed that wall thinning in the feedwater and that identifies and pnoriti/cs inspection k) cations on the condensate system is more prevalent in pressurized-water basis of plant-specific factors such as fluid velocity, piping reactors (PV7Rs)than boiling-water reactors (BWRs). As geometry, system temperature, and water chenastry. Ar-shown in Table 6.1,26 PWRs and 6 BWRs have identified cas that are subjected to flow disturbances, such as cl-various degrees of wall thinning in feedwater piping and bows, branch connections, and piping and fittmgs down-fittinE8' stream of control valves or flow orifices, are preferentially selected locations for inspection. By letter dated June 12, 1987, the NRC staff informed NUMARC that, with mi-The systems and components reported as experiencing nor comments, these guidelines were acceptable.
pipe wall thinning are listed below.
17 NUREG-1344
6 Responses to Bu?ctin 87-01 l
l 1
i Sinele Phase Line Two Phase Line main feedwater lines, straight runs, fittings main steam line 1
i main feedwater recirculation to condenser, straight runs, fittings turbine crossover piping feedwater pump sutaon line, straight runs, fittings turbine crossunder piping feedwater pump discharge line, straight runs, fittings extraction steam lines condensate booster pump recirculation line fittings moisture separation reheater steam generator letdown lines, straight runs, fittings feedwater heater drain piping Wall-thinning problems in single-phase piping occurred to-condenser line (minimum-flow line) in the feedwater
~
primarily in the feedwater and condensate system; the and condensate system has experienced pipe wall-problems in two-phase piping, although varied in extent, thinning degradation most frequently.The line is used to have been reported in a variety of systems in virtually all protect the pump duringlow-power operation and is iso-operating plants. Although inspection of single-phase lated by a minimum-flow valve during high-power opera-lines is not scheduled until the next refueling outage for a tion. Specific information regarding a minimum-flow number of plants, the available data from plants already line degradation incident at the I;tSalle County Station is inspected indicate a widespread problem.
provided in the bulletin to alert licensees about this The staff's review further indicated that the recirculation-problem.
Table 6.1 Plants experiencing wall thinning in the feedwater condensate system Type of Initial Plant / Unit Reactor Criticality Date Degraded Comp <ments, Fittings, or Straight Runs Dresden 2 BWR January 1970 cibows Duane Arnold BWR March 1974 elbows, reducers,8traight runs Pilgrim 1 BWR June 1972 cibows Oyster Creek 1 BWR May 1969 elbows River Bend 1 BWR October 1985 recirculation line Perry 1 BWR June 1986 straight runs Arkansas 1 PWR August 1974 elbows, drain pump discharge piping Arkansas 2 PWR December 1978 undefined Calvert Cliffs 1 PWR
. October 1974 cibows, reducers, straight runs Calvert Cliffs 2 PWR November 1976 cibows, reducers, straight runs Callaway 1 PWR October 1984 recirculation line elbows
. Diablo Canyon 1 PWR April 1984 c! bows, straight runs Diablo Canyon 2 PWR March 1978 elbows Ft. Calhoun 1 PWR August 1973 elbows, straight runs lladdam Neck PWR July 1967 recirculation line Ilarris 1 PWR October 1986 recirculation line Millstone 2 PWR October 1975 elbows, heater vent piping North Anna 1 PWR April 1978 clbows, straight runs North Anna 2 PWR June 1980 cibows, straight runs Robinson 2 PWR September 1970 recirculation lines San Onofre 1 PWR June 1967 reducers, heater drain piping
)
San Onofre 2 PWR July 1982 heater drain piping San Onofre 3 PWR August 1983 heater drain p;pmg Salem 1 PWR December 1976 recirculation ime Salem 2 PWR August 1980 recirculation line Surry 1 PWR July 1972 fittings Surry 2 PWR March 1973 fittings Sequoyah 1 PWR July 1980 cibows, straight run Sequoyah 2 PWR November 1981 cibows Trojan PWR December 1975 cibows, reducers, straight runs I
Turkey Point 3 PWR October 1972 feedwater pump suction line fittings
)
Fort St. Vrain llTGR*
January 1974 straight run m emergency feedwater line l
Rancho Seco 1 PWR September 1974 straight runs downstream of main feedwater (MFW) loop isolation valve or MFW pump miniflow valve r
'high-temperature gas reactor NUREG-1344 1F
7 ONSITE INSPECTION 7.1 Inspection Criteria Review of Licensee's Implementation of Erosion /
Corrosion Monitormg Program Ten plants were selected for inspection as part of the Inspection procedures and guidelines e
overall staff actions to address the pipe wall-thmning is-sue.The NRC inspection team included consultants from are properly reviewed and approved before the Brookhaven National 12boratory (BNL). The staff their implementation assessed how licensees are implementing their erosion /
cover periodic monitoring of high-energy corrosion momtonng program to ensure that proper 1cch-safety-related and non-safety-related carbon niques were used by qualified personnel for pipe wall-steel thickness measurements and to ensure that adequate guidance was provided for corrective actions and other ac-provide for qualification or certification of per-sonnel and equipment tivities regarding repair and replacement of degrading are consistent with commitments pipmg and components. Ihrough selective examination The equipment used to perform NDE has been cali-of each licensee's program for monitoring pipe wall thin-ning, the procedures and administrative controls defining biated against krown standards for types of metals the activities to be accomplished 'cre verified for consis-as i range of thickness to be measure 1.
l tency with the licensee's progrwa commitment. The Pipe wall thickness is being measured in accordance l
stafft criteria to evaluate the licensees' erosion / corrosion e
monitoring programs and their implementation are with established instructions and results are being j
briefly described below, appropriately documented.
Qualified personnel are evaluating pipe wall meas-e Review of Licensee's Erosion / Corrosion Monitoring urements to determine the need for corrective ac-Program tion and the frequency of continued periodic moni-e The licensee has developed an crosion/
10fI"8-corrosion monitoring program.
A schedule has been established to repair and con-tinually monitor piping that has shown evidence of The licensce's program has well-defined criteria for wall thinning.
selecting inspection points Administrative controls are in place and manage-e determining inspection frequency ment support is evident.
defining method of inspection The licensee's commitments in response to NRC making replacement / repair decisions e
BuHedn M am Mg met.
e The licensee's program meets the intent of NUMARC guidelines.
7.2 Summary of Inspection Results The licensee's program includes e
7.2.1 Erosion / Corrosion Monitoring high-energy single-phase lines, including long-Program Most licensees in the 10-plant inspection developed their two-phase lines, includmg guidelines and com-initial erosion / corrosion monitoring program for two-puter codes phase lines in 1982 after the 24-inch pipe rupture (extrac-large moderate-energy single-phase piping sys-tion steam line)at the Oconec Unit 2 plant. A few of them tems later expanded their program to include high-energy The licensee has established a plant-specific history single-phase piping after the failure of a heater drain dis-charge pipe at the Trojan Plant in 1985. After the feed-of pipe wall thinning. including failure analysis and w ter line rupture at Surry Unit 2 in 1986, a majority of damage mechanism.
licensecs again expanded their program to include large
""E "E
"8
'Ihe licensee has a well-developed training program e
and personnel conducting NDE examinations have Most licensees established their owa crosion/ corrosion been properly certified.
multidisciplinary task force in early 1987 shortly after the 19 NUREG-1344
7 Onsite Inspection Surry incident to address the issue of pipe wall thinning.
Grade P22). or other low-alloy steel was used for replace-
'Ihc task force's objectives were to ment piping.
develop a predictive method to select inspection lo-In some instances, licensees have chosen to weld overlay a e
cations on the basis of pipe configurations, materi-thinned area of piping so that it exceeds the minimum als, velocitics, and water chemistry wall thickness requirements of the piping system. In cases where a pipe is repaired by weld overlay, the pipe wall inspect cach unit using various NDI! methods at thickness is monitored closely to ensure the pipc integrity, e
points established by system modeling otherwise the overlay will only stay in place until the next outage at which time it will be replaced with pipe of the develop a baseline from the m. formation gathered same specification or of a more resistant low-alloy steel.
e during inspections and use it for trending purpose All replacement pipe welds were welded without backing expand the overall program to include the utility's rings. the root passes for these replacement welds gener-e fossil units ally were made by the gas tungsten arc weld (GTAW) test-
'Ihe task force also established an action plan to inspect ing method while the subsequent passes were made with piping of 1 inch and larger with a system temperature the shicided metal are weld (SM AW) testing techniques.
greater than 100 C. In general, piping over 8 inches in di-ameter was given priority. This action plan covers main 7.2.3 Inspector Qualification and Training steam, extraction steam, heater dens, condensate, feed-Program water, and reheat systems.
Majority of the licensees in the 10-plant inspection used The locations for inspection were selected on the basis of outside contractors to perform the ultrasonic testing in-NUM ARC guidelines or ca:culated flow rates, piping ge-spection of the secondary piping systems. Only a few li-ometries, and past experience. The EPRI "ClIEC" com-censees used their own NDLi personnel for the pipe wall-puter code was used in most cases to determine the most thickness measurements. However, in both cases, the susceptible areas to crosion/ corrosion damage. However, inspectors who conducted the crosion/ corrosion examina-most licensees indicated that the results were more con-tion were certified to the ASNTTC-1 A Standard, sistent when the selections were made on the basis of op-erating conditions and engineeringjudgment rather than 7.2.4 Overall Program Assessment on the basis of the compi er code alone.
The 10 licensees in the inspection have developed and I,or small bore piping systems, some h.censees are usmg a have in place an crosion/ corrosion monitoring program through-insulation radiography technique for detecting that meets the intent of the NUMARC guidelines for wall thinning. When wall thinning was identified by radi-crosion/ corrosion monitoring in single-phase lines. In ad-ography, insulation was removed and actual depth of wall dition, all licensees have completed their initial inspec-thinning was determined by ultrasonic (manual) mcasure' tion on the feedwater and condensate system. Although ment. In actual practice, however, some licensees have these inspections were carried out by qualified and certi-chosen to replace all piping 8 inches in diameter and less fied NDl! inspectors, none of these licensees have for-rather than measuring the actual thickness if crosion/
malized their procedures and administrative controls to corrosion thinning is detected.
implement their erosion / corrosion monitoring programs.
7.2.2 Corrective Actions and Repair /
7.3 Plant-Specific Inspection Findings Replacement Critena All licensees in the 10-plant inspection either implictly or explicitly have adopted an acceptance criteria for making The mitial program for erosion / corrosion was instituted repair / replacement decisions consistent with the in 1982 after the 24-inch pipe rupture of an extraction NUMARC guidelines for crosion/ corrosion in single-steam line at Oconec Unit 2.This program was expanded phase lines.
in 1985 after the heater drain discharge pipe failure at the Trojan plant. A third expansion of the program was made in general, the licensees have been replacmg all piping after the Sarry incident in 1986. This modification was that shows a significant amount of wall thinning. Most re-made to include large moderate energy piping systems placement piping has been made of the same cat bon steel with single phase flow conditions. The program has since (ASTM A-106 Grade H and ASTM A-234 Grade WPH) evolved to include small-diameter high-and moderate-or the ASME SA equivalents. Ilowever, depending on energy piping systems as well as the condensate system availability, in some cases 2-1/4 Cr-1 Mo steel (SA 335 piping.
NUREG-1344 20
7 Onsite Inspection
'Ihe following systems of both units were examined for current time frame for an overall implementing proce-wall thinning:
dure to be written and ap[ roved is the summer of 1989.
feedwater and condensate system l'lant No. 2 (l'WR) Inspection Date: June 7-8,1988 heater-drain pump discharge piping
.g ;;
,s program of crosion/ corrosion monitoring extraction steam pipmI was initiated in the spring of 1985, with early detection of turbine crossunder pipmg two-phase (wet steam)line degradation being the primary small piping downsteam of steam trap concern. During the fall of 1985, the licensee began its in-spection of steam piping in accordance with the guide-To date, pipe wall thinning has been discovered in the lines given in EPRI Report NP-3944. After the 1986 heater-drain discharge piping, the feedwater and conden-Surry pipe rupture event, the licensee organized a multi-sate piping, and a 'wnstream of steam traps on lines com-disciplinary task force to expand the original pipe wall-ing off the m.a steam lines of Unit 1. Only a small thinning monitoring program and to develop an action amount of wall thinning was discovered in the piping of plan for implementing the NUMARC recommendations the feedwater and condensate system. Failure analyses of for monitoring crosion/ corrosion in single-phase lines. In damaged piping or components were conducted routinely addition, the licensee also established a program of accel-by the licensee.The inspection team reviewed a typical m-erated monitoring of heat drains and vents.
vestigation report during the visit; it appeared to have been done in a thorough and professional manner.
To date, the licensee has replaced pipe in various areas of the plant for two-phase crosion/ corrosion but has not ex-Ultrasonic testing (UT) and radiographic testing (RT) perienced any incidents of single-phase crosion/corro-were used for wall-thickness measurements.The licensee sion. A section of piping downstream from a throttle valve contracts with outside vendors to do UT inspection as well (entering the steam generator blowdown flash tank) was as the RT work.The inspection team reviewed NDE cer' replaced recently with stainless steel pipe. In another tifications of vendor personnel who have conducted the case, a main feedwater pump cibow (90* minimum flow wall-thickness measurements and found their qualifica-line) was replaced whh a P-11 cibow, which was found tions satisfactory.
leaking after only 6 months in service. The licensee is monitoring this elbow by frequent inspections and will The licensee is replacing all piping that shows a significant conduct a failure analysis after replacement at the next amount of wall thinning. As of the last refueling ouuge refueling outage.
before the NRC inspection, all replacement piping was of the 2-1/4 percent chromium,1 percent molybdenum vari-The licensee has looked at all but one of the potential ar-ety of steel (SA 335, Grade P22). In some instances, the cas of erosion / corrosion. It is currently evaluating the fea-licensee may opt to weld overlay to a thinned area of pip-sibility of inspecting the auxiliary feedwater system and ing so that it exceeds the minimum wall thickness require-the main steam drain lines. Ia addition, the licensee plans ments of the piping system. In such cases, the repaired to inspect the safety-related portion of the feedwater sys-piping will stay in place only until the next outage at tem as well as the blowdown and essential service water which time it will be replaced with SA 335, Grade P22 systems at the next refueling outage.
material. In addition, all replacement pipe welds were l
done without the use of backing rings. The staff deter-The staff found that the licensee normally uses UT tech-mined that this new welding procedure will reduce signifi-niques for its inspection and it has very elaborate and pre-cantly the propensity of crosion/ corrosion-induced pipe cise procedures for the layout of inspection grids for UT.
wall thinning.
When UT did not provide meaningful results, the licen-see's practice was to use RT. Only Level II inspectors cer-The staff found the licensee's crosion/ corrosion monitor-tified to ASNT Standard TC-1 A, are used in the pipe ing program meets the intent of NUMARC guidelines.
wall-thinning inspection program. Records review and The licensce's program is above industry standards. Spe-pers(mnel interviews indicated that the licensee has es-cifically, replacing carbon steel pipe with chromium-tablished an adequate qualification and training program molybdenum steels without backing rings in welds is a for its NDE inspectors.
positive step towards reducing the pipe wall thinning problem.
The licensee's program is above the industry standards.
There was excellent assignment of responsibilities and All of the appropriate comrols appear to be in place at concise, explicit procedures approved and issued for con-this site.The only potential problem with the program is ducting pipe wall-thickness measurement, and there was the lad of formalization with implementing procedures.
evidence of good management support and well-defined The utility is currently using a series of memorar.da to organizational responsibilities. Ilowever, because of document the practices to be used. As of this date, the intrinsic difficulties with UT inspections through weld 21 NUREG-1344
7 Onsite Inspection overlays, the licensee should prepare UT standards to be program, involving various engineering and quality assur-used for various thin-wall /overlayed conditions.
ance groups,is found to meet the intent of the NUMARC guidelines. 'the staff recomrnended that the licensee Plant No. 3 (llW10 - Inspection Date: July 6-8,1988 should examine the safety-related portion of the feed-water lines for potential crosion/ corrosion problems as The licensee's initial interest in the crosion/ corrosion soon as possible and should formalize its implementation phenomenon developed in 1978 when a small crack was procedures and administrative controls to ensure that its discovered in a feedwater discharge pipe downstream of a p pc wall-thinning inspection program will be conducted reactor feedwater pump. Subsequent UT showed that just like any other inservice inspection (ISI) program.
wall thinning had occurred on all three feedwater pumps, and three feedwater pipe reducers were replaced. In late I'lant No. 4 (PWit) Inspection Date: July 12-14,1988 1978, a 6-inch pipe downstream of a reheater drain tank
,Ihe crosion/ corrosion monitoring program initially flow control valve was inspected and wall thinning was de-evolved in 1978 as an informal smgle-pomt-tected. During 1983 and 1984, a total of 40 locations in per<omponent production plant maintenance monitor-various parts of the main steam, extraction steam, and ing program. 'l,he second phase started during the Unit 2 feedwater piping systems were inspected. As a result of refueling outage m 1984. Crossunder and extraction this inspection, three feedwater expansion elbows, lax 16 steam piping was selected for inspection; the results inches, were replaced. Although a total of 46 locations showed an active crosion/ corrosion mecham,sm m the sys-were identified for inspection in 1986, only five main tem. The inspection program then was expanded for two-steam line drains were inspected and replaced because of phase pipmg. Heater dram tank pump discharge piping schedular priorities. During the 1987 outage,43 locations and feedwater pump discharge pipmg were added in 1985 were inspected; six components were found to require as the result of INPO SER 23-85. A sample of large bore reinspection within the next three outages. The staff cooling water piping was added in 1985 after a leak was found that, to date, the licensec has not inspected the experienced near a cooling water pump discharge piping safety-related portion of the feedwater lines. However, weld. The licensee s current program covers turbmc the licensee indicated that this will be done during the crossunder piping, high-pressure extraction piping, feed-next refueling outage.
water pump discharge pipmg, and cooling water piping.
The licensee has found areas of pipe wall thinning.
The licensee primarily uses UT techniques in monitoring pipe wall thinning. The licensee's inspections were con-The licensee has replaced two sections of piping as a re-ducted by its own N DE personnel and those of its contrac-sult of the thickness sun'ey program. However, in both tors. The NRC inspection team reviewed the licensee s cases the licensee determined that the piping removal was quah,neation and training program for the NDE person-necessary because of lamination and not because of nel responsible for conductmg pipe wall-thmnmg exami-erosion / corrosion-induced wall thinning. No piping sam-nation. The results showed that only inspectors (Level 11 plcs or written failure analysis reporis were available and lil) certified by ASNT Standard TC-1 A were used on for review.
the inspections.
The inspections were performed by using various forms of To date, the licensec has replaced all degraded compo-NDE such as UT, RT, and visual examination. However, nents with original carbon steci replacements. However, UT remains the primary inspection method. Radiography the licensee is making a concerted effort to locate and has been attempted with limited results. A procedurc l
procure chromium-molybdenurr. materials for future out-would have to be developed to qualify the use of RT to age replacements. Although there is no written procc-determme actual wall thickness of damaged pipe sections.
dures that preclude the use of backing rings in replac" Inspection areas were selected by reviewing results from ment situations, the welding supervisor could recall no computer code, the findings of an independent survey by instances where eroded / corroded pipe or components an outside vendor, and the requests of the Plant Exten-were replaced using backing rings.
sions Group. Forty-five areas were scheduled for inspec-tion. Single-and two-phase systems greater than 4-inch The licensee also took initiative to develop its own com-diameter with a design temperature cer 200 F are pres-puter program for selectmg inspection locations. The ently included in the inspection scope.
NRC inspection team found that the licensce's computer program was developed by basing the Keller equations Outside contractors were used to conduct UT inspec-with modifications as well as considering the parameters tions. All inspectors were certified in accordance with recommended in the NUM ARC guidelines for monitor-ASNT Standard TC-1 A requirements. In addition, data ing crosion/ corrosion in single phase lines.
interpretation and evaluation were done by ASNT-certified Level 11 or Level 111 inspectors. However, spe-The staff finds that the licensee has a rnultidiscipline ap-cialized training is required for personnel using Ultra Im-proach to address the pipe wall-thinning problem. The age til UT equipment. The NRC inspection team also NUREG-1344 22
7 Onsite inspection reviewed the licensee's training program and certification analysis reports were available for the NRC inspection records of selected NDE personnel and found them satis-team to review.
factoq.
The licensce's crosion/ corrosion monitoring program is The licensce's corrective actions and repair / replacement governed by three memoranda instead of a formalized criteria for pipe wall thinning is based on the NUM ARC well-integrated master inspection program plan. These guidelines for single-phase lines. '1he responsible plant memoranda establish a 2-inch grid pattern for UT meas-system engineer evaluated the inspection results and de.
urements. To date, no permanent reference points have termined if a repair / replacement was required. Replace.
been marked on the pipes, only point stick grids have ment piping materials are currently planned as like kind been drawn. The licensee's selection of inspection loca-replacement (i.e., carbon steel A-106 Grade 11 replacing tions was primarily on the basis of engineering judgment the original carbon steel A-106 Grade 11 material).
and operating experience. Use of the EPRI CliEC com-puter code for inspection point selection was not encour.
The staff finds that the licensee's crosion/ corrosion moni.
aging.
toring program meets the intent of the NUMARC guide-The licensee uses qualified (ASNT Standard TC-1A, lines and is above the industry standards. Specifically, the Irvel II) personnel to perform UT inspection. Certifica-licensee took initiative to evaluate the advantages of se.
tion records and training program for NDE personnel lecting inspection locations on the basis of using the com.
were reviewed and found satisfactory.
puter code or on the basis of operating experience and en-gineering judgment alone. The licensee concluded that The staff finds the utilitJ's program meets the NUM ARC the benefit gained by use of the computer code does not guidelines for m.spection of erosion / corrosion m smgle-justify the resources and efforts spent on providing data phase lines. However, the lack of coherent corporate (for-input for the computer code. Similar conclusions by other malized) program and associated implementing procc-licensees also were reported to the staff.The appropriate dures is a potential problem area.
controls appear to be adequate to meet program require-ments. Ilowever, a potential problem with the program is I'lant No. 6 (Il%R). Inspecuon Date: August 9-11,1988 the lack of formalization of the program implementing procedures and the level of responsibilities assigned to An crosion/ corrosion inspection program at both units the plant systems engineer. Approval of procedures cur-was started in 1982 in re sponse to industry reports of wall-rently in a draft / review status and a shift of material thinning problems occurring in the cross-around steam repair / replacement responsibilities to the Materials &
piping. The extraction steam piping was examined (both Special Processes Group should go far to provide formal units) in 1985 and heater drain piping was examined at program implementation.
Unit 2 in 1986. In the fall of 1987, the licensee performed a baseline inspection at Unit 1. A similar inspection was I'lant No. 5 (I'MR). Inspection Date: July 26-28,1988 performed at Unit 2 in the spring of 1988.There were in-stances of crosion/ corrosion problems in single-phase The licensee initiated its original crosion/ corrosion moni-lines at both units although more so at Unit 2.
toring program in 1983/1984 outage. However, the pro-gram was confined to two-phase lines only. The main The licensee used three contractors during the fall 1987 steam reheat crossunder piping was inspected and nu-inspection to determine which arcas of the plant were the merous components were either repaired by weld buildup most likely candidates for wall thinning. A total of 36 or replaced with new components of the same specifica-single-phare h> cations were chosen from the feedwater, tion.
condensate, condensate booster, and heat er drains. In ad-dition,21 t wo. phase k> cations were chosen, as well as four After the Surry incident in December 1986, the licensee additional elbows, in response to leaks found in the reac-inspected 70 fittings from systems, such as the condensate tor feed pump minimum flow lines of another licensee's bypass, condensate and feedwater, and the heater drain plant. Of the 61 locations inspected,11 had UT grid re-pump discharge recirculation, of one unit. Three loca-duction performed; these all were found to be acceptable tions were found to be below the minimum wall thickness after engineering evaluation.
of the piping material specification; these were all re-placed. During the 1987 inspection,88 fittings in the sec-Durmg the spring 1988 outage, a pipe wall-thinning in-ondary systems of the second unit were inspected. Of the spection was performed at Unit 2. Of the 46 locations (20 30 hications identified with a wall thickness below the ac-single-phase and 26 two-phase) inspected,41 were found ceptance criteria,3 h> cations were repaired by weld over-to be acceptable (either on initial inspection or after engi-lay,13 locations were replaced, and 14 locations were neering evaluation); in 2 locations nozzles were repaired; evaluated to be adequate for continued service. To date, and in 3 locations, 6-inch diameter pipes were re-the licensee has not identified any cases of single-phase placed. UT methods were used in these inspections, crosion/ corrosion. During the inspection, no failure So far, no radiography has been used for pipe 23 NUREG-1344
7 Onsite inspection wall-thinning inspection. Outside contractors were used Sixty-nine locations of the Unit 2 feedwater and conden-for these inspections and all evaluation of data was per-sate system piping also were examined in the 1986/1987 formed by ASNT-certified Level II or Level 111 inspec-timeframe. This included 10 locations recommended by tors. The NRC inspection team reviewed certification computer code. The steam extraction piping for Unit 3 documentation and the inspector training program and was re-examined in 1987. Although some evidence of wall found them to be adequate.
thinning was observed, all of the locations exceeded the minimum design requirements for wall thickness. Also in No repairs or replacements were necessary following the 1987,102 locations of the Unit 1 feedwater and conden-Unit 1 inspection. Three 6-inch pipe replacements were sate system piping were examined: 3 areas were found to necessary following the Unit 2 inspection. The damaged have readings below manufacturer's tolerance but no ar-P-1 material (carbon steel) was replaced with P-5 (5 per cas exceeded the replacement limit. In a 1988 inspection cent chronium) material making the line entirely P-5.
at Unit 3,98 locations were inspected. Two expanders The inspection team reviewed the work package for the were replaced with chromium-molybdendum material replacement work, welder performance qualification re-because of significant wall thinning and two other fittings cords, the NDE inspector's certification records, and the were replaced with chromium-molybdendum fittings be-training program; all were found to be satisfactory.
cause of experience gained at Unit 2.
UT is the riormal method of inspection for crosion/
The staff finds the licensee's overall crosion/corros. ton corrosion pipe wall thinning. Both A-scan (CRT display) program meets the intent of the NUM ARC guidelines for and ultra image equipment were used during examina-smgle-phase imes. All appropriate controls for an effec-tion. Grid patterns were applied by " paint stick" to facili-tive pipe wall-thinning program were in place at both tate examinations, units under one program manager.The ability of the h-censee to draw from experience at its other sites is a defi-Although the licensee has implemented the EPRI I
nite advtmtage. However, overall administrative proce-
"CHEC" computer program for determination of the dures for pipe wall thinning need to be written. Currently, most susceptible areas for pipe wall thinning, no signifi-there is not a procedure available to control this program.
cant wall thinning had occurred at these chosen locations.
l The utility stated that such a procedure would be written In contrast, crosion/ corrosion-induced wall thinning had j
and in effect before the June 1989 outages.
occurred at various other k> cations, predominantly at ex-panders downstream of control valves. To date, there has I'lant No. 7 (PWR). Inspection Date: August 30-31,1988 been no instance of significant crosion/ corrosion in safety-related feedwater lines.
The licensee's crosion/ corrosion experience dates back to the extraction steam pipe inspections at Unit 1 in 1978. In The licensee used outside contractors for t'e baseline in-1980,70 fittings of the extraction steam lines were again spections at all three units.The NRC inspution team re-i 1
inspected at Unit I without instance of pipe wall thinning.
viewed the licensee's qualification program and certifica-A similar inspection was conducted at Units 2 and 3 in tion documents of selected NDE personnel. In addition, 1985. Seventy-six h> cations in each unit were examint;l certification and calibration documentation for test and all locations were found to be above minimal wall equipment and test bk)cks were reviewed. The results thickness. A 1986 insp;ction of Unit 1 steam extraction were all satisfactory. However, the licensee has no train-piping disclosed an elbow with a wall-thinning problem.
ing or retraining program for the NDE personnel.
Eight other elbows also were examined and they were found to be satisfactory. In addition,12 main steam (tees Concerning repair / replacement criteria, the licensee at-and cibows) and 5 feedwater elbows were examined on tempts to repla e the ongmal carbon steel components with more corrosion-resistant materials (e.g., stainless Unit 1 in 1986. All fittings were within the manufacturer's tolerance for wall thickness.
steel or chromium-molyodenum alk>ys). In addition, backing rings have not been used in welding or repairs at the site since the construction phase was completed.
In the 1986 inspection of Unit 2,93 areas on the steam cxtraction piping were examined; this included the 76 lo-The staff finds that the licensee's erosion / corrosion moni-cations previously examined in the 1985 baseline inspec-toring program meets the intent of the NUMARC guide-tions. No significant wall thinning was observed.
lines and is above industry standards.The replacement of carbon steel piping with more corrosion-resistant materi-The Unit 3 feedwater and condensate piping was exam-als is a definite step toward mitigating pipe wailthin-ined in the 1986/1987 timeframe as a result of the Surry ning caused by erosion / corrosion. Review and root incident. Thirty-six locations were examined and six cause evaluations are done by a company metallurgist, reducers replaced. Two red ucers were replaced with Type which also is a positive attribute of the licensec's pro-304 stainless steel and four reducers with a 2-1/4 percent gram. These two items plus the utility's large inspection chromium,1 percent molybdenum alloy.
populatior, definitely show that appropriate controls are NUREG-1344 24
7 Onsite Inspection in place at the site.The only area lacking completeness in aggressive program of component inspection with good the program is absence of an overall administrative proce-management.
dure that would delineate the responsibilities of the sari-ous groups involved with the pipe wall thinning monitor.
Although a great many inspections have been performed for crof on/ corrosion pipe wall thinning, the repairs have ing program.
been pcrformed with a minimum of procedures.The util-I'lant No. 8 (PWit). Inspection Date: August 24-25,1988 ity tercJs to assign a cognizant engineer to a project, who in turn " runs the show" for his/her area of responsibility.
The licensee performed its first pipe wall-thinning in-Previous record keeping (pre-1987 outage) shows a spection in 1981 on the extraction steam lines. Thirteen somewhat haphazard approach to documenting crosion/
elbows and one " tee" connection were examined. Three corrosion inspections. A more formalized program with elbows were replaced and one unow was repaired and appropriate procedures would tend to firm up and fill out eventually replaced. In 1982,56 elbows from the extrac-an otherwise good program.
tion steam lines and 4 cibows from the main steam line were examined. Six cibows were repaired and three el-Plant No. 9 (llWR)- Inspection Date: September 6-9,1988 bows were replaced because of wall thinning. In 1984,39
.lhe licensee began its development of a crosion/
elbows and 2 drains of the extraction imes were exammed; corrosion monitoring program in 1985 as a result of NRC 10 cibows were replaced and 2 locations on the 5th-point IE Information Notice 82-22, INPO SER 41-82, and extraction dram lines were temporarily patched. Sixty-INPO Significant Operating Experience Report 82-11.
nine elbows were examined in 1985.The systems covered 1hese documents address erosion and crosion/ corrosion extraction dram lines, the high-pressure turbme skim of steam piping at several nuclear power plants. Phase 11 tank vent, feedwater heater steam vents, and heater g7 g g
,s program evolved as action response to drains. Pipe wall thinning occt.rred m, all systems with NRC Bulletin 87-01. Thirty additional components were eight elbows replaced and one repaired.'lhe 1987 inspec-selected for inspection bringing the total number of com-tion encompassed: 149 elbows,9 tees,14 dram locations, ponents to thirty-eight. These components will be in-10 vent locations, and 2 pipe-to-nozzle connections. The spected at every reactor refueling outage (RFO) with ex-systems inspected mcluded feedwater lines and many pansion of the crosion/ corrosion monitoring program other two-phase lines. This inspection required 55 cibows determined on the basis of field inspection results.
to be replaced and 1 vent location plus the entire second-point vent to be replaced. To date, formal failure analyses in June 1987, an 18-inch condenser drain leak was investi-were not performed on damaged components. However, gated. The condenser was in service since 1982. Steam durmg the plant audit, sigmfictmtly thmned sections from crosion was determined to be the cause of the leak.The small bore dram imes, the heater drain system, and the line was repaired using a Pil ASME S A-335 insert. Im-high pressure moisture preseparator drain system were pingement of water droplets was determined to be the shown to the NRC inspection team. Visual exammation root cause of the material degradation. Ilowever, no for-of these components by the NRC team confirmed the h mal failure analysis has been reported.
censee's proposed erosion / corrosion failure mechanism To date the licensee has used digital ultrasonic thickness The licensee uses UT techniques to determine pipe wall measurement equipment (D-meter) and visual examina-thinning, where feasible, and supplements this by visual tion for pipe wall-thinning inspection. Automated ultra-examination. During the next refueling outage, the licen-sonic equipment has not been used because of the short see plans to inspect 283 locations (e.g., pipe, fitting, el-operating time of Unit 1.The next scheduled inspection bow, and tec). Inspections were done by outside contrac-will be performed at RFO 3, March 1989.
tors and only ASNT-certified. Level 11 inspectors were used. The NRC inspection team reviewed certification The staff finds that the licensee's crosion/ corrosion moni-documents of inspectors and calibration records for toring program meets the NUM ARC guidelines. The ap-equipment and instruments used for wall-thickness meas-propriate engineering and plant operational controls are urements and found them to be satisfactory.
in place and provide adequate instruction, response, and corrective actions when required. The inspection pro-The' licensee is replacing components that show a signifi-gram engineering commitments are good and provide op-ctmt amount of wall thinning. Although some replace-erational restraints when program requirements are ex-ments have been "in kind"(carbon steel for carbon steel),
ceeded. However, administrative / management interface the licensee is attempting to replace croded/ corroded for the erosion / corrosion program does not appear to be components with P-22 materials (2-1/4 Cr-1 Mo).
clearly defined either by procedures or organizational flow charts. The licensee is in the process of incorpo-The staff finds that the licensee's current program meets rating the erosion / corrosion inspection program the intent of NUMARC guidelines for crosion/ corrosion requirements into a plant specification or procedure, monitoring in single-phase lines. The licensee has an which would provide the administrative / management 25 NUREG-1344
7 Onsite Inspection organistional interface that the program is currently accordance with ASNT Standard TC-1 A. All evaluation weak in.
of data'is performed by ASNT-certified Level 11 or Level Ill inspectors. The licensee reviews and approves all con-Plant No.10(PWH) Inspection Date:
tractor procedures and personnel qualifications. How-September 28-29,1988 ever, the licensee currently does not have in-house proce-The licensee did its first pipe wall-thinning inspection in dures or equipment to perform NDE examinations or 1983. Two-phase lines at 36 locations were inspected, verify the contractors' results.
which resulted in replacement of only one reducer. In 1985, the licensee inspected another 21 locations on two-phase lines and no significant wall thinning was discov-The licensee generally will replace all pipm.g that shows a cred. In 1987,69 locations on single-phase lines were ex-significant amount of wall thinning. Repair / replacement to date has been with carbon steel and 2-1/4 Cr-1 Mo ma-amined.
Four locations showed wall-thickness discrepancies. Two components are currently being terials. The licensee is reviewing design changes and the redesigned, another one was replaced, and the fourth one use of stainless steci piping materials for future repair / -
was found acceptable to the next refueling outage. The replacement. All nonconforming inspections for safety-licensee has not completed any failure analysis on the de.
related and non-safety-related systems are documented graded corpponents.
on nonconformance reports and transmitted to the Nu-clear Engineering Department for resolution.
Currently, selection of inspection points and sample ex-pansion are based on criteria from the NUMARC guide-lines and the llechtel Power Corporation "WNDiEK" The staff finds that the licensee's crosion/ corrosion mori-computer code. Prioritization of sampling for single.
toring program meets the intent of N U M ARC guidelines.
phase systems is determined using onsite crosion/
However, inspections have shown that there is a definite corrosion events, industry experience, and consideration need to improve record keeping (reproducibility of wall-of factors that affect the rate of erosion / corrosion.
thickness inspection data). Appropriate implementing procedures and positive program commitments also The licensec uses outside contractors for all N D E-related would fill out an otherwise good erosion / corrosion l
activities. The contractors' personnel are qualified in program.
1 N UREG-1344 26
8 CONCLUSION l
As this review has shown, erosion / corrosion is a complex were completed at all 113 plants at the end of October phenomenon and its rate can be affected by factors such 1988. Results of the 10 plant audits completed by the l
as piping material, geometry and hydrodynamic condi-NRC indicate that all licensees have conducted initial tions, and operating conditions of the secondary systems crosion/ corrosion monitoring inspections and the pro-such as temperature, pH, and dissolved oxygen content, grams meet the intent of the NUMARC guidelines for The problem is widespread for both single-phase and two-erosion /corrosionmonitoringinsingle-phaselines. A few phase high-energy carbon steel piping systems. Although of the 10 licensees are in the process of formalizing proce-the Surry incident was the first time for such a cata-dures or administrative controls to implement their iong-strophic failure occurring in a nuclear power plant, it was term programs; however, alllicensees have not commit-no worse than previous failures in large steam turbine ted to these procedures or controls for a long-term pro-reheat lines or in feedwater lines. The failure mechanism gram. In many instances, licensees had not specified re-was different to some extent because the Surry incident cord keeping during previous inspections to allow for the was caused by crosion/ corrosion-induced pipe rupture future reproducibility and trending. There aise were in-whereas the wet steam pipe failures were caused by a stances where work was done by outside contractors usmg cavitational type of erosion damage.
the contractor's guidelines rather than the licensee's Many nuclear utilities initiated inspection programs to monitor erosion / corrosion on their own initiative shortly Therefore, in the absence of formalized procedures and after the Surry incident. However, the extent of the in-administrative controls, there is not adequate assurance spection programs varied until NUMARC and EPRI de-that licensees will continue to meet their licensing basis veloped uniform guidelines for inspection, repair, and re.
by maintaining the structuralintegrity of high-energy car-placement of piping and components degraded by bon steel piping systems. In addition, codifying pipe wall-erosion / corrosion attack.These guidelines were endorsed thinning examinations is not an easy task. Past experience i
by the NRC in June 1987, and theyform the basis for most shows that it will take several years before the final code erosion / corrosion monitoring programs developed by requirements regarding pipe wall-thinning inspection are utilities. In addition, as requested by the NRC, the ASME formalized and implemented. Consequently, the staff Section XI Committee is developing a new requirement recommends, as an interim step to ASME imposing its for pipe wall-thinning inspection. The new requirement code requirements, that the NRC (through a generic let-will cover both Class 1 and 2 piping of the safety-related ter) require licensees to formalize their procedures and portion of the feedwater lines.
administrative controis to ensure that the NUMARC program or another equally effective program is imple-Recent utility reports to NUMARC have indicated that mented so that the integrity of all high-energy piping sys-inspections using the new guidelines or their equivalent tems is maintained.
1 i
l l
27 NUREG-1344
9 REFERENCES American National Standards Institute (ANSI), Standard
--, EPRI NP-3944, " Erosion / Corrosion in Nuclear B16.5," Flanges and Flange Fittings," 1981.
Plant Steam Piping: Causes and Inspection Program Guidelines," 1985.
--, Standard B31.1, " Power Piping," 1967.
Iluijbregts, W. M. M., Erosion-Corrosion of Carbon Steel
'* "C#'
'EE
--, Standard B31.7, "Nuctcar Power Piping," 1969.
American Society of Mechanical Engineers (ASME)
Institute of Nuclear Power Operations (INPO), Licensee Boiler and Pressure Vessel Code, Section 111, " Rules for Event Report (LER) 305-85006 " Steam Generator Tube Construction of Nuclear Power Plant Components,"
incorrectly Plugged," March 15,1985.
1977.
--, l.ER 305-G5017, " Reactor Tripped Manually to
--, Code Case N-7, "High Yield Strength Steel,"
Isolate Rupture Excess Steam Vent Line," September 6, 1975.
1985.
-~, l.ER 366-86010, " Relay Actuated Tripping of
--,Section IV, " Nuclear Components," 1971.
Mam.l'urbme Which Resulted in Reactor Scram,,,
--,Section XI, " Rules for Inservice Inspection of Nuclear Power Plant Components," 1986.
--, LER 244-86004,"ReactorTrips Manually When 12rge Steam Leak Discovered in Turbine Building,"
American Society of Nondestructive Testing (ASNT),
August 2t,1986.
" Recommended Practice SNT-TC-1 A," August 1984.
--, Significant Event Report (SER)41-82," Erosion Bignold, G. J., et al Erosion-Corrosion in Nuclear Steam of Steam Piping and Resulting Failure," July 15,1982.
Generator, British Nuclear Engineering Society, pp. 5-18,
--, SER 23-85, " Water Pipe Wall Erosion Down-1980, London, England.
,, May 16,1985.
Bouchaourt, M., " Flow Assisted Corrosion in Power
--, Significant Operating Experience Report 82-11.
Plants," International Atomic Energy Agency (IAEA)
" Erosion of Steam Piping and Resulting Failure,"
Specialist Meeting on Erosion / Corrosion, Septem-November 17,1982.
ber 12-14,1988, Vienna, Austria.
Kastner, W., H. llenzel, and B. Stellwag. " Erosion Corro-Electric Power Research Institute (EPRI), Users Manual sion in Power Plants." The Third International Topical NSAC-112.
"CH EC" (Chexal-Horiwitz Erosion.
Meeting on Nuclear Power Plant Thermal Hydraulics and Corrosion). June 1987.
Operations, November 14-17,1988, Seoul, Korea.
29 NUREG-1344
l i
NUMARC TECHNICAL SUBCOMMITTEE WORKING GROUP ON PIPING EROSION / CORROSION
SUMMARY
REPORT June 11, 1987 NUREG-1344 Appendix A 1
________-__-_--________-__a
TABLE OF CONTENTS SECTION PAGE EXECUTIVE
SUMMARY
.1.
BACKGROUND 1-1 2.
GUIDANCE 2-l' 3.
COMMITMENTS 3-1 APPENDIX A: NUMARC WORKING GROUP MISSION APPENDIX B: GRID LAYOUT SKETCHES i
1 l
NUREG-1344 iii Appendix A
EXECUTIVE
SUMMARY
The need exists to establish industry initiatives to identify potential evidence of single phase erosion / corrosion and thus help ensure personnel safety and minimize unnecessary plant challenges resulting from potential further failures.
It is paramount that timely and appropriate action be taken.
Schedules for completion of these industry efforts should be aggressive for those units where the potential exists indicative of unacceptable wall thinning.
For other units where no indication exists, analysis and inspections are necessary but may be performed on a schedule commensurate with normal 1
refueling cycles.
Simply stated, the recommendations of the NUMARC Working Group are threefold.
They are 1) conduct appropriate analysis and a limited but thorough baseline inspection program, 2) determine the extent of thinning, if any, and repair / replace components as necessary, and 3) perform follow-up inspections to l
confirm or quantify thinning and take longer term corrective actions (i.e.,
adjust chemistry, operating parameters, or others), as appropriate.
To assist in accomplishing these efforts, the Working Group consistent with their stated mission is providing guidance to industry in the following areas.
SAMPLE SELECTION
- Initial sample - 10 most' susceptible locations and 5 additional locations SAMPLE EIPANSICN
- For each component below acceptance guidelines, the inspection shall be expanded to similar fittings / components based on engineering judgement and appropriately documented.
INSPECTION
- Technique selected by the inspecting utility using recommendations provided ACCEPTANCE
- T
- Wear > code allowable min. wall at end of next refueling cycle or eEpEcted operating cycle (+ 10% margin of that time)
COMMITMENTS
- Analyze within 3 months
- If analysis shows unacceptable conditions, perform inspection on initial sample selection within 6 months.
- All others perform analysis and inspections within next refueling cycle or 18 operating months.
PROGRAM FOLLOWUP
- General inspection results through " Nuclear Network"
- INFO include programmatic review in plant evaluations j
Appendix A L_________-_____-__--___-_--__--_-_______-____-__--__-_-__-_____________-__-_-__-____-____-____--_______
r -
SECTION 1 BACKGROUND Many nuclear utilities initiated various inspections and investigations of erosion / corrosion phenomena in piping containing single phase, high energy fluid subsequent to the Surry incident in December 1986. These activities were initiated since this was the first incident of this kind for a nuclear utility.
The need existed for the nuclear industry to establish initiatives to identify evidence of single phase ero. ion / corrosion and thus help ensure personnel safety and minimize unnecessary plant challenges resulting from potential further failures.
It was paramount that timely and appropriate action be taken.
Although some action has already been taken, further action is required.
Hence, analysis and limited but thorough baseline inspection program is necessary.
Schedules for completion of these industry efforts should be aggressive for those units where the potential exists indicative of unacceptable wall thinning.
For other units where no indication exists, analysis and inspections are necessary but may be performed on a schedule commensurate with normal refueling cycles.
To assist in performing the above analysis and inspection tasks, the NUMARC Working Group, consictent with their stated mission (Appendix A), is providing guidance to industry in the following areas:
Selection process, including analytical methods, for inspection points Inspection methods and techniques including related acceptance criteria Possible remedy options for near term
- Nature and extent of future inspections I
l 1-1 NUREG-1344 1
Appendix A
l SECTION 2 GUIDANCE l
i SAMPLE SELECTION AND EXPANSION Introduction A large amount of information exists regarding this subject and was discussed in detail at the EPRI Workshop on Erosion / Corrosion April 14, 15 and at various industry briefings.
The purpose here is to provide. summary information and direction for the ensuing industry efforts to determine the scope of the concern and appropriate actions to be taken.
The following are summary recommendations and will be discussed below:
- The initial sample size should be, as a minimum, the 10 "most susceptible locations" and 5 additional locations based on unique operating conditions or special considerations.
- A structured approach should be employed to increase sampling size upon indication of unacceptable thinning.
Sample Selection The following guidance should be used to determine the locations to be investigated:
- 1) Determine population of piping products listed in Table I which are contained in the portions of piping systems listed in Table 2 (page 2-7).
have
- 2) The piping subsystems may be grouped into examination categories that similar characteristics.
- 3) Based upon EPRI methodology (reference summary provided below) or engineering judgement / analysis, determine which piping products / locations most susceptible to erosion / corrosion by considering the effect of the are parameters listed in Table 3.
List these in descending order of severity (likelihood of erosion / corrosion to exist). Unusual operating conditions (e.g. extended recire line flow or others) which are different from normal operating conditions should be considered in generating the list.
- 4) Choose the 10 most susceptible, at a minimum, of piping product / locations from the list generated in Step 3 above. This will conservatively bias the sample.
In addition, 5 piping product / locations should be selected at random.
EPRI Model Summary The erosion-corrosion damage process is fundamentally flow assisted corrosion.
is observed only when specific combinations of piping material composition.
It model water chemistry, and hydrodynamic conditions co-exist.The EPRI empirical l
I was developed by correlating a large amount of experimental erosion-corrosion 2-1 NUREG-1344 2
Anoendix A
rates..both from laboratory and plant ' data, with the pertinent measurable system parameters.
The result is a series of factors which when multiplied together yield the~ predicted erosion-corrosion rate. As some of the factors are interrelated,'the model is not linear. The model formulation is comprised of 6 factors. incorporating various plant variables affecting each factor.
As can be seen, the following system and component variables must be identified to
{
use the model:
j l
- The piping material alloy content.
- The water chemistry variables of operating temperature, pH, oxygen ~
-concentration and the water treatment used.
l
- The hydrodynamic variables of pipe diameter, component geometry
]
(fitting type and configuration ) and flow rate.
s A personal computer based program, CHEC (Chexal-Horowitz-Erosion-Corrosion),
was developed for prediction of erosion-corrosion in single-phase piping systems using specific plant data. The program was developed by EPRI as part of its ongoing research effort on erosion-corrosion in both nuclear and fossil power plants. Use of the program will be detailed in technical report NSAC-112 (The EpRI Computer Program for Erosion / Corrosion User Manual).
For a given plant or unit, the program will:
1.
Rank components in the piping system in order of susceptibility to erosion-corrosion.
2.
Choose most susceptible locations for inspection based on several criteria.
3.
Use wall thickness inspection data to develop a plant specific model to predict time to reach minimum required wall thickness.
The computer program is a user friendly, interactive program in simple language with multiple data files included in the program to provide input on nominal wall thickness, geometry factors, and other important factors needed to supplement computations.
It represents the most complete data base available on single phase erosion / corrosion.
Specific plant inspection data is combined with the e=pirical model to refine useage of the program to the specific unit or plant.
2-2 I
l l
Appendix A
]
L_
Saeple Expansion:
For each piping product (component) found that has a vall thickness which is below code minimum requirements or expected within the next refueling cycle or expected operating cycle (+10% margin of that time) to be below code minimum requirements and which is known to be caused by erosion / corrosion, the inspection shall be expanded to additional susceptible components in the examination category based on engineering judgement (Reference " Acceptance Guidelines" pg. 2-5).
The additional samples shall be taken from similar or like components in the examination category (i.e. sister train or similar arrangement) or components in proximity to the area of concern.
The inspection of additional samples by this criteria is also required for piping product / component determined to be unacceptable in any additional test lot.
When the EPRI model is used, additional samples may be taken from the next
" tier" of susceptible components.
i l
2-3 NUREG-1344 4
Appendix A
INSPECTION AND ACCEPTANCE Introduction EPRI Document 1570-2 (Nondestructive Examination of Ferritic piping for Erosion / Corrosion) provided a compilation of various methods that nay be employed in performing inspections to detect erosion / corrosion.
Most methods used within the industry to determine pipe / component wall thickness are well developed, provide repeatable and accurate reeults, and are governed by standard practices.
Upon component inspection, decisions must be made regarding determination of thinning and resulting potential replacement or continued monitoring. Acceptance guidelines at this stage are not well defined and consistently applied.
The following summary recommendations are provided:
- UT or RT may be utilized for the initial round of inspections with personnel qualified in accordance with SNT-TC-1A.
NDE procedures should be reviewed by qualified Level III.
- Utilize current industry experience gained in performance of inspections.
Establish benchmark or zero wear point at T,,, 100% wear at ecode min.
Inspection Guidelines If UT methods are utilized, the following guidance is provided.
1.
Grids - The area of interest is laid out in a grid and thickness readings are recorded for the points where the grid lines cross.
Grid lines may be close together (1") or far apart (6").
One method that is particularly effective in giving both thorough coverage of an area and a sufficient number of data points is the use of the large grid (greater than 2")
where the entire area of interest is scanned. (Reference attached sketches, Appendix B).
The first step would be to collect data at the grid line intersection points.
The second step would be to scan the entire area of interest and record the location and thickness for any point that is more than 20% below the average of the four adjacent grid intersection readings.
(Reference attached sketches Appendix B).
2.
Partial Grids - These are primarily used for elbows but may be similarly applied to other components.
There are five areas of interest for this type of scan. The first two would be segments centered on the intrados and the extrados.
The remaining three would be bands running circumferentially around the elbow at each pipe weld and one midway between the welds (Reference attached sketches, Appendix B).
Areas of interest would be marked with a grid a'nd readings would be taken similar to the method described above.
3.
Quick Scan - This type of scan may be used as a preliminary inspection or scoping work where a large amount of pipe is being evaluated for f.urthe r examination.
The key element in this type of inspection is setting the A-scan of the ultrasonic scope such that a small change in pipe thickness 2-4 NUREG-1344 5
Appendix A
will make a large change in the sweep position of the first back wall j
reflection. The inspection sequence would begin by placing couplant on the i
area to be scanned.
The technician would then place the transducer on the pipe and moving it a specified distance clockwise then. counterclockwise around the circumference of the pipe. At the end of each circumferential scan, the technician would index the transducer axially by a specified distance and perform another circumferential scan.
this process would continue until the area to be quick scanned was complete.
During the scanning the technician would watch for change in the sweep position of the A-scan presentation of the scope to see if there were any significant changes on the pipe wall thickness.
{
4 Automatic Scanning - This provides an excellent scan coverage as well as thorough record keeping for the inspection parameters, but this method is i
more time consuming than conventional methods.
-5.
Marking - Components should be appropriately marked (i.e. with high temp.
paints, low stress stamps, etc.) for reference for future inspections.
If RT methods are utilized, refer to EPRI Document 1570-2 and related references.
Acceptance Guidelines l
l The NDE results should be used to calculate the approximate erosion / corrosion rate and the number of cycles remaining before the component reaches minimum wall thickness.
If the calculations indicate that an area will reach code minimum allowable wall thickness within'l refueling cycle or expected operating cycle
(+10% margin of that time), the questionable component must be repaired or replaced unless the results of an engineering analysis show that there is an acceptable safety margin for continued operation beyond that point.
1.
Erosion / Corrosion (E/C) Rate Calculation:
(For each piping product / location) d)/ Time E/C Rate
= (t
-t T
= Nominal wall thickness
- nom T
= Actual wall thickness from the in-plant inspections measured (min value).
l Time
= Operating time, hours critical
- The actual nominal wall thickness may be significantly different than the value listed in the manufacturer's specification.
If the wall loss is I
localized it is often possible to determine a more realistic nominal wall thickness value through actual measurement on areas of the component that are not showing signs of wall loss.
Determined or measured nominal vall thickness should not be used unless it is equal to or greater than the manufacturer's specified, or as supplied, nominal wall thickness.
2-5 NUREG-1344 6
Appendix A
i l
2.
Engineering Evaluation:
Each utility should choose its own method for engineering evaluation'.ithin Code requirements.
3.
Subsequent Inspection:
Each utility should perform future inspections based on the results of the initial inspections. Timing of the inspections woulo be based on predicted
-wear rates utilizing initial inspection data and related acceptance guidelines.
l l
l l
l l
i l
2-6 NUREG-1344 7
Appendix A
e f
TABLE 1 SUGCESTED FITTINGS
- Closely Coupled Fittings or Configurations
- Entrant Tee, Combining Tee, Splitting Tee
- 90' Elbow
- Reducer / Expander
- Straight Section of Pipe Downstream of:
- Reducer
- Flow Control / Throttling Valve
- Restricting Orifices
- Multiple Thermove11s, etc.
TABLE 2 i
SUGGESTED PIPING LOCATIONS l
- Feedwater Suction
- Feedwater Discharge
)
- Heater Drain Pump Discharge i
- Condensate from FW Heater i
TABLE 3 KEY PARAMETERS
- C.S. piping & components - major parameter, chromium content
- pH
- 02 * "**""
l
- Fluid temperature
- Local / Bulk flow rate
- Piping product geometry factor
- Joint configurations (backing rings, etc.)
1 NOTE:
Information extracted from EPRI Workshop Information (April 14-15) and EPRI Report NP-3944.
2-7 NUREG-1344 8
Appendix A
_}
SECTION 3 COMMITMENTS INSPECTION SCHEDULES In order to provide sufficient.information to further assess the ex:ent of the l
concern regarding single-phase erosion / corrosion and to gather this information j
in a controlled and timely fashion, the following schedules are proposed-1.
All units should conduct an analysis in accordance with EPRI method within i
3 months of release of the EPRI model or provide other acceptable evaluation methods.
2.
Where analysis and current operating parameters / conditions are indicative of unacceptable wall thinning, they should perform inspections within 6 months after release of the EPRI model.
3.
For units where programs have previously performed inspections-based on other methodology, they must ensure the most susceptible locations identified by the analysis performed in accordance with 1.
above are inspected in accordance with 2. above or 4. below as applicable.
4.
All other units should perform inspections at their most susceptible l
locations as a minimum at the next refueling outage or within 18 operating months, whichever is sooner after release of the EPRI model.
5.
Future inspections will be based on the results of the first inspection and inspections continued accordingly.
INSPECTION AND PROGRAM FOLLOWUP General inspection results for each nuclear unit will be provided by " Nuclear Network" entries by the utility. Program followup will be provided through ongoing INPO plant evaluations beginning in June 1987.
3-1 NUREG-1344 9
Appendix A
APPENDIX A NUMARC TECHNICAL SUBCOMMITTEE WORKING CROUP ON PIPING EROSION / CORROSION (SPECIFIC OBJECTIVES)
MISSION Review current industry activity regarding inspection plans for piping containing single phase, high energy fluid that is susceptible to erosion / corrosion phenomena.
Review and evaluate current technical information from EPRI and others regarding inspection
- criteria, extent of inspections, and scheduling of inspections.
Identify parameters affecting erosion / corrosion in nuclear power plant piping and their relative importance.
Provide screening
- criteria, inspection and acceptance guidelines, and possible remedy options for near term concerns.
Decemine whether an industry-wide program to monitor pipe wall thinning is technically justified.
Consolidate and coordinate industry positions and plans to ensure any potential generic concerns are addressed.
Formulate actions and provide required irinstry liaison with NRC.
1 i
i NUREG-1344 10 Appendix A
APPENDII B CRID LAYOUT SKETCHES l
NUREG-1344 11 Appendix A
SCAN n
N FLOW J
- WELD JOINT DATA POINTS
- WELD JOINT i
w Grid-Type Inspection
- NOTE: SIMILAR PHILOSOPHY SHOULD BE APPLIED TO OTHER COMPONENTS (i.e. REDUCER / EXPANDER, TEE, etc.)
i
- SEE SECTION 2, PAGE 4 NUREG-1344 12 Appendix A aL________________-_.
SCAN O
m FLOW
)
- WELD JOINT DATA POINTS
- WELD JOINT w
Example O" A Partial Grid Inspection Of An Elbow *
- SEE SECTION 2, PAGE 4 NUREG-1344 13 Appendix A
WELD JOINT I
\\
t i TEE
, wetg 30ig7 REDUCER
~
[
N::=
n i -
f
- WELD JOINT
- WELD JOINT Other Applications Of The Partial Grid Methoc
- SEE SECTION 2, PAGE 4 NUREC-1344 14 Appendix A
l l
NOTE: SIMILAR PHILOSOPHY SHOULD BE APPLIED TO OTHER COMPONENTS (i.e. REDUCER / EXPANDER, TEE, etc.)
/*
O ge Sce/e' Qi V
Quick-Scan Inspection
- SEE SECTION 2, PAGE 4 NUREG-1344 15 Appendix A
- u. =ve6... out.,o,co.. w oN i ai oa t Nu se a..,-.. reoc, v.,
y;*o= =
E'i 8 BIBLIOGRAPHIC DATA SHEET NUREG-1344
$tt INSimVCTIONS ON tag atvinst 2 TITLE AND $v81#1Lt 3 (gaveggaNa Erosion / Corrosion-Induced Pipe Wall 5.nning in U.S. Nuclear Power Plants e Dat t mgs'O*T COMPititD MON T ee
.gan March 1989 l
..vi o..,
. DA W E ALPO*185SvtD Paul C. S. Wu g
0%,.
April 1989
? PERf DRMsNG QRf,ANi2 AliON Naut AND MAsuNG ADDfsE 55 tearse,sela Comi e PROJEC* 146E *QAK WNif hvMPt h Division of Engineering and Systems Technology Office of Nuclear Reactor Regulation
.,,No o..N, vo...
U.S. Nuclear Regulatory Commission Washingtc'i, D. C.
20555 N/A SO SPON50meNG omGaNelaT EON Nave av wait NG avas se,,,,,,,, r. ca,.'
s ie T v't 08 H E*08'T Same as 7, above.
,,,00co,,,,,0,,,,,,,,,.,,,,
N/A 12 EVPPLE MENT AA v NOTts 13 A55 TRACT 4100 e oras or esst Erosion / corrosion in single-phase piping systems was not clearly recognized as a potential safety issue before the pipe rupture incident at the Surry Power Station in December 1986. This incident reminded the nuclear industry and the regulators that neither the U.S. Nuclear Regulatory Commission (NRC) nor the American Society of Mechanical Engineers (ASME)Section XI Boiler and Pressure Vessel Code require utilities to monitor erosion / corrosion in the secondary systems of nuclear power plants. This report provides a brief review of the erosion / corrosion phenomenon and its major occurrences in nuclear power plants.
In addition, efforts by the NRC, the industry, and the ASME Section XI Committee to address this issue are described.
Finally, results of the survey and plant audits conducted by the NRC to assess the extent of erosion /corrosior.-induced piping degradation and the status of program implementation regarding erosion /
corrosion monitoring are discussed. This report will support a staff recom-mendation for an additional regulatory requirement concerning erosion / corrosion monitoring.
.. oocout~,.N.s.4......o.os vic.,,10.,
,,...,,,m,.
Erosion / Corrosion Flow-Assisted Corrosion Nondestructive Examination Ultrasonic Testing Unlimited Pressurized Water Reactors Boiling Water Reactors
..o Nv.... s,o,e N.No' o """5 Unclassified (IAar report, Unclassified
,, u... o...G u 14 D A ec{
w
.z UNITED STATES senci.oovam ctass a,,,
NUCLEAR REGULATORY COMMISSION
' 5'^ &*J'5 "^'"
E WASHINGTON, D.C. 20555
,,,,,,,,c.,
p L
OFFICIAL BUSINESS y
a PENALTY FOR PRIVATE USE, G00 1 1 h 120555139531 US NRC-0ADMPUBLICATIONS SVCS DIV FOIA TPS pop-NUREG P-209 DC 20555 WASHINGTON
$0E0Z oO
$OEO Y
m 8
9 0
3 m
E F
=
zO E
P z
C P!
Of E
i a
4a t-3