IR 05000483/1999011
| ML20212G031 | |
| Person / Time | |
|---|---|
| Site: | Callaway  |
| Issue date: | 09/22/1999 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML20212G027 | List: |
| References | |
| 50-483-99-11, NUDOCS 9909290060 | |
| Download: ML20212G031 (40) | |
Text
r.
'
.
ENCLOSURE l
U.S. NUCLEAR REGULATORY COMMISSION
REGION IV
Docket No.: 50-483 License No.: NPF-30 Report No.: 50-483/99-11 Licensee: Union Electric Company Facility: Callaway Plant Location: Junction Highway CC and Highway O Fulton, Missouri Dates: August 12-20,1999 inspectors: J. F. Melfi, Senior Resident inspector J. D. Hanna, Resident inspector J. Medoff, Materials Engineer C. G. Santos, Materials Engineer G. W. Johnston, Senior Licensed Examiner Approved By: W. D. Johnson, Chief, Project Branch B ATTACHMENTS:
Attachment 1: Supplemental information Attachment 2: Table and Figure Attachment 3: Inspection Plan Attachment 4: Photographs lJ
~
9909290060 990922 PDR ADOCK 05000483 0 PM
p g.
!
l I EXECUTIVE SUMMARY Callaway Plant NRC Inspection Report No. 50-483/99-11 Operations
- Operator performance was good and was exemplified by cautious decision making in regard to identifying the need to trip the reactor, close the main steam isolation valves, and evacuate the turbine building following rupture of a 6-inch drain line between a moisture separator reheater first stage reheater drain tank and feedwater Heater 6 Equipment affected by the event did present some challenges to the operators, but the loss of availability of equipment affected by the drain-line break did not compromise the ability to shut down the plant (Section 01.1).
- Safety-related equipment performed as expected and the plant transient response was as expected (Section O1.2).
- Because of the intrusion of steam and moisture, some nonsafety-related equipment did not function as designed. The fire suppression system did respond as expecte Certain radiation monitors failed, but the ability to perform manual backup samples was always available (Section 01.3).
- Two loud noises were noted in the turbine building within 70 minutes of the rupture which generally correlate with load swings scen by the plant. The inspectors concluded that these noises were not associated with the steam-line rupture (Section 01.4).
- The method for calculating the shutdown margin in the presence of the axial offset !
anomaly was conservative and adequately bounded exhibited plant behavior !
i (Section O2.1).
- The inspectors determined that the scope of the licensee's damage assessment was reasonable and adequate. The observed damage was corrected (Section O2.2).
Maintenance
,
- The licensee's root cause failure analysis for the pipe rupture, from the visual inspections and laboratory tests, showed combination of two corrosion mechanisms; flow-accelerated corrosion and water impingement. The licensee's analysis of the sequence of the piping failure was plausible (Section M2.1).
- Wall thickness predictions generated for components by the CHECWORKS model were
,
generally in agreement with wall thickness values generated from actual measurements l of the components; this provides indication that the flow assisted corrosion control program is predicting wear in the carbon steel piping systems (Section M2.1).
l
- The inspectors identified a weakness in the licensee's flow-accelerated corrosion program in that the program did not require inspecting the pipe upstream or downstream of a component (e.g., elbow tee, expander) scheduled for inspection. The licensee's l
.
c
} -2-program was not consistent with industry guidance to inspect two pipe diameters upstream and two pipe diameters downstream of the component scheduled for inspection. ir specting the downstream pipe within two pipe diameters of the elbow cou!d have identified the excessive wear (Section M2.1).
- Due to a failure of the extraction steam system, the licensee imposed additional monitoring on extraction steam system as required by the maintenance rule. The licensee did make a reasonable effort to monitor the system using the erosion / corrosion program, but the program did not model this particular pipe condition satisfactorily, and did not account for the multiple-corrosion mechanisms that could affect the pipe (Section M2.2).
- The inspectors concluded that a similar industry event was adequately reviewed, and concluded that the licensee's quality assurance reports did not identify any previous program weaknesses (Section M7.1).
Enaineerina a The licensee and the inspectors identified an instance where some pipe material changes were not shown on plant isometric drawings after a minor modification. The system was not safety related. The problem is being resolved by the licensee's corrective action program as Suggestion-Occurrence-Solution Report 99-1641 l (Section E2.2). l
I
-
Report Details l
Summary of Plant Status l
l Operators manually tripped the plant from 100 percent power on August 11,1999, following l l
rupture of a pipe in the extraction steam system. The failed line was from moisture separatur i Reheater D to feedwater Heater 6B. The unit was maintained in hot shutdown for repairs and was subsequently returned to power operations.
! l
'
1. Operations 01 Conduct of Operations (93702)
'0 Steam Extraction Line Ruoture Event Inspection Scoce (40500. 71707)
The inspectors reviewed the sequence of events for the reactor trip on August 11,1999; assessed operator response to the event; and determined the adequacy of safety-related and nonsafety related equipment performance during and following the reactor trip. Interviews wc:s conducted with plant operators and control room logs were 4 reviewe Observations and Findinas On August 11,1999, the plant was operating at 100 percent power with no unusual !
activities in progress. At approximately 9:17 a.m., operators saw unusual alarms related to feedwater Heater 6B, with noise noted coming from the turbine building. After
~ determining that a major steam leak was in progress, operators manually tripped the reactor. The reactor trip resulted in a turbine tri The following is the sequence of events for the steam extraction line rupture that occurred and possible related event Time Line of Events ;
Time Events Description 8:08 A chemistry technician in the turbine building cold lab noticed an unusual noise. No report was made to the control roo :09 Operators indicated that when this occurred they noticed a decrease in electrical load of 100 megawatts and a swing up in load of 80 megawatts (1081 to 1225.6 megawatts electric) before the plant stabilized at nominal full power. Alarm MSEAH84 was received indicating high grid voltage.
!
l 8:37 Another instance of a noise in the turbine building was noticed by L the chemistry technician in the turbine building cold lab. No f electrical load swing was noted in the control room at this time.
I-
q
)
{
.
-2-9:17 The first control room indication of the event was when Annunciator 103C, second stage Reheater B drain tank, came i The plant computer indicated that the high level dump Valve B was not closed. Also at this time, the field supervisor heard a loud bang and saw steam in the turbine building. Concurrently the operators in the control room noted a 15 to 20 megawatt electric dro :18 The PKO1/2 ground detector alarm came in and flashed, and continued to come in intermittently. Around this time the work control operating supervisor entered the control room and reported excessive noise coming from the turbine building. He stated that a relief valve may have been lifting from the reboile The reboiler relief valve is located just above the area where the leak occurre :19-9:24 Somewhere in this time frame, a plant announcement was made to evacuate the turbine building. (Security actions time line indicates this may have occurred at 9:25 a.m.).
9:20 Annunciator 121 A, main feed Pump A oil pressure low, came i This pump is below the area where the drain line rupture occurre :22 Several fire alarms were received. The diesel and an electric fire pump started. A computer uninterruptible power supply failure occurred, the computer display screens went blank, and the plant computer went offline. This did not affect the operators' response to the event. However, important information was lost because of the computer failur :23 The unisolable steam leak was still in progress, and the lo:ation was unknown. The control room staff began working on a strategy to reduce power and to isolate high pressure feedwater extractio :25 The field supentisor entered the control room to report that the steam leak was large. He conferred with the control room supervisor. Based on the observations in the turbine building, the control room supervisor decided to trip the reactor. The field supervisor concurred. The turbine tripped automatically af ter the reactor was manually tripped. The field supervisor returned to the field office to coordinate an effort to locate the leak and initiate a search for potentially missing personne :25-9:35 a t- Entry into Emergency Operating Procedure E-0, " Reactor Trip or Safety injection," was made following the manually initiated reactor trip. While performing the actions of step 4, the operators m
r
.
.
-3-noted indication of steam flow. Several condenser dump valves indicated open and closed erratically. Position meters were bouncing and a random pattern of open and closed lights were received. Controller c;emand output was zero at this tim :29 After more precise reports came to the control room, the operators determined that, with steam flow still present in the turbine building and the erratic performance of the condenser steam dumps, that the best course of action was to close the main steam isolation valves. The valves all close :37 Immediate boration of the reactor coolant system was commenced in accordance with Attachment 12 of the emergency operating procedure. A briefing was held by the shift supervisor to review plant conditions. Since safety injection was not required, the operators transitioned to Procedure ES-0.1," Reactor Trip Response." j i
9:38 The emergency duty officer decided to initiate personnel I accountability to determine if there were any missing persons in the turbine building. The plant emergency alarm was sounde There was a problem in making the announcement because the Gaitronics phone at the shift technical advisor's desk was j inoperabl I 10:00 Emergency Procedure ES-0.1 " Reactor Trip Recovery," was exited and Procedure OTG-ZZ-00005," Plant Shutdown 20%
Power to Hi t Standby," was entere :10 Accountability was completed, with no personnel reported missing. After the initial count, six people were unaccounted for, I but all responded quickly to a plant pag :50 Procedure OTG-ZZ-00006, " Plant Cooldown: Hot Standby to Cold Shutdown," was entered. A decision to remain at normal I reactor pressure and normal no-load temperature nas made at ,
this time because of the unavaHabihty of the condense Essentially ali emergency actions were complete at this tim Operator Response !
The event review team assessment of operator performance determined that the j i
operators demonstrated good decision making. This determination was made following interviews with the crew members and a review of the event. Examples cited by the team included the manual reactor trip, turbine building evacuation, and the fast closure l of the main steam isolation valves to isolate the secondar I l
j
-
,
.
.
After a review of the events, the inspectors independently interviewed the crew
'
members. The operators' decision making was good when conditions were know When the plant conditions were not known, the operators did not take untoward actions that may have jeopardized plant recovery. The decision to trip the reactor was l
'
appropriate given the potentialimplications of a serious secondary plant leak on the overall plant. When the decision was made, the operators proceeded through the immediate actions of Procedure E-0, " Reactor Trip or Safety injection." They did halt for a period of time at step 4, which requires confirmation that a safety injection is not required. The conditions did not require a safety injection. The operators were apprized i
'
that the size of the secondary leak was large and most likely could not be isolated. The decis'on was then made to fast close the main steam isolation valves. This action was completed and confirmed. The inspectors noted that this was a prudent action, and that taking that action at that step in Procedure E-0 was appropriate, given the
'
circumstances. Subsequent operator actions were routin Operator response with regard to the fire protection system actuation was appropriat Securing the system, following the fire protection system actuation was appropriat The pumps were stopped and placed in automatic, and a line that had a pinhole leak was isolated until repairs were made. Although a fire watch was not in place !
immediately following the event, it was established about 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> after the even Access to the turbine building was restricted during that time for safety reasons. There were fire brigade membero in the building for a significant portion of the time, and operations personnel were in the proximity performing recovery actions. The fire protection system caused little consequence to the outcome of the even Conclusions Operator performance was good and was exemplified by cautious decision making in regard to identifying the need to trip the reactor, close the main steam isolation valves, and evacuate the turbine building. Equipment affected by the event did present some challenges to the operators but the loss of availability of equipment affected by the drain line break did not compromise the abi!ity to shut down the plan .2 Safetv-Related Plant Eauipment Performance Associated with the Eve.nt Inspection Scope (93702)
The inspectors reviewed the safety-related equipment response to the August 11,1999, reactor trip. Both primary and secondary plant indications and responses were evaluated. The performance and reliability of important equipment were assessed. The potential for steam and moisture interaction with other plant equipment that was not directly affected by the rupture was also assesse > .
l .
1- )
-5-l Observations and Findinas All safety-related equipment functioned as designed during this event. The steam and water from the rupture did not affect any safety-related structures such as the control building or auxiliary buildin Licensed Reactor Power l l
Reactor power did exceed 100 percent but it was transitory and did not contribute to the I event. The peak was 100.78 percent of full power on the nuclear instrumentation and l thermal power went to 3591.8 megawatts thermal (licensed reactor thermal power is j 3565 megawatts thermal). The power level persisted throughout the event, about 9 to '
10 minutes, but was not noticeable on normal plant instrumentation. An examination of recorder traces did not reveal a significant level above the normal trace for 100 percent powe j Main Steam Pressure At no time were the lift set points of the main steam safety valves challenged. Peak pressure following ths reactor trip was slightly over the no-load set point pressure for the condenser steam dumps. Later, following the closure of the main steam isolation valves, steam line pressure did go to the atmospheric dump valve set point value (1125 psig), which is well below the first main steam safety valve set point. The sequence of events precluded the potential for a challenge to main steam safety valve set point Conclusions Safoty-related equipment performed as expected and the plant transient response was as expecte .3 Nonsafetv-Related Plant Eauipment Performance Associated with the Event Inspection Scope (93702)
l The inspectors reviewed the plant and equipment response to the event. Both primary and secondary plant indications and responses were evaluated. The performance and l reliability of important equipment were assessed. The potential for steam and moisture l interaction with other plant equipment that was not directly affected by the rupture was l also assessed.
l l Observations and Findinas Some nonsafety-related equipment in the area of the oipe rupture was in the damage
,
area of the break, but continued to operate satisfactorily. This included the heater drain, I condensate, and main feedwater pumps. The following are nonsafety-related systema that were affected by the pipe ruptur l F.
,
l
6-
l Steam Dumo System The steam dump system performed as designed, but operators noticed erratic indication on some of the steam dump valves. The licensee subsequently discovered that water entered terminal boxes associated with these valves. The licensee dried the water out of these terminal boxes and the steam dump valves indicated correctl I Fire Suporessien System The fire suppression system responded as expected for the conditions seen. A preaction valve opened in the south 2000 foot elevation of the turbine building, in response to a signal from a rate compensated thermal detector (setpoint 190 F). The l environmental conditions present in the turbine building at the time of the drain line i break caused the actuation. The system is a dry-pipe system and requires that a l proaction valve open to supply water into the line. The sprinkler heads in the fire l suppression system did not actuate, because the area probably did not see temperature conditions of greater than 212 F, which was required for the fusible links in the sprinkler heads. A pinhole leak was found in a 1-inch branch line in the overhead of the 2000-foot elevation. The line was briefly isolated and replaced the following day with a i like-kind galvanized pipe. The installed design of the pipe was determined to be the cause. There is a drop in the line from the main header that does not allow the line to drain after being filled with water following a preaction valve opening. The licensee has placed the concern in their corrective action program and will determine what possible modifications are required to correct the proble Radiation and Effluent Monitors Several components were affected. They were radwaste effluent Monitors GHRE10A and B, plant vent Monitors GTRE21 A and B, and Monitors ABRE111 and FCRE38 The power supplies for these monitors were from inverter SPO1. The failure occurred from water drawn into the cabinet, causing momentary shorts that resulted in a voltage spike. Protective features (fuses and varistors) in the monitors prevented damage to them; however, the event did render them inoperative. Repairs were completed quickly by replacing the fuses and varistors. The monitors were out of service for most of the event. The ability to perform backup manual grab samples was always availabl Therefore, if monitoring had been necessary, the impact would have been minimal. The licensee is considering modifications to certain electrical panel Plant Computer The failures of the plant computer and other systems energized from the same power supplies were caused by the high moisture environment from the break. Uninterruptible power Supply POO2 failed due to steam and moisture condensation, this cauced a 400 ampere fuse to blow in the alternate AC feeder. The failure caused meteordgical tower communications to fail and the alarm and utility printers to lose power. Also losing
,
power were emergency plan response equipment, the Magnem and Sentry computers, and the radio. Other affected equipment included four cathode ray tubes in the control room and the equipment on the shift technical advisor's desk. The shift technical
_
,
l' ,
.
l 7- )
)
advisor's desk equipment which failed included the Gaitronics phone, which hampered making plant announcements. Announcements had to be made on a phone outside the control roo Conclusions Because of the intrusion of steam and moisture, some nonsafety-related equipment did not perform as designed. The fire suppression system did respond as expecte Certain radiation monitors failed, but the ability to acquire manua; backup samples was always availabl .4 Evaluation of Turbine Buildina Noises Before Line Ruoture Event Insosciion Scope (40500. 71707)
The inspectors reviewed the circumstances surrounding noises in the turbirie building prior to the steam rupture to assess if these noises may have been related to the pipe ruptur Observations and Findinns l A chemistry technician in the cold lab heard two loud noises in the turbine building about 70 minutes and 40 minutes prior to the rupture. The noises were loud enough and unusual enough for the technician to note them in the log. The licensee correlated these noises to electrical load swings which occurred to approximately 145 and 50 megawatts electric, respectively. These load swings would have affected the turbine ,
'
generator and may have caused a pressure transient in the secondary side of the plan The inspectors do not believe that these load swings induced the rupture. Plant !
operation was steady for a significant length of time prior to the rupture. Examination of the ruptured pipe indicated a rapid break, and did not indicate a leak before the brea Personnel walking through the area several minutes prior to the break, did not observe any steam leaks. An NRC inspector was in the turbine building when the break occurred and heard the rapid release of steam. Based on these observations, the inspectors do not believe that the load swings were related to the steam ruptur Conclusions Two loud noises were noted in the turbine building within 70 minutes of the rupture. The noises generally correlated with electrical load swings seen by the plant. The inspectors concluded that these noises were not associated with the steam-line ruptur .
,
a 8-02 Operational Status of Facilities and Equipment O2.1 Boron / Shutdown Maroin Response Followina Reactor Trio Inspection Scope (92703)
In April 1999, the licensee adopted a modification to the method for calculating j shutdown margin in the presence of the axial offset anomaly. The new technique j assumed a delayed release of some of the beron from the crud following a reactor trip, where the previous method assumed a complete and instantaneous release. The inspectors confirmed that the shutdown margin required by Technical Specifications was maintained during the reactor trip, and they also determined whether exhibited plant I response was bounded by the time-release boron mode ' Observations and Findinas The licensee calculated shutdown margin throughout the transient using a core model which was updated to the most recent flux map. The calculations used the following assumptions:
- The most reactive rod was stuck
- - Rod worth uncertainty was equal to 7 percent
- An instantaneous release of 100 percent of the boron from the crud occurred Technical Specification 3.1.1.1 requires that shutdown margin greater than 1300 pcm be maintained while in Mode 3. The lowest shutdown margin during the reactor transient, using the conservative rapid release model, was calculated to be 1356 pc The licensee then validated the time-release model using two comparisons. In the first,
' the posttrip crud samples for the recent reactor scram were compared with those for Cycle 5. Cycle 5 crud data was used as the comparison because it was the basis for the creation of the time-release model, and because the mass of boron immediately released to the coolant is a function of the crud immediately released. .The results shown in the table of Attachment 2 indicate that the behavior of the core during the Cycle 10 trip was consistent with the expected respons In the second validation, the predicted mass of boron released from the crud (based on the release rate model) was compared to the results of samples taken after the Cycle 10 trip. . Reactor coolant system lithium samples were used to calculate the mass of the boron released. The samples were corrected for other mechanisms affecting lithium '
concentrations and translated to boron mass using the relative atomic weights. The
. figure of Attachment 2 compares the actual boron release to the values predicted by both models.' The licensee determined that the boron mass release inferred from the samples was conservatively bounded by the predicted boron mass releas .
i .
-9-l Conclusions l The method for calculating the shutdown margin in the presence of the axial offset anomaly was conservative and adequately bounded exhibited plant behavio .2 Licensee Plant Damaae Assessment insoection Scoce The inspectwc reviewed the licensee's actions to assess and repair the damage from the pipe ruptur Observations and Findinas The licensee conducted a review and walkdown of equipment surrounding the pipe rupture. The licensee defined this area on plant maps for the different elevations. This included multi disciplinary (i.e., civil, mechanical, electrical, instrumentation and control)
walkdowns. The licensee identified various pieces of equipment that were affected by water and repaired these component Most of the plant damage was to structural supports associated with the ruptured lin The licensee replaced the line and installed new supports in the area. Electrical walkdowns identified moisture in some electrical cabinets, which were dried ou Instrumentation and control technicians opened instruments in the affected area to verify 1 that no water was present, or remove any water that was foun l l
All equipment functioned satisfactorily during startup. Approximately 1 week after i startup, one piece of equipment in the turbine building suffered a failure that may be related to the steam rupture. This piece of equipment was a solenoid valve associated with testing of a bleeder trip valve. The leads to the solenoid are Teflon coated and water had degraded the Teflon enough to cause an electrical short. The bleeder trip valve continues to operate, but the valve cannot be teste l Conclusions )
The inspectors determined that the scope of the licensee's damage assessment was ,
reasonabie and adequate. The observed damage was correcte l l
,. .
!
I l .
l 10-II. Maintenance
M2 Maintenance and Material Condition of Facilities and Equipment M Flow-Assisted Corrosion f Erosion / Corrosion) Issues Ippection Scope (49001. 62706)
The team reviewed the aspects of the licensee's maintenance rule and flow-assisted corrosion control programs as they relate to the circumstances of the rupture of the j first-stage reheater drain Tank A drain line. The team's review served the following purposes: ]
To determine whether the licensee had performed a plausible failure analysis for j the failed piping in the first stage reheater drain tank drain lin I
To determine whether the licensee had designed a sufficient flow-essisted corrosion control program for predicting and controlling wear in safety-related l and nonsafety piping systems fabricated from carbon or low allow steel )
- To determine whether the licensee's implementation of the flow-assisted corrosion control program was sufficient to predict the wear of both safety-related and nonsafety-related carbon (low alloy) steel piping system I Observations and Findinas The ruptured first-stage reheater drain tank drain line was fabricated from 6-inch diameter, Schedule 40, A-106, Grade B, carbon steel pipe. This system operates at a temperature of -410*F and a pressure of 264 psig. The line configuration consists of a straight horizontal 96-foot stretch of pipe welded to a 90 elbow at its discharge en l The 90* elbow is oriented to direct the flow downward at a 45 angle from horizontal. A 1-foot pipe section'is welded to the discharge side of the 90* elbow that is in turn welded to a 45' elbow. The 45* elbow redirects the flow back to horizontal, and is welded at its I discharge end to a 13 foot straight stretch of pipe. The 13-foot stretch of pipe is l
restrained by a pipe restraint (i.e., pipe clamp) located about 1.5 feet away from the ;
weld at the discharge end of the 45* elbow. The pipe sections or elbows were welded to '
each other using backing rings and E7018 filter metal. The pipe ruptured in the straight I piping section immediately downstream of the 45 elbow, with the minimum metal thickness about an inch away from the elbow's downstream weld. The stored energy in the system was sufficient to cause two guillotine breaks of the piping; the first at the location of the pipe restraint, and the second at a point approximately an inch away from
- the weld at the discharge end of the 45' elbow. The system's steam quality at the point ;
where the pipe failed is approximately 80 percent steam by volume (i.e., l 4.5 mass-percent steam; wet steam).
The licensee performed a root cause analysis to determine the mode of failure in the first-stage reheater drain tank drain line. The licensee's analysis reviews included: (1) a detailed analysis of the damage to structures, systems, and components in the general i
l
-
p 7
.
-
-1 1 - l l
vicinity of the pipe failure; (2) removal of the degraded piping components from the nondegraded portions of the drain line and visualinspections of the inside and outside surfaces to check for evidence of damage, erosion, or corrosion; (3) removal of material testing samples from the sectioned components and performance of material property testing (i.e., Rockwell hardness testing, chemical composition analysis, metallography, scanning electron microscopy of the fracture surfaces) at an independent laboratory; and (4) ultrasonic testing wall thickness measurements of the degraded components, in addition, the licensee conducted a CHECWORKS flow assisted corrosion analysis of the degraded system using the additional data from the ultrasonic testing measurements of the failed component, and expanded the ultrasonic testing scope to !
'
include nn additional 12 systems and 42 component locations. The selection of locations was based on identifying those locations that had similar geornetries. system operating conditions, and flow characteristics as those associated with the failed componen {
The licensee concluded that the failure of the first-stage reheater drain tank drain line !
pipe resulted from a multiple-mode degradation failure mechanism. In this mechanism, l the licensee attributed the failure of the line to the following events' !
i
Axial fishmouth rupture of the pipe approximately an inch away from the weld at I the discharge side of the 45* elbo *
Propagation of the axial flaw downstream until the flaw contacted the pipe clamp (about 18 inches along the pipe), then redirection of the flaw circumferentially along the pipe clamp. The pipe tore in a ductile manner completely around the pipe, leading to a guillotine break of the pipe at the pipe clam *
Separation of the two pipe sections due to jet forces, flattening of the pipe wall at the live end of the guillotine break, leading to more ductile tearing near the downstream end of the 45* elbow. An 18-inch long flattened section was still attached to the upstream pipe section along the top of the pip * The upstream pipe section hit a pipe support, and the 18-inch long flattened section then finished tearing off the upstream pipe. The upstream pipe section thus completed a separate guillotine brea An 18-inch long section of flattened piping was removed from the system by the rupture and the two guillotine breaks. The licensee discovered the flattened section of piping during the walkdown of the area following the event. The licensee concluded that the ,
following evidence supports this failure mechanism sequence:
)
- CHECWORKS modeling predicts two phase (water / steam) flow characteristics in this section of pipe with a steam quality that is approximately 80 percent steam by volume (i.e.,4.5 mass-percent steam, wet steam in this case).
i
- Visualinspections of the 90* elbow, the 1-foot pipe section, the 45* elbow, and the flattened 1.5-foot pipe section revealed the presence of magnetite (Fe3 0 4, a )
i i
l-12-black oxidation product in carbon steels) on the inside surfaces of the pipe This was indicative that oxidation has occurred in the system.
.
Visualinspection of the weld at the discharge side of the 45 elbow (including the portion of the straight pip 0 0.5 to 1.0 inch away where the pipe failed) shows extensive wear and pitting of the pipe, the weld, and the backing bar. Visual inspections of the inside surface of the 45* elbow demonstrate that the pitting extends into the elbow up to the elbow's upstrearn weld. No pitting was exhibited upstream of the backing-bar at the elbow's upstream weld.
Visual inspections of the flattened 1.5-foot pipe section also indicated extensive wall thinning and pitting of the pipe material; wall thickness measurements of the flattened pipe section indicated that the spool had worn to a thickness as low as 0.010 inch (nominal wall thickness for this pipe is 0.280 inch).
- Design calculations indicate that the code minimum allowable wall thickness (tmin) was 0.109 inch, and the burst pressure thicknesses ( bt urst) for the pipe conditions was 0.016 inch.
- Scanning electron microscopy results of the fractured surface on the flattened pipe section (magnification of 164X) displayed a scalloped surface; this type of scalloped appearance is characteristic in carbon steel components that have been eroded as a result of flow-assisted corrosion.
- Scanning electron microscopy results of the sample taken from a second area at the edge of the fracture surface on the flattened pipe section (magnification of 600X) displayed a dimpled surface in the fractured surface which is characteristic of metals that have failed by ductile overload (i.e., by ductile tearing).
The inspectors independently examined the outer and inner surfaces of the sections taken from the failed pipe and reviewed the licensee's independent laboratory analysis of the sections. The inspectors noted that the evidence supported the licensee's determination of the degradation mechanism for the piping, a combination of flow-accelerated corrosion and water impingement.
The inspectors also reviewed the licensee's design and implementation of the flow-assisted corrosion control program at the plant. The inspectors determined that the licensee designed its flow-assisted corrosion control program in accordance with the specific details in Engineering Department Procedure EDP-ZZ-01115, " Flow-Assisted Corrosion of Piping and Components Predictive Performance Manual," Revision 1 '
This procedure provides the plant-established guidelines and criteria for scheduling and performing ultrasonic testing inspections of components in the flow-assisted corrosion control program. The flow-assisted corrosion control program is designed using a CHECWORKS sof tware application program and models degradation from corrosion in plant system piping. The flow-assisted corrosion performance engineer models the following secondary systems for wear analysis:
l
,
(
'
.
l
.
-13-
- Third stage extraction !
- Fifth stage extraction j
- Seventh stage extraction
- Heater drain lines
- Moisture separator reheater drain tank drain lines
- Feedwater system a Condensate system l l
The inspectors confirmed that the systems modeled by the licensee were the systems which use carbon steel and operate in the pressure and temperature ranges conducive to flow assisted corrosion. The inspectors also confirmed that the flow-assisted corrosion performance engineer designed the flow-assisted corrosion model with the following plant-specific inputs:
System operating temperatures, pressures, enthalpies, and water / steam quality factors are inputs for the heat balance portion of the program as obtained from plant-specific thermal performance program data or from plant-specific heat balance diagrams;
- Materials and design inputs for the line modeling portions of the program are l obtained from plant-specific isometric drawings, valve manuals, Bechtel design specifications, and design calculations;
- Oxygen, amine, and pH inputs for the chemistry portion of the program are obtained from plant chemistry records; and
- System flow rate inputs for the flow analysis portions of the program are obtained from plant-specific thermal performance programs and from plant-specific heat balance diagram During discussions with the licensee, the inspectors determined that the 45* elbow in the first-stage reheater drain tank drain line was originally inspected in the late 1980s and again during Refuel 7 (March 1995). Neither inspection included portions of the straight pipe immediately downstream of the elbow. In order to determine whether the licensee's flow-assisted corrosion model could provide a reasonable prediction of the wear in carbon steel components, the inspectors compared the predicted wall-thickness results of other components modeled in the CHECWORKS program to the wall thickness values obtained from ultrasonic testing measurements of the component The inspectors verified that the predicted thickness results for components modeled in the licensee's CHECWORKS model were generally in agreement with the actual ultrasonic testing thickness measurements of the component The inspectors identified a weakness in the licensee's flow-assisted corrosion control program. The inspectors noted that the system had failed in the straight section of pipe within 1 inch of the discharge end of the 45' elbow. The licensee's inspections of the 45" elbow in the Refuel 7 did not include any portions of the straight piping immediately
.. !
!
.
I
'
-14-downstream of the elbow. The inspectors identified this weakness at the exit meeting that was held with the licensee on August 20,1999. The following section discusses this weakness in more detai Electric Power Research Institute Document NSAC-202L-R2, " Recommendations For an Effective Flow-Assisted Corrosion Program" (April 1999), in part, makes the following statements and recommendations in the guidelines for developing flow-assisted corrosion control programs:
- "It is . . . beneficial to inspect the area on both sides of each pipe-to-component weld. It is desirable to start the grid line on both sides of the weld, as close as possible to the toe of the weld, in order to locate potential thin areas adjacent to the weld. This will help detect the presence of backing rings, the use of counterbore to match the two inner surfaces, or the localized wear that is sometimes found adjacent to welds."
" Maximum wear in straight pipe downstream of components typically occurs within two pipe diameters of the connecting weld. Consideration should be given to extending the grid two diameters downstream (or two diameters upstream for expanders and expanding elbows), at least for the first two inspections."
l
'
Furthermore, during CHECWORKS training sessions, the designers of the CHECWORKS software programs note that the software programs are not infallible and recommend that utilities use engineering judgement in addition to the model's predictions when selecting piping locations for inspection. The CHECWORKS user's group and software designers, therefore, urge utilities to apply engineering judgement and identify cbnormal piping configurations that have the potential of being more ,
susceptible to flow-assisted corrosion than the models would predict (e.g., tee to i reducers, elbow to elbow, etc.). Thus, both the Electric Power Research Institute guidelines and the designers of CHECWORKS stress the importance of making sure that the inspection scope coverage for carbon steel or low-alloy steel components is large enough to identify thinning in those areas where excessive erosion would be expected; these areas include portions of the downstream components that are within two pipe diameters of the component scheduled for inspection, or within two pipe diameters upstream if the component scheduled for inspection is an expander or expanding elbo In contrast, the sample selection guidelines of Engineering Department Procedure EDP-ZZ-01115," Flow-Assisted Corrosion of Piping and Components Predictive Performance Manual," Revision 15, do not administratively require the flow-assisted corrosion engineer to inspect downstream of a component that is scheduled for examination. Instead, sample selection Guideline 2.4 in the procedure states that the flow-assisted corrosion engineer should consider including components l- within two pipe diameters downstream of any component exhibiting excessive wear, j Furthermore, the sample selection guideline does not provide any criteria for concluding l
when it is necessary to expand the inspection scope into the portion of the downstream component that is within two pipe diameters of the component scheduled for inspectio Thus, the licensee's procedure places the decision to expand the inspection scope r
I,
.
-15-I entirely on the judgement of the flow-assisted corrosion engineer. Although the licensee did inspect the 45' elbow in the ruptured first-stage reheater drain tank drain line during Refuel 7, the licensee concluded that it was not necessary to inspect into the straight pipe downstream of the elbow, and did not include any portions of the straight length of pipe in the inspection scope. If the licensee had inspected the downstream section piping that was within two pipe diameters of the 45* elbow, the inspections could have indicated that excessive wear was occurring in the straight run of pip l l Conclusions Following its review, the team concluded the following with respect to the licensee's root cause analysis of the failure event and design and implementation of the flow-assisted corrosion control program: ;
The licensee's root cause failure analysis for the pipe rupture, from the visual inspections and laboratory tests, showed combinations of two corrosion mechanisms; flow-accelerated corrosion and water impingement. The licensee's analysis of the sequence of the piping failure was plausibl *
Wall thickness predictions generated for components by the CHECWORKS model were generally in agreement with wait thickness values generated from i
'
actual measurements of the components; this provides indication that the flow-assisted corrosion control program is predicting wear in the carbon steel piping system * The inspectors identified a weakness in the licensee's flow-accelerated corrosion program in that the program did not require inspecting the pipe upstream or downstream of a component (e.g., elbow tee, expander) scheduled for inspection. The licensee's program was not consistent with industry guidance to inspect two pipe diameters upstream and two pipe diameters downstream of the component scheduled for inspection. Inspecting the downstream pipe within two pipe diameters of the elbow could have identified the excessive wea M2.2 Maintenance Rule Issues Inspection Scope (49001)
The team reviewed the aspects of the licensee's maintenance rule program that pertained to the extraction steam system and the circumstances surrounding the pipe rupture even Observations and Findinas The licensee determined that the failure of the piping in the fourth stage extraction steam system was most likely due to flow-accelerated corrosion combined with steam impingement of water. This combination is not modeled well by the CHECKWORKS computer program. The licensee considers the extraction steam system to be within the scope of the 10 CFR 50.65," Requirements for Monitoring the Effectiveness of
,
.
.
-16-Maintenance at Nuclear Power Plants." Ten CFR 50.65(a)(1) requires that licensees monitor their systems, structures, and components against licensee established goals such that there is reasonable assurance that the system, structure, or components are l capable of performing their intended functions. Ten CFR 50.65(a)(2) notes that monitoring as noted in 10 CFR 50.65(a)(1) is not required if their structures, systems, or components are being effectively controlled through appropriate preventative maintenance such that the structure, system, or component remains capable of performing its intended functio The licensee did establish goals for the nonsafety-related extraction steam system, using the erosion / corrosion program to monitor wear in this system, since wear rates generally correlated well to computer program predicted values. Goals for this systems included: (1) not more that three maintenance preventable functional failures, (2) no more than one reactor trip due to this system, and (3) no unplanned capacity loss of greater than 2 percent. The inspectors considered that the licensee's program was !
generally adequate to monitor the extraction steam system. However, the computer program used to predict erosion / corrosion did not model the combined effect of two corrosion mechanisms. The inspectors concluded that monitoring the extraction steam system using the erosion / corrosion program was reasonable, but could not have predicted a combination of corrosion mechanism ,
i The system performance goals were not met since this rupture caused more than a 2 percent loss-of-capacity factor. The licensee is evaluating additional actions to take regarding monitoring of this syste Conclusion Due to a failure of the extraction steam system, the licensee imposed additional monitoring on the extraction steam system as required by the maintenance rule. The licensee did make a reasonable effort to monitor the system using the erosion / corrosion l program, but the program did not model this particular pipe condition satisfactorily, and could not account for the multiple-corrosion mechanisms that could affect the pip M7 Quality Assurance in Maintenance Activities M7.1 Licensee Review of Previous Industrv Experience a.' Insoection Scope (62706)
The team reviewed the licensee's assessment of a previous similar industry event, and previous quality assurance audits of the erosbn/ corrosion program to determine if the licensee had previously identified problems in their program and had adequr ely evaluated the industry even _
O
.
- 17- Observations and Findinas i The inspector reviewed the licensee's response to NRC Information j Notice 97-84: " Rupture in Extraction Steam Piping as a Result of Flow-Accelerated j Corrosion," and recent quality assurance surveillance reports that reviewed the erosion / corrosion progra q The inspectors determined that the licensee adequately reviewed the information contained in Information Notice 97-84. The inspectors also reviewed previous quality assurance surveillances on the erosion / corrosion program, and these reports did not indicate any problems with the program. A previous review of the erosion / corrosion 3 program revealed a satisfactory program, and the teams assessment was that the '
program was generally satisfactor Conclusions The inspectors concluded that a similar industry event was adequately reviewed, and concluded that the licensee's quality assurance reports did not identify any previous program weaknesse IV. Enaineerina
!
E2 Engineering So,nport of Facilities and Equipment E Licensee Enaineerina Evaluations for Pioe in Similar Conditions Insoection Scoce (37551)
Following the pipe rupture, the licensee selected adciitional pipe sections to pericrm ultrasonic measurements. The licensee also reviewed several downstream pipe sections that had never been inspected and were not inspected prior to startup. The team reviewed the licensee's selection and rationale for inspection of these pipe section Observations and Findinas The licensee reviewed isometrics for the 12 lines associated with the 5th,6th and 7th feedwater heater extraction lines. The licensee looked at these pipe sections since these lines might have two-phase flow, and might be susceptible to the same failur The licensee excluded pipe sections consisting of chrome-moly steel since the wear rates from this material are orders of magnitude less than for carbon steel. The licensee selected all the piping configurations that might have changes in direction similar to the
' failed componen The inspectors agreed with the licensees' assessment and choice for inspectio Measurements made by the licensee did not find any other problems. Most of the pipe was .still at the nominal pipe thickness within the tolerance for manufacturing the pip Some sections were less than the manufacturing tolerance, but were much greater than
-
<
,
-18-
!
code minimum requirements. The licensee performed engineering evaluations of the pipe that was less than manufacturer thicknesses, and the inspectors reviewed four of these evaluations. The inspectors identified no problems with these evaluations The licensee also identified 21 pipe sections downstream of inspected components that j were not inspected. The inspectors reviewed the conditions for these pipe sections, and concluded that they were not in a two-phase flow, and should not be susceptible to the I
same erosion mechanis Conclusions The inspectors reviewed the licensee's inspection sample and rationale for not inspecting certain pipe sections prior to startup. The inspectors concluded that all the pipe sections were operable. The inspectors identified no problems with the licensee's evaluation E2.2 Isometric Drawina Issue Inspection Scope (37551)
l The team assessed an issued identified by the licensee related to the accuracy of the I licensee's isometric drawing Observations and Findinas
The licensee and the inspectors noted that the isometr:c drawings did not always note l where the sections of pipe that had been upgraded to chrome-moly steel. The licensee j evaluated this in Suggestion-Occurrence-Solution Report 99-1641, and concluded that a l design change package noted the correct changes to the drawings, but the changes i were not incorporated into plant drawings. This type of information is usually included on an isometric drawing within a month after modification is complete. The pipe involved is nonsafety-related, only one example was identified, and the licensee will take corrective actions on this item. Suggestion-Occurrence-Solution 99-1641 has a corrective action to review other past modification packages to verify that pipe material changes were incorporated onto the isometric Conclusions The licensee and the inspectors identified where some pipe material changes were not
, shown on plant isometric drawings after a minor modification. The system was not l safety related. The problem is being resolved by the licensee's corrective program as Suggestion-Cecurrence-Solution Report 99-164 .
.
-19-V. Manaaement Meetinas X1 Exit Meeting Summary The exit meeting was conducted on August 19,1999. The licensee did not express a position on any of the findings in the repor The inspectors asked the licensee whether any materials examined during the
, inspection should be considered proprietary. No proprietary information was identified.
<
!
4
_j l ATTACHMENT 1 SUPPLEMENTAL INFORMATION PARTIAL LIST OF PERSONS CONTACTED Licensee R. D. Affolter, Manager, Callaway Plant J. D. Blosser, Manager, Operations Support H. D. Bono, Supervising Engineer, Quality Assurance Regulatory Support D. G. Cornwell, General Supervisor, Electrical Maintenance M. S. Evans, Superintendent, Emergency Preparedness R. E. Farnam, Supervisor, Health Physics, Operations M. R. Faulkner, Superintendent, Security (Acting)
J. M. Gloe, Superintendent, Maintenance G. A. Hughes, Supervising Engineer, NJclear Safety L. H. Kanuckel, Supervising Engineer, Operations Quality Assurance W. O. Jessop, Superintendent, Business Planning and Development J. V. Laux, Manager Quality Assurance W. A. McElduff, Supervisor, instrumentation and Control J. A. McGraw, Superintendent, Engineering R. D. Miller, Supervisor, Radiological Wasto and Environmental T. A. Moser, Superintendent, Systems Engineering D. W. Neterer, Assistant Superintendent, Operations C. L. Nurnberg, Supervisor, Instrumentation and Control J. T. Patterson, Superintendent, Work Control (Acting)
J. R. Peevy, Manager, Emergency Preparedness ;
G. L. Randolph, Vice President and Chief Nuclear Officer )
M. A. Re!dmeyer, Senior Engineer, Quality Assurance Regulatory Support R. R. Roselius, Superintendent, Radiation Protection and Chemistry L. S. Sandbothe, Superintendent, Operations T. P. Sharkey, Supervising Engineer, Systems Engineeiing M. E. Taylor, Manager, Nuclear Engineering W. A. Witt, Assistant Manager, Callaway Plant NRC D. N. Graves, Chief, Project Branch B IN_SPECTION PROCEDURES USED 37551 Onsite Engineering 40500 Effectiveness of Licensee Process to identify, Resolve, and Prevent Problems 49001 Inspection of Erosion / Corrosion Monitoring Programs
,_
'A
.
2-62706 Maintenance Rule 71707- Plant Operations 93702 Prompt Onsite Response to Events and Operating Power Reactors ITEMS OPENED. CLOSED. AND DISCUSSED Non LIST OF DOCUMENTS REVIEWED Industrv InfoiLqating EPRI Heport NSAC-202L-R2, Recommendations for an Effective Flow-Assisted Corrosion Program," April 1999
' A_merican National Standard Code for Pressure Fiping ANSI /ASME B31.1 - 1980 Edition
= Procedurgg Revision Title EDP ZZ-01115 15 Flow-Assisted Corrosion of Piping and Components Predictive Performance Manual Drawinas . Revision Title M-23AC06 5 . Piping Isometric, H.P. Turbine Extraction to Htrs SA & SB, turbine building M-23AC07 '8 Piping Isometric,5* Stage Extraction to Heaters 6A & 6B, turbine building M-23AE01 6- Piping Isometric, Main Feedwater, turbine building M-23AF01 7 Piping isometric, M.S. Drain Tank A Drains, turbine building M-23AF02 3 Piping Isometric,15' Stg. Rhtr. Drn. Tk. A Drains, turbine bdilding M-23AF03'~ . 7 Piping Isometric, M.S. Drain Tank B Drains, turbine building
.
M 23AF04 J 5- ' Piping Isometric,1 Stg. Rhtr. Drn. Tk. B Drains, turbine building
' M 23AF09 '2 : Piping Isometric, L.P. Feedwater Heater Drains & Vents, turbine building
- .-
? M 23AF17 8 Piping Isometric, M.S. Drain Tank C Drains, turbine building
< M-23AF18 7 ' Piping 1sometric,15' Stg. Rhtr. Drn. Tk. C Drains, turbine building i
,
-
.
.
3-l M-23AF20 9 Piping Isometric, M.S. Drain Tank D Drains, turbine building M-23AF24 2 Piping Isometric, Scavenging Steam M.S.R 'B', turbine building M-23AF31 8 Piping Isometric,2" Stg. Rhtr. Drn. Tk. A Drains, turbine building M-23AF32 7 Piping Isometric,2"d Stg. Rhtr. Drn. Tk. B Drains, turbine building M-23AF38 ' 6 Piping Isometric, HP Heater Vents, turbine building M-23BM02(O) 3 Piping Isometric, Steam Generator Blowdown System, reactor and auxiliary building i M-23BM12 4 Piping Isometric, Steam Generator Blowdown System, turbine building M-23FB03 4 Piping Isometric, Auxiliary Steam System-Return, turbine building Other Documents Inputs to the CHECWORKS Computer Software Program for modeling flow-assisted corrosion in piping systems fabricated from carbon or low-alloy steels Callaway Nuclear Plant Unit 1 CHECWORKS Wear Rate Analysis: Wear Rates / Input Data Report dated August 11,1999 (Run Time 3:32 p.m.)
Callaway Records information System Calc Record Number AE-51, Revision 000 Callaway Records Information System Calc Record Number AF-38, Revision 000 Callaway Records Information System Calc Record Number AF-37, Revision 000 Callaway Records Information System Calc Record Number AF-39, Revision 000 Ultrasonic Thickness Report for the 15' Stage Reheater Drain Tank C Drains (Report Number 99-056)
Ultrasonic Thickness Report for MS Drain Tank A Drains (Report Number 99-074)
Ultrasonic Thickness Report for MS Drain Tank C Drains (Report Number 99-077)
Ultrasonic Thickness Report for MS Drain Tank D Drains (Report Number 99-066)
Ultrasonic Thickness Report for Feedwater HP Heater SA Nozzles N3-&-N5 (Report Number 99-075)
Ultrasonic Thickness Report for Feedwater HP Heater 5B Nozzles N3-&-N5 (Report Number 99 071)
y
,.
.
-4-L
'
Centralized Action Tracking System Report 57565, Response to Information Notice 97-84, January 26,1998 Assessments OA Reports Creation Date Subject SP90-062 09/04/90 Erosion / Corrosion monitoring program SP92-003 02/28/92 Assessment of GL 89-13," Service Water System Problems Affecting Safety-Related Equipment" SP93-036 ' 05/12/93 Flow-Accelerated corrosion AP98-012, 10/28/98 Quality assurance audit on inservice inspection and special processes SOS Reports Creation Date Sublect 99-1583 August 12,1999 Steam rupture and manual reactor trip 99-1590 August 11,1999 Apparent crud release after reactor trip 99 1596 August 11,1999 Fire protection system response after pipe rupture 99-1597 August 11,1999 Plant computer failure after pipe rupture 99-1603- August 13,1999 Event review team critique of pipe rupture event 99-1606 August 12,1999 Rad monitor failures following ruptured steamline l 99-1630 August 16,1999 Investigation into electrical system disturbances 99-1641 August 14,1999 Assessments and corrective actions to flow-assisted corrosion program I
i i i
! .
.
.
ATTACHMENT 2 I Table Cycle Co-58 (uCi/g) Total Suspended Solids (ppm)
Cycle 5 .0 Cycle 10 .300 REACTOR TRIP CRUD BURST COMPARISON l
.................................v............................
ggg....AAAA'
1500 -
.....**"" I l
1000
!'
,.'.*,
. ..**'...
y e a a i
!
i l
.
500 - -
'
0 - * **** *~ ~ r- i ; ,
- i ;
-10 0 10 20 30 40 50 60 Minutes Actual Boron
-
Time Release Model
-
Immediate Release Model Figure
-
l
a
.
ATTACHMENT 3 inspection Plan I inspection Procedure References 49001," Inspection of Erosion / Corrosion Monitoring Programs" 62702, * Maintenance Program" 62706, " Maintenance Rule" 71707, " Plant Operations" 93702," Prompt Onsite Response to Events and Operating Power Reactors"
'
OPERATIONS AREA TIME LINE Develop a sequence of events (time line) with operator respons . Assess licensee event review team (ERT) on operator performance and compare with our observations of their performanc . Evaluate operator response to the actuation of the fire system (verification of actuation, validity, isolation, termination, etc.).
EQUIPMENT RESPONSE Verify safety-related equipment response during the event Licensed reactor power not exceeded Main Steam pressure did not exceed MSSV setpoints after MSIV closed Rad-Monitor failures Core response with regard to boron release model following trip Verify Non-Safety-Related Equipment Response During the Event Review fire suppression system response Review failure of plant computer failure PLANT DAMAGE ASSESSMENT Review licensee's assessment of plant damag . Walk down area around break, and assess damage to other plant equipment from this break, including: Damage to other pipes and pipe support Water spray on heater drain pumps and other pumps Electrical switchgear and junction boxes i Instrumentation and Controls in the area ! Steam Dump Valves (moisture found in electric boxes to 3/4 steam dumps) !
l
.
MAINTENANCE AREA PREVIOUS INDUSTRY / PLANT EXPERIENCE Review the licensee's review of IN 97-84, " RUPTURE IN EXTRACTION STEAM PIPING AS A RESULT OF FLOW-ACCELERATED CORROSION,' and assess the licensee's action . Review previous NRC inspection (s) for erosion / corrosion related problems previously identified and licensee act!ons to addres . Review previous OA assessments on the erosion / corrosion program EROSION / CORROSION PROGRAM Verify that the correct input parameters were utilized in the plant Flow Accelerated Corrosien predictive program (CHECWORKS): Systems inputs for Line Modeling i review operating parameters (temperature, pressure, enthalpy, water / steam quality, flow rates) for the system review system modeling Materials and Design inputs for Components in Line Modeling i Size and geometry Material of fabrication for components . Design Pressure and Temperature l Valve / Orifice Opening Size 1 Wear rate assumptions if no inspection data (TDAT)
l Tcrit definition for determining service life Bending moments for structural acceptance
'
7.
l Insulation type and thicknest,
! NDE Inputs to modeling review ultrasonic volumetric inspection results for past inspections of this location Pipe inspection Program Compare CHECWORKS predictions with actual field measurement Review selection of monitoring point Review whether required, susceptible piping was included in the analytical models, Analytical models current and/or verified? Review system design, fabrication ad in-process inspection records
o
.
3- Review previous inspection results on this pipe and assess if the inspection interval was appropriate 4. Review impact of efficacy of the plant flow-assisted corrosion program for safety related system MAINTENANCE RULE Review failure of the moisture separator reheater drain system in the context of compliance with the Maintenance Rul METALLURGICAL RESULTS Review metallurgical laboratory test results to verify the failure mode for the affected piping including: metallography chemical analyses - I ' scanning electron fractography and energy dispersive analysis ! mechanical testing results fracture toughness test results profile characterization of pipe wallin the failed component Review fracture mechanics analyses to verify: material properties are as expected cr ical flaw size / thinned area are consistent with material properties and loading
!
conditions ENGINEERING AREA PIPE FAILURE ROOT CAUSE ANALYSIS A. Review Licensee Root Cause Investigation and Corrective Actions I
STEAM PIPE REPLACEMENT Review program to change out/ replace carbon steel pipe including frequency, materials, replacement criteria, et !
Approved: (oriainal sianed by David N. Graves 8/13/99)
David N. Graves, Chief, Project Branch B, DRP ,
L
[-
'
.
.
ATTACHMENT 4
)-
Photographs i
l
1 i
L
, . _
.-
-
l l
.2-i
l l
\
, p, . . . .. . .
JJ Q):
. .
'
.
N.* ..
' .k'.
.
g . - . ,.'.-
.
.4
,
i
,
0 :, $, ,. 7 . .
,
-
.. .
<- . .
__ -., 7
-
,
.
.
/ .
.y,, - .
l
, .
., -
- * . mg +y
- w3, ;
-
- -
. .
%yAn
.
,
,
, .,
.
, .m .
.
'y
< .. - -
.
-
Y- ~
-
l:,; 1
, . , ..
. . . -
. . . *.. *1
.
-
ky yy
-
+
, ,<%Q M
,
,
-
- .
, .. ,,
7 _
'
..,.ir-
, ,
.
. -
..
'
,
,)'
.
, , ,.
. '
'
v ~.a.__..._ _ M/
._
-
.. v .
..
,. g UPSTREAM SECTION OF PIPE AT 45 ELBOW
- ._
..
.
-3-
, . NC.-;
-
f s '
. .
j .. . . .
, ,,,
'
' .
. .
i, , '; -
- -
..
, % ,
ry x,,.. . . .
.
..
g3, ..g...
6..r ,11 ,y, .
.
, , - -
.
.
u a.c , ,. ..
..
, .
.
.,
. s, i X ,. . .
.
,
,
y t. - x,'. .
.
_
,/ , : v -
./ s.y - e 1
. . . .
.
,
.
.
.
- ' ..
-
.
{%,, %_,_ . .
.
,.
-
we ,
,
.
i(>e
- '
_
f. . .
.
.
. . . - ..
. .
, -
2 . ,.
,
-
-
e ,-
. . . . .
.
s v . j
-
. . . . . . :....,..
.
.. .
-.... -. . _ Y--.
Downstream Section of pipe, showing deflection
_ . - - - . .. - - - --
-
.
.
l I-4- l l
l l
I l
l j+>,,
. . c .,. ..
-'. .
.
.; c a: \
'
,
\
1
,- l
-
k meusk .i s..- "
.
>
Closeup of downstream pipe end
.
i
.
a 5-
l '
, ,
,
, , ,,. .x .,,my-
.., . . .
.
,_
o
- (.p .
c . O).Q*(".flmlly,(; v;t f.f'.- o g gg m j$h
< .
- -
-
. ,. ; ::': y; y,phywgpMg ywyn,4,
,
, ':- < -: . f : , = , .n; q;f t y;.,,,
2 .4 p w;r;y;g4 9 p,,.
.
, , ,
a; ; % yuy ,
uq
,
, ..i .
r;,- ny,,1p-;f e
.
..
se @ w.n.wn ., 1
.,
i i
v y m,<1:
. ..
. ...,jw v**. a;y g. n?,
a n g:? +e, g f. , J;,a w ..y
-: $% f 4f> rs; ?; **A' vQ*p.J .oyesz,q3',m
,
, w;fiyy
}.f::.;, y y :/ *
- ;
' '- *
,
1 %qn . Zi-?) yyE/. ; $;&T""
-
r Y
'
.
. , nin .lM.r;.:w,Y %%
w w t$- MQ:'s a r.y G-
1 gA >g y . , . 1 f g,y ,y!,s at* ! a ,# h.!. '
e, ..% s'Y ' -
. . y' , .
,- T. . p f, ,49,
,
pp: . < .
II JZhf$5fN?fl: lO+$
M1
/vy,{
(n@. x pe-(x%$ c s p
-,-
g. g
<
. ,##~qs aa u gn,n+g.,.3 a , .;p s
.bg-d:n ge gy : M. c k.w wwcxdMc , aw.kn sw;r, . ..e'7 M . .,,da ., &a: o u..S W*3
,
. . . .
w,r , ,e ,-
,f~ ;
[g
'
E .,
\ c
~,#
.%,,Y' wl ,# f
.. .
,lr4%g a
), e } ~ i,
. ',kk.igf ,', l e,ct. .
.' a- ai h
-
1. a , ;
, ; , . ,N,. ' !
3 -s 3,, 3 /
.
a .
,
i t
Upstream Edge, showing irregular edge on top of pipe ll l
l
!
i
i
!
. . . . .. . .
.
.
l y ... : , c .,. . .
. ; ,- ;
.
.$
v..-
.1
,
, . . . . .
. . : ; ,
.
." 3 a
2e w . : ...
'4. * , . ,m g
..
,
- .
-
- .
~
'
.
' '
.
'
e 49 .
,
,
.. ; + . -
x ,
. - a i
.;. j
V. f ' i * - : -'
'
,
.
,
-
1 -
.
.- , - .
. ~
$q,.. ;.J
~
. .
,
, <
- , . , .
3,. s J- ' , ' ,
>
4'
l3t ,j ',
.
. .
' , '
<.
.
.
-.. .y -
- . .. . .
L l
..> ; g
.- _
.
.
.
,}' ' ' ' ' . 4 -
,,,5
, ,, . ('
.; - ,
,
, ...
,
+ i ,- -
a
.* . J%pn! Q \
- .
. ~;,y,:4
- = g, - ... , (- , .
.,; .
- ...
.4,. . .
-
e :
Y -
'
,, .
.
-
.g
.'.
.,
'
.
.' ., .
~ v .
,.,
. .
..
- .
> s ..
.,
O'
. .
.
.
. . ~.
. *. );4 s *
,
.
.
-
.-
.
, . -
Downstream Pipe Section, after remova Pipe tear was along pipe clamp edg l
.
<
,
l-7-
,.. .
.g .n .u..,. -~s ;;:n/ . w, :,.wn s,:qa m.J . 1. . ; ,z. .,,qV4
.
.. . .
-%w n .tw ,r n.~,9;;,, q ...e
,,
w n. p; gg( .;my
-
., :.
<
y ,., ,
.
-nfp;> . . : ._, ,
-* * l ~* iY f.R' .?':.,s y . ' :-
3;.3 7;ll+ ~,;f h_,4 %,hE.O*/.R pf.%f f ?(9 % ,i fa'
,
, , , .
i
w ..,.-. .- c
.
.a. m ~ w n & :3, g y'* _%
yyfs
.
',,
.qrs yyy>tyxjgggg?. i~; - ::4 %Ykygx.sykT n ;,.e,v. a <; r >{
c - .,
. . .
..
'
1 .%v
.1 . a. . a:u~u a =,e,wr
- t .
-
,
T -
m;: . ' * Ay*C;*:q::,**c^;:
.M
.n , 7 % ; ij,,. ; .. .. ;.. . ,,e . .? ; ,,. at 6s,... ,
, .:
(: . .
,.,o ...
.,
<_ :;. - ' .
, ., . ,,
- .
, .; .,.4
_._'.---*-~w-
.,
,
Y g
_
' . s . ., -
_ ,, , , .. .
+ + .
.
., .s .
..
' ~
. p!% . '
. ;c e
.s .
.. .a.; ,_....{7 ..,.
.
.t
.,, h ., '. '4 .
. -. -
.i
~, '
.
f't
?
p.; ' . i, ' . Q ' .
_
y
, . '..?p, w ,
.. s -
,~,. .,
s ( p. X , _. _ ,. s .r
.
2, .. y,.y , e ,
., . . , .
s
.- ,... i '.
~
.
,
' ' b. w . ' ,.
/ . Y se e..,
e'. , '%
.*' ., .
,
.. .' ^.. ,
s g,..'_
., ,- .- : , . .
.=-.
y. . .
.
-
e. g ,~, :c , .
L ..
,
., : _
- ,+ .
e,
'
n
-.
w
.#ra- .
. . , ,
.
-
.g.-
.*
w, *
.
.
. w,
.
spuut
.
.
.,.#%t.* 1' . . (96,1 .
.
.
.
' .
! ' . .:- , . . . ,
, . .. 4, . .p. ~, . c f,? ' ...L., .
A -rl_ : W
.
. c..p~.- .
,
)A.h. .1.' 6.<
- . ** . . .
'- 4% *
.>
'
.q , . .
i
.. y,' ",g' a r
.
. 's _
. *
,k * -
'; *
- .% ; . ,. M). : ~y.,'o- g,& T
,
g_
'
'
g .: ,.
_;
J , . ; AtR; Oy ,.~~ f ' g + i-7
'
,_ G ; .. .J .
Y
-% ,,;,.39 g g j;pp
'
gg;%g Pipe " Unwrapped Section" Interior !
Bottom Edge in Picture was by Pipe Clamp
<
.. --
.
..
.
.
,
-8-g .
.
$ ' .
'- g
.. ,
.
. .
j
,. .
.
. ..
I ,
I 4
., _ .
'g
- . . ,
- .- . -
.. . l
.. . ,
.
a ' ., rg s .i s.,, . .
~
-
., y,, 'h ...-
. .. ; - :gr- '
.,
. .
,.y - .
g
,. . .. .
w.- , .
.
.
. , ,, . t . .- .. .
. .
.
.. .. .
r> .:.
.
3-
.-. .< - -
g . 4
,
- .i t i,- .'
.
y '9 **
<
ste-
-
- < 2, *
..
,,.
.,
, . ,
,..,,,.<f +. . ;. .y
. '.,.<,., gu
.
4 t , . . , . , ,
'
~
4 ,-
.
.
... . , . .
,, ,
P.vgi = . 4
'
,. . . ,
. ,
- 'x* ,;f ,.
_. . , , , ge trss '- '
C'
21 ._:. 5 y,
.'. .,p f ..- l ; -M-f_, . ,
'
g w . .:..%. 9 .
_
A)u -
?
. _ ,. -
%
s$lY&l%. . $
m- '" a},.g.r>..s :. , %-~ ' w.c w:; u : a e m a c y =; w ;,1e; .
R }% 42 % M ; .:
Lag *
m:e 7 u .
y N73 .
^
3 ,qjn;L:
v c
_ -,'
..:a+M;Oig%w'.,'i,,,4"% ' d
.,
Og;; 4
'
.-
kfb;qgM .n q' bM =:.. end.-&+;.yQ:r+:a,-
,
a* :
'.'p
.
< --*
. - - ~ .;.,
ff/g% : Win.t:4F : W W "T-. ~ " ';* -
- u;.fkMv ' ?rw .
.. . .
'
--
,
^ ..
.
,
.
, . . .
._ <
. .. .
.
Pipe " Unwrapped Section" Exterior Bottom Edge in Picture was toward elbow
.
.
. .
.
..
.
.
.
nr~ # ^V g.v: mm
-
_
,.
> ;;,, ,,,e.,.,x*
3 .g p, mmy
, *4
- ,. f a
~{::. ' . , , ' ^ ' ?'
.
., y +;[ ry, - . _,..q ?}*
g
_
i,
.,g'* e -} 'N4 ,
2,
-
(-
? '
.
'
u.ws
, , ,e ('- 7, ,
.,,, l
., f -
'&'s (
if,;${d
,
- -
i'
,.,
..
y+3 .
'
T.,.e
...
y r
,.,
g7
,
1p e .. 3
-
.
.
s bb
-
a
,
yi ,
2 < t
. i a
', u
- '# ;
4 c . +:
' #
A;;. J
.
-
- j r
y ,yn)' '.g g .
- .
' g/gg, 4. , 'Wm..dr ;g 3%
.
,
- .
/ ,
%gggyrg,. a,..qs.wngssE,sk.;
- .
,, _ /
< j
e v %u- , !
!
Inside view of 45 Elbow, tooking upstream !
\ %
.
.
.
<
l
'
- 10-
-
.
.
,,
.
.
.
.. .
. .
. .
. .
s b
-
.
w ,
- .
.
.
-
s ,
'
. . . .
'
g
-
-
.i. . .
.
.
. ..
%, .
'
-
.; . = '!,
y.y'q f)y; h;.
.
. .
3f1;
_
-
-
.. .
.
mh :.+> [, .. . .- .
_
.-
-
>
- 4.g. Y,r .
-
.
s f. . - ,
- -
x ,g .
-
- - -
.
.
,.. 1.rc ,
.
3, ; . , , <.
.
- . . .
- /Esg.g.?V g. , $,46:;.e Wj.h:,',' - 'g# y
_
s ...,,7'
-
..
- . .- .
hlk f 0 sh
%gg- swam
-
i .
- ,; . . ;
.
-
. .
'- +-
Q:.=A:E., aq l
- ,???W W st .,
,.,;..., - . '
2, l
,
~~*jC ;,,, p .
~,,."*t ~f'% ~
- ['h!ye?$$$
a.s. ca:-
s,m cu c -
,
Inside view of 45 Elbow, looking upstrea Closeup of pipe wall.
t