ML20203D782
| ML20203D782 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 12/12/1997 |
| From: | Rainsberry J SOUTHERN CALIFORNIA EDISON CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| Shared Package | |
| ML19317C790 | List:
|
| References | |
| NUDOCS 9712160288 | |
| Download: ML20203D782 (31) | |
Text
,
.c gl
]
g EDl SON lia"r'-, 1 l
n uw nmwun u n .,.- }! l _
December 12, 1997 U.S. Nuclear Regulatory Comnission Attention: Document Control Desk
} i Washington, D.C. 20555 Gentlemen: .
Subject:
Docket Nos. 50-361 and 50-362 Mechanical Nozzle Seal Assembly Code Replacement Request for Relief from 10 CFR 50.55a San Onofre Nuclear Generating Station, Units 2 & 3 -
~
z
References:
- 1) Let ter from Mel B. Fields (NRC) to Dwight E. Nunn (SCE), dated November 20, 1997; Subj ec t : Mechanical Nozzle Seal Assembly l
l
.j I Code Replacement for San Onofre Nuclear Generating Station, _
[
Units 2 and 3, Request for Additional Information (TAC
- Nos. M99558 and M99559) .
- 2) Letter f rom J. L. Rainsberry (SCE) to Document Control Desk '
(NRC), dated July 11, 1997; Subj ec t : Docket Nos. 50-361 and .
L 50-362, Mechanical Nozzle Seal Assembly Code Replacement , l ,
Request for Relief f rom 10 CFR 50.55a, San Onofre Nuclear (
E Generating Station, Units 2 & 3 .
j h
f This letter provides, as an enclosure, the response to the November 20, 1997, l ji NRC request for additional information (Reference 1) conctrning the mechanical [ .I
[
nozzle seal assembly joint (MNSA). This information supports the Southern p'p
- California Edison Company (SCE) request for NRC approval to use the MNSA at j !f the San Onofre Nuclear Generating Station (SONGS) made by the July 11, 1997, j 1 SCE letter (Reference 2). The American Society of Mechanical Engineers (ASME) y
- main Code committee has Interpreted the design and use of the MNSA to be in '
compliance with the ASME Code. The er. closed information includes declassified f
.I
~
documents from ABB Combust,an Engineering (ABB CE) The MNSA addressed in l this submittal is patented by Combustion Engineering, Inc. l is 5,619,546. Thepatentnumber,p'; -
'l 3 1 Additionally, this letter 1) clarifies information that was provided by the ]
original relief request (Reference 2) and 2) revises our original request to 1 include three additional locations where the MNSA could be used. These h locations are the steam generator cold leg channel head instrument nozzles, 1 l;
- reactor coolant system (RCS) hot leg sample nozzles, and RCS hot leg pressure ta p differentialtransmitter(PDT) nozzles.h~'l,,o ( , , . , , ) '
san onohe Nuaw venmang woon F
~
P O Na W san cienwnw. ( 4, w4 m u
, g As {Il{lig 1
I ll f: !{
i 1 \
9712160288 971212 PDR ADOCK 05000361 P PDR ,,
( .
Document Control De~,k !
INTRODUCTION As a-result of findings during-the most recent Unit 3 refueling outage, SCE is taking-an aggressive approach to address alloy 600 instrument nozzles at SONGS. Each-RCS instrument nozzle has a typical outside diameter of approximately one-inch and penetrates the loop piping, steam generator shell.
-_or pressurizer shell'.- SCE has established the goal of replacing all reactor coolantsystem(RCS)looppipingalloy600nozzleswithalloy690nozzlesby the completion of the next (Cycle 10) refueling outage for each unit.
Alloy 690 was chosen for the replacement nozzles since it has substantially greater resistance to stress corrosion _ cracking. This replacement plan includes all alloy-600 resistance temperature detector (RTD) nozzles, pressure differential transmitter (PDT) instrument- nozzles, and sample nozzles in the hot leg and the cold-leg of the rnctor coolant system piping. The instrument and sample nozzles will be r'ferred to simply as nozzles throughout the rettiinder of the letter and enclosure.
Since each unit is scheduled for a mid-cycle outage this winter, SCE intends to use these outages as much as practical to support this replacement goal.
During these mid-cycle outages, reduced inventory (i.e., mid-loop) operation is planned to allow nozzle replacement and access to the steam generator tubing for inspections. It is planned to replace a significant portion of the
-remaining alloy 600 RCS locp piping instrument nozzles with alloy 690 nozzles during the mid-cycle outages, nearly one year before the refueling outages.
By the completion of the mid-cycle outages, approximately 55fs of Unit 2 and 66fs of Unit 3 RCS loop piping nozzles will be replaced with alloy 690. The remaining alloy ~600 RCS loop nozzles will be ,eplaced during each unit's next refueling outage.
Should mid-cycle inspections identify leakage in one of the RCS nozzles located below the mid-lcop water level, SCE intends to utilize a MNSA. The MNSA would'be used as an interin: repair instead of replacing the nozzle because nozzle replacement would require a full core offload. The MNSA design is a practical alternative to the time and radiation dose intensive effort of reactor disassembly, core offload, reload, and reassembly. Should an interim MNSA be used, it will be removed and replaced with an alloy d90 nozzle during
.the next refueling outage.-_The installation and the removal of the MNSAs will not interfere with tre implementation of the permanent nozzle replacement.
The mid-cycle reduced inventory condition has been evaluated for its risk implications. SCE has determined this mid-loop RCS water level condition will
__s
Document Control Desk result in an acceptable level of risk consistent with previous refueling outages and is within the overall 1998 core damage risk goal (6E-5/ year) established by SCE for San Onofre. This risk management goal includes the at-power and shutdown core damage risks due to internal initiating events for the entire year.
Further, during the mid-cycle outages, SCE intends to install five MNSAs on the only pressurizer instrument nozzles that are still alloy 600 (the two bottom head nozzles on both Units 2 and 3, and the lower side shell nozzle on Unit 3). SCE plans these pressurizer MNSAs to be permanent. The MNSA provides a straightforward alternative to a welded replacement. SCE believes the MNSA is a prudent, appropriate repair approach for the pressurizer because the welded replacement of a pressurizer nozzle is difficult to perform for the following reasons:
The welded replacerrent consists of time consuming welded pad build-up process The bottom head nozzles are located in a very restricted area that is difficult to reach and to work because of congestion created by the pressurizer heaters The risks of foreign material entry into the RCS are eliminated by the MNSA Additionally, the MNSA installation is performed with substantially less radiation exposure (estimated to be 3-4 person rem savings per nozzle).
1 1
l Similar to the pressurizer nozzle approach, SCE has c;ntingency plans to install MNSAs on the steam generator cold leg channel head instrument nozzles.
Fach of the steam generator primary cold leg channel heads contains four instrument nozzles. Because these nozzles operate at RCS cold leg temperatures, they are not expected to develop leakage. Even though no leakage is i.xpected. SCE has 4 MNSAs available, as a contingency plan, for the 16 steam generator cold leg channel head instrument nozzles. Should leakage be detected, a MNSA would be used for the repair, and plans will be developed to address the remaining cold leg channel head instrument nozzles at the next l refueling outage.
l l
l REVISION OF INFORMATION IN THE ORIGINAL RELIEF REQUEST
.J
- Document Control Desk -4 As-quoted below, SCE's; original relief request (Reference'2) called for disassembly of-'the Mtv5A bolted connections to complete the 10-year In-service Inspection _(ISI) requireme_nt, "Section XI requirements applicable to the MNSA are the system leakage test -at the end of- each refueling outage (Table IWB-2500-1, .
Category B-P, VT-2 visual examination with acceptance _per IWB-3522) and the bolting examination at the end of each 10-year examination cycle
_(Table IWA-2500-1 Category. B-G-2, VT-1 visual examination with acceptance per IWB-3517). Bolting examination rules mean that the MNSA bolted connections will be disassembled for the 10-year examinations if they are located on the pressurizer; however, the MNSA's located on the pipes would only be disassembled for inspection of the bolts if the section of pipe containing the MNSA is scheduled for examination during the pai ticular 10-year inspection. Because the "J" weld would no longer q be the pressure boundary, the "J" weld would no longer be included in l the ISI program."
l SCE has concluded disassembly is_not required to meet the requirements of Table IWB-2500-1, Examination Category B-G-2, " Pressure Retaining Bolting, 2 inches and less in diameter," footnote (1), restated below:
" Bolting may be examined (a) in place under tension, (b) when connection is disassembled, or (c) when the bolting is removed."
Therefore, SCE is completely revising Item 3 of the Relief Request
" Inspection" as follows:
Section XI inspection requires 2 examinatfor.s applicable to the MNSA.
- 1. The system leakage test visual examination at the end of each refueling outage, in accordance with Table IWB-2500-1, Category B-P, VT-2 visual examination with acceptance per IWB-3522, and
- 2. The bolting examination per Table IWB-2500-1, Category B-G-2, VT-1 visual examination with acceptance per IWB-3517.
-J
F Document-Control Desk- - - It should be noted that one of'the acceptance criteria for the visual.
- examination of-the MNSAs is "no visible leakage." Therefore, if any leakage should be fout.d the MNSA would be disassembled and inspected, and appropriate repair, _ replacement, or reinstallation would be performed.
Subsequent -to the submittal of the relief request (Reference 2), SCE has completed a risk analysis of the outage activities. Based on this analysis, SCE determined that midloop water level should be increased six incnes above the _ level originally planned. This change raised the level above the ;
horizontal centerline of the RCS piping. Therefore, SCE has . identified the RCS '
hot leg sample nozzle and the RCS hot leg PDT nozzle, located on the .
horizontal centerline of the RCS piping, es potential candidates for the installation of a MNSA in the event of leakage. As discussed above, the contingency plan for the steam generator channel head instrument nozzles requires the installation of the MNSA. Therefore, this letter adds these nozzles to the original relief request (Reference 2).
SUMMARY
SCE requests NRC approval:
- 1) to use the MNSAs as an interim repair on the lower hot leg nozzles, the pressurizer instrument nozzles, and the steam generator channel head instrument nozzles by January 23, 1998, to support the Unit 2 and Unit 3 mid-cycle outages, and
- 2) to use the MNSAs permanently on the pressurizer instrument nozzles and the steam generator channel head instrument nozzles by September 30, 1998, to support planning for the Unit 2 and Unit 3 Cycle 10 refueling outages.
+
a - -, = , + - - - - - -- -
. _ . _ _ . . __ _ . . _ . . ~ . _ . . . . - - _
- Document Control-Desk :
If.lyou have any questions on this subject, please call me.
' Sincerely,.
4 '
%, [4 6-i Enclosures-cc:- - E.- W. Merschof f, Regional Administrator, NRC Region . IV-K. E. Perkins, Jr., Director, Walnut Creek Field 0ffice, NRC Region IV M.-B< Fields, NRC Project Manager, San Onofre Units 2 and 3 J. A. Sloan, NRCLSenior Resident inspector, San Onofre Units 2 & 3 1
3 4
J-D I >
v 1
a
- - . ~ __
- n. .
+
ENCLOSURE T11E SOUTilERN CALIFORNIA EDISON cvMPANY RESPONSES TO TIIE NRC'S REQUEST FOR INFORMATION MECIIANICAL NOZZLE SEAL ASSEMBLY (MNSA) l
1-Mechanical Nozzle Seal Assembly (MNSA) Enclosure F
- 1. NRC Request: (BACKGROUND)
Provide the total number of proposed replacement instrument nozzles and their detailed location on the reactor coolant system (RCS) hot leg and the pressurizer.
- 1. SCE Response: -
To be complete, SCE has summarized below the nozzle replacement plans and the MNSA installment plans for RCS piping nozzles, pressurizer instrument nozzles, and steam generator channel head instrument nozzles. The outages discussed below are currently scheduled to begin as follows:
Unit 2 mid-cycle - January 24,1998 Unit 3 mid-cycle - March 7,1998 Unit 2 Cycle 10 refueling outage - December 1998 Unit 3 Cycle 10 refueling outage - March 1999 Unit 2:
RCS Cold Leg e The 12 RTD nozzles are planned to be replaced with alloy 690 nozzles during the mid-cycle outage.
RCS Hot Leg:
e All 12 hot leg nozzles located above the midloop water level are planned to be replaced with alloy 690 nozzles during the mid-cycle outage.
- The 20 hot leg nozzles (RTD, PDT, and Sample) located at or below the midloop water level are scheduled for replacement with alloy 690 nozzles during the Cycle 10 refueling ouiage. A MNSA will be installed as an interim repair ifleakage is detected in the mid-cycle outage. These MNSAs would be removed and the nozzle repaired using alloy 690 at the next refueling outage.
Pressurizer:
-e Two lower level nozzles in the bottom head of the Pressurizer will have MNSAs installed during the mid-cycle outage. All other pressurizer instrument nozzles have presiously been replaced with alloy 690 nozzles.
i
2 Mechanical Nor.zle Seal Assembly (MNSA) Enclosure Steam Generator Channel liead:
- Each steam generator contains 4 cold leg channel head instrument nozzles. Should leakage be found a MNSA will be installed and plans developed to address the remaining channel head nozzles.
- For the mid-cycle outage,4 MNSAs have been procured as a contingency. Those not used in Unit 2 will be available for Unit 3.
t e SCE intends to use this plan as a contingency for future outages.
Unit 3:
RCS Cold Leg:
- Eleven RTD nozzles are planned to be replaced with alloy 690 nozzles during tiie mid-cycle outage. One RTD nozzle was previously replaced with an alloy 690 nozzle.
RCS llot Leg Nozzles:
- Eight hot leg nozzles located above the midloop water level are planned to be replaced with alloy 690 nozzles during the mid-cycle outage. The remaining 4 nozzles have already been replaced with alloy 690 nozzles.
- Filleen hot leg nozzles (RTD, PDT, and Sample) located at or below the midlo., water level are scheduled for replacement with alloy 690 during the Cycle 10 refueling outage.
The remaining 5 nozzles have already been replaced with alloy 690 nozzles A MNSA will be installed as an interim repair ifleakage is detected in the mid-cycle outage. These MNS As would be removed and the nozzle repaired using alloy 690 at the next refueling outage.
Pressurizer:
- Two lower level nozzles in the bottom head of the Pressurizer and one RTD nozzle in the side shell will have MNSAs installed during the mid-cycle outage. All other pressurizer instrument nozzles have previously been replaced with alloy 690 nozzles.
Steam Generator Channel licad:
- Each steam generator contains 4 cold leg channel bead instrument nozzles. Should leakage be found a MNSA will be installed and plans developed to address the remaining channel head nozzles.
3 Mechanleal Nozzle Seal Assembly (MNSA) Enclosure
- As stated under Unit 2, four MNSAs have been staged as a contingency. Additional MNSAs can be ordered if the need arises during the outage.
e SCE plans to use this plan as a contingency for future outages.
- 2. NRC Request: (DESCRIPTION 1.)
Provide a detailed listing of all instrument nozzles on the RCS hot leg and the pressurizer where the "J" welds leaked, and the repair and inspection procedures of each weld. Provide the results of metallurgical examinations showing that the cracks were caused by stress corrosion cracking.
- 2. SCE Response:
SCE believes there was only one instance of"J" weld leakage (1995, Unit 3, pressurizer steam space nozzle). All other alloy 600 instrument nozzle leakage is attributed to axial through-wall cracking. For your information, all SONGS alloy 600 instrument nozzle leakage history is provided in chronological order below. The repair method for the "J" weld leakage utilized a full length nozzle replacement and a complete ID "J" weld using alloy 690 material. This technique is referred to below as a full length nozzle repair. It v as used only on the pressurizer.
Another repair technique was developed whereby the repair cc aid be made from the outside of the vessel or the pipe. This method significantly reduced the r tdiation exposure. In this method the original installed nozzle is partly removed. A replacemen nozzle is inserted from the OD of the pipe or vessel and a complete OD "J" weld is made using mioy 690 material. When this technique is used on the pressurizer or the steam generator a weld pad buildup is required. This weld pad is not required for the RCS piping nozzle repairs. This technique is referred to below as a half nozzle repair.
SCE perfonns visual inspections for leakage in accordance with ASME Section XI Code,1989
- Edition with no Addenda, in addition to inspections required by the Code, SCE has developed a careful visual examination procciure to inspect the bare metal adjacent to all alloy 600 RCS penetrations. These inspections are performed at the beginning of each outage and repeated during startup operations in accordance with the enclosed inspection procedure SO23-V-8.16.
SCE has performed root cause analyses for three instances ofleakage from alloy 600 nozzles.
Metallurgical examinations were performed as documented in the following two reports:
e SCE internal memorandum from C. Chiu to J. T. Reilly, dated November 5,1986;
Subject:
Corrective Action for the Unit 3 Pressurizer Nozzle Failure in March 1986
~
O
-4 Mechanical Nozzle Seal Assembly (MNSA) Enclosure e SCE intemal memorandum from C. Chiu to M. P. Short and L. O. Cash, dated June 30, ,
1992;
Subject:
Root Cause Evaluation 92-019, SONGS 3 l>ressurizer Level Instrument Nozzle Leakage
-All three root cause reports are enclosed and concluded that the leak path developed through the instrument nozzle wall from an ID initiated crack, not through the_"J" weld.-
Allov 600 Nozzle Leakage and Repriir History 1986 Unit 3 The first through-wall instmment nozzle leak at SONGS occurred in 1986 when a pressurizer steam space nozzle was discovered to be leaking. A root cause evaluation (Encired 1986 Report) was performed which identified that the source of the leak was intergranular cracking caused by PWSCC. The evaluation concluded that the high yield strength of the material (heat 54318) made it highly susceptible to PWSCC. All of the remaining penetrations fabricated from heat 54318 were scheduled for replacement.
There were two Unit 3 pressurizer steam space nozzles, one Unit 3 lower shell nozzle, and one Unit 2 lower shell nozzle made from this heat. Replacement activities viere completed in 1988 using alloy 600 material that was expected to have good characteristics with respect to resistance to PWSCC. These replacements were full length nozzles welded to the inside of the pressurizer in accordance with the original design.
1992 Unit 3 The next nozzle leak was found in 1992 during the Unit 3 Cycle 6 refueling outage in one of the pressurizer steam space nozzles which had been replaced in 1987 as a result of the first nozzle leak. During the inspection and repair activities inside the pressurizer, cracking was identified in two additional nozzles by PT exams (Enclosed 1992 Report).
All four pressurizer st:am space nozzles were replaced with alloy 690 nozzles. However, the weld material used was similar in composition to alloy 600, since the alloy 690 weld material had not yet been accep'ed by the ASME Boiler and Pressure Vessel Code.
These replacements were full length nozzles welded to the inside of the pressurizer in accordance with the original design.
1992 Unit 2 During a forced outage of Unit 2 in 1992, two pressurizer steam space nozzles were found to be leaking. A half nozzle repair was installed on both of these nozzles using an alloy 690 nozzle piece welded to the exterior of the vessel.
1993 Unit 2.
In 1993, during the Unit 2 Cycle 7 refueling outage, all four pressurizer steam space nozzles were replaced with full length alloy 690 nozzles using weld material similar in composition to alloy 690, which had by this time been accepted by the ASME Boiler and Pressure Vessel Code. The first evidence ofleakage from a hot leg nozzle (2PDT0978-1)
. . ~. - . - . .-_ -. . _ . . . . . . _
S-Mechanical Nozzle Seal Assembly (MNSA) Enclosure was found during this 1993 refueling outage; This nozzle was repaired using a nozzle .
piece fabricated from alloy 690 and welded to the exterior of the pipe.
1995 Unit 3 In 1995 during the Unit 3 Cycle 8 refueling outage, two hot leg nozzles and one pressurizer steam space nozzle were found with evidence ofleakage. During the inspection'and repair activities in the pressurizer, surface indications were Mtified in . !
the weld material of two steam space nozzles, one of which was through-wall crack. This i is the only suspected "J" weld leak. Since these nozzles had been previously replaced with full length alloy 690 nozzles and defects were identified in the welds (which were of a composition similar to alloy 600), SCE concluded that the leak path in this case was through the "J" weld. All four pressurizer steam space nozzles were replaced with full length alloy 690 nozzles using weld material similar in composition to alloy 690. The two hot leg nozzles were repaired using the alloy 690 half nozzle technique.
1997 Unit 2 During the return to service of Unit 2 after the Cycle 9 refueling outage in 1997, the pressurizer side shell nozzle (2TE0101) was found to be leaking. This nozzle had been proactively replaced in 1987 with alloy 600 material as a result of the first nozzle leak in 1986. An inspection was perfonned which located the crack using eddy current and ultrasonic testing. A root cause evaluation determined the most likely cause to be PWSCC. The text of this root cause report is provided in the enclosed pages from Action Request (AR) 970300092-02;
Subject:
Temperature Element 2TE0101. This nozzle was repaired using the half nozzle repair technique in 1997 with a weld pad build up on the exterior of the pressurizer using alloy 690 material.
1997 Unit 3 At the beginning of the Unit 3 Cycle 9 refueling outage, five nozzles (four hot leg and one cold leg) were identified for repair. As the unit was being returned to service, four additional hot leg nozzles were identified for repair. All nine nozzles were repaired using the half nozzle repair technique with alloy 690 material.
- 3. NRC Request: (DESCRIPTION 2.)
Provide a detailed listing of all events where a "J" weld cracked completely and the instrument nozzle was ejected. Provide the results of the metallurgical examinations of these welds State
- . the date when the nozzles were last inspected prior to the occurrence of each event, and the repair procedures which were followed in each case. State if these events were reported to the NRC.
- 3. SCE Response:
L There has been no such event at SONGS.
l i
6 Mechanical Nozzle Sert Assembly (MNSA) Enclosure
- 4. NRC Request: (DESCRIPTION 3.)
At each nonje location, provide the latest results of the ISI, and the date when performed.
- 4. SCE Response:
Unit-2. Inservice Insnection date and results:
The nozzles were inspected in accordance with ASME Section XI Insenice Inspection, VT-2 System Pressure test, performed on March 17,1997; no leakage was detected. Inspection results were satisfactory in accordance with ASME Section XI requirements.
Unit-3. Inservice Inspection date and results:
The nozzles were inspected in accordance with ASME Section XI Inservice Inspection, VT-2 System Pressure test, performed on July 16,1997, no leakage was detected. Inspection results were satisfactory in accordance with ASME Section XI requirements.
- 5. NRC Request: (DESCRIPTION 4.)
Provide all dimensions and the materials of all components shown in Figures 2 through 7.
- 5. SCE Response:
Figure 2, Existing Ilot Leg RTD Nozzle The piping material is carbon steel (SA 516 - GR 70) with a stainless steel liner, The internal "J" groove weld is inco-182 and the nozzle is Alloy 600. The nozzle shank dimensions are smaller than %"_ schedule 160 piping. The nozzle, bore and weld dimensions are provided on enclosed drawing SO23-923-19.
Figure 3, Existing Side Pres.eurizer RTD Nozzle The vessel material is low alloy steel with a stainless steel liner. The internal "J" groove weld is Inco-182 ad the nozzle is Alloy 600. The nozzle dimensions are consistent with one inch schedule 160 piping. Dimension details are provided on enclosed drawings SO23-919-12 and SO23-919-16.
Figure 4, Existing Bottom Pressurizer RTD Nozzle The vessel material is low alloy steel with a stainless steel liner. The internal "J" groove weld is inco-182 and the nozzle is Alloy 600. Dimension details are provided on enclosed drawings SO23-919-13 and SO23-919-16.
Figure 5, llot Leg RTD . Mechanical Nozzle Seal Assembly The llot Leg RTD MNSA details are provided on enclosed drawing SO23-411-57-6 and
7 Mechanical Nozzle Seal Assembly (MNSA) Enclosure component details are provided on enclosed drawings SO23-411-57-7 through 10.
Figure 6, Side Pressurizer RTD Mechanical Nozzle Seal Assembly The Side Pressurizer RTD MNSA details are provided on enclosed drawing SO23-411-57-5 and component details are provided on enclosed drawings SO23-411-57-7 through 10.
Figure 7, Bottom Pressurizer Mechanical Nozzle Seal Assembly The Bottom Pressurizer MNSA details are provided on enclosed drawing FO23-411-57-4 and component details are provided on enclosed drawings SO23-411-57-7 through 10.
In addition to the requested information, details of the Steam Generator PDT, Hot Leg PDT, and llot Leg Sample MNSAs are provided in the following enclosed drawings:
-S O23-411 57-20 E-MNSA-228-014, Sht. I of 2 / Steam Generator PDT MNS A S O23-411-57-21 E-MNSA-228-014, Sht,2 of 2 / Steam Generator PDT MNSA SO23-411-57-2f E-MNSA-228-015, llot Leg PDT MNSA S O23-411-57-26 E-MNSA-228-016,110t Leg Sample MNSA S O23-411-57-22 E-MNSA-228-013, Sht.1 of 3 / MNSA Details S O23-411-57-23 E-MNSA-228-013, Sht 2 of 3 / MNSA Details S O23-411-57-24 E-MNSA-228 013, Sht. 3 of 3 / MNSA Details
- 6. NRC Request: (DEGRADATION MECilANISMS a.)
Provide your basis for stating that the boric acid will not be replenished, thereby limiting conosion of the carbor, steel.
- 6. SCE Response:
If a through-wall crack were to develop within or outside a nozzle "J" weld, primary coolant containing boric acid could flow into the annular space between the nozzle and the penetration.
Yhis would expose the inside surfaces of the nozzle penetration to a borated water erwironment.
Since the pressurizer and steam generator materials are low alloy steels, there is a potential for boric acid corrosion (BAC) of these surfaces. Similarly, the hot and cold leg RCS piping is carbon steel, so it is similarly susceptible to BAC interior to the penetration. The design of the MNSA device is such that any leakage that occurred through a crack at or outside the "3" weld would be confined to the annular region. No leakage would be expected beyor.d the GRAFOIL seal ring, so no BAC of exterior surfaces of the piping or steam generator / pressurizer walls will occur.
o At the onset ofleakage into the annulus, boric acid laden primary coolant would flash or evaporate until the annulus space becomes pressurized. Borie acid crystals would form in the annulus as the water evaporates. Eventually, as pressure increases, coolant could condense in the annular space and absorb air (oxygen and nitrogen) previously present in the annulus. Upon
8 Mechanical Nozzle Seal Assembly (MNSA) Enclosure conderaatien and absorption of air and redissolution of boric acid, the environment in the annulus could become conducive to corrosion of the adjacent base metal. However, the extent of-corrosion of the base metal is ex; ected to be low based on experience and evahiation of the likely environment within the crevice:
EPRI Report TR 104748," Boric Acid Corrosion Guidebook," provides descriptions of BAC that have occurred as a result ofleaks in RCS nozzles. The key feature of the BAC morphology is that the corrosion occurs at the exterior of the penetration where the boric acid solution is exposed to air. Oxygen is required for BAC. Iron and steel will not rust in dry air or in water depleted of oxygen, with or without boric acid. The MNSA device prevents air from reaching the annulus by providing a seal between the nozzle and the base metal. Therefore, corrosion amounts would be limited by the amount of oxygen initially present in the annulus.
SCE intends to remove an alloy 690 nozzle which has been in service since 1993 to inspect the low alloy carbon steel annulus for corrosion during the Unit 2 midcycle outage. Minor corrosion is expected on the order of 3-5 mils per year. A bounding calculation is in progress which will yield the maximum limit of wastage acceptable. An evaluation of the results of the inspection will be documented.
- 7. NRC Request: Several questions relate to the test program and its results. To avoid repetition and confusion they are treated together.
(DEGRADATION MECHANISMS b.) Provide a summary of the testing performed and its results.
(TEST PROGRAM 1.) Provide the details of the three MNSA test configurations. State whether the tests of the MNSA were performed with the RTDs and the attached tubing in place.
(TEST PROGRAM RESULTS 1.1.1) Provide a summary of the test results.
(TEST PROGRAM 1.1.2) Provide the details of the special effects test with the simulated 360*
crack. State the pressure at which the nozzle slipped and contacted the anti-ejection device.
(TEST PROGRAM 1.2) Provide the test results demonstrating that the joint is capable of maintaining both mechanical and seal integrity under all sersice conditions, for the life of the plant.
(TEST PROGRAM 3.) Show that the fatigue tests envelope all expected operating and transient conditions for the life of the plant.
(TEST PROGRAM 4.) Provide details about the seismic shake tests. These tests were performed in the cold condition. State why they are applicable to seismic shaking in the hot condition.
9 Mechanical Nozzle Seal Assembly (MNSA) Enclosure
- 7. SCE Response:
This response covers the seven questions liste.1 above in reference to the testing of the MNSA's.
A discussion of the test program follows.
These laboratory tests were performed on the MNSA to verify its capability to perform its intended design functions of sealing and anti-ejection under different conditions ofinternal pressure, thermal transients, and seismic acceleration. Three types of tests were performed:
hydrostatic pressure testing, thermal cycle tests, and seismic qualification testing. A sumraary of the test results is provided below. Detailed descriptions of the test configurations and results can be found in ABB CE reports TR-PENG-033 and TR-PENG-042 (enclosed).
Design Requirements for Testing ASME Section 111, NB-3671.7 " Sleeved Coupled and Other Patented Joints" requires that "a) the joint design make provision to prevent separation under all Service Loadings, and b) be accessible for maintenance, removal, and rcplacement after service, and c) either of the following two criteria are met: 1) A prototype joint has been subjected to performance tests. . . under simulated service conditions. . . . The mechanical joints shall be sufliciently leak tight to satisfy the requirements of the Design Specification. or,2) Jointo are designed in accordance with th t rules of NB-3200."
The design of the MNSA joints is in accord mce with NB-3671,7 (1989 Edition with no Addenda) and the rules of NB-3200, option 2) above. Prototype testing was performed to ,
seismically qualify the MNSA design. Additional testing (hydrostatic and thermal cycle) was perfonned at the option of the designer to validate the seal integrity of the MNSA joint.
SCE's initial request for the design for the hot leg RTD and pressurizer instrument nozzle MNSA joint was prompted by the leakage of a pressurizer side shell nozzle at SONGS in March 1997. Later, after the discovery of more leaking nozzles in July of 1997, the original request for MNSAs was modified to add MNSAs for the steam generator channel head instrument nozzles and the hot leg PDT and sample line nozzle. Only the MNSAs for the h t leg RTD nozzles and the pressurizer le"el tap were subjected to the design verification testing, herein described. This testing is sufficient to envelope the additional MNS As designed for the pressurizer side shell RTD nozzle, the hot leg sample line nozzle, the hot leg PDT nozzle and the steam generator PDT aozzle. Prototypes for all designs will be subject to hydrostatic testing for seal integrity. All MNSAs are designed in accordance with the ASME Code,1989 Edition, paragraph NB-3671.7, and meet all ASME Code requirements.
TESTING PROGRAM DISCUSSION i Both MNSA configurations (the hot leg RTD and pressurizer level tap instrument nozzle) were subjeted to three types of test:. 1) a hydrostatic pressure test,2) a theraal cycle test, and 3) a seismic test that represent the significant design transients for the life of the plant.
1
10 Mechanical Nozzle Seal Assembly (MNSA) Enclosure
- 1. Ilvdrostatic Pressure Test The hydrostatic pressure test was performed at essentially 3,175*!O psig at ambierit temperature on both the hot leg RTD nozzle MNSA prototype r.nd the pressurizer instrument nozzle MNSA prototype. The RTD and the attached tubing were not included in the tested assemblies. The leak tightness under pressure conditions is not impacted by the presence or absence of the RTD and the attached tubing since they have no sealing function, and their weights are negligible under these conditions. Acceptance enteria stated if any pressure decay of more than 50 psig could not be attributed to leakage from other than the test seal, the test was considered a failure.
To simulate a 360 crack, the nozzle was not weided to the test block. For further details of the hydrostatic test and acceptance criteria, refer to the following sections of the enclosed ABB CE Report TR PENG-042, Rev. 00:
- Sections 4.0 and 5.0 provide a detailed description of the test apparatus and test setup.
- Section 6.1.1 provides a detailed description of the hydrostatic testing of the hot leg mockup. A schematic is shown in Figure 3.
- Section 6.1.2 provides a detailed description of the hydrostatic testing of the pressurizer instrument nozzle.
Results Summarv e A total of 4 tests were performed on the hot leg RTD nozzle MNSA. The final test began with a pressure of 3200 psig. There was a 100 psi pressure decay over a 3 hour3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> period, but there was no sign ofleakage from the seal area. Leakage was attributed to other than the seal. Therefore, the test was successful.
- Pressurizer instrument nozzle: 2 tests were performed. In the final test, the pressure decay from 3175 psig was less than 50 psi with no sign ofleakage from the seal area.
Therefore, the test was successful.
- 2. Thermal Cvele Test The thermal cycle test was performed on the RTD MNSA and the pressurizer instrument nozzle MNSA, by heating the test rig from ambient to 650110 F at rate of 150 *F/ hour. A pressura of 2,500150 psig was maintained for at least 60 minutes. The nozzle was allowed to cool to less than 200 F. Three thermal cycles were conducted to verify seal performance under expected plant temperature extremes. Simulation of a total of three test cycles has been adopted by the designer as the common, sufficient practice to demonstrate sealing functions of GRAFOIL components. The RTD and the attached tubing were not included in the tested nozzle assemblies, ne leak tightness under combined thermal and pressure conditions is not impacted i by the presence or absence of the RTD and the attached tubing since they have no sealing l
l l
11 Mechanical Nozzle Seal Assembly (MNSA) Enclosure fimetion, and their weights are negligible under these conditions. The three thermal cycles were used te verify the integrity and the design of the seal. Since there is metal to metal contact between the compression collar and the lower flange, the seal does not relax during thermal cycling. 'lhe bolts did see the thermal cycles during the test, however, thermal cycle evaluation of the bolts was done by analysis. The hiNSAjoint seals were examined for signs ofleakage.
None was found. The test was successful.
- Section 6.2 of the enclosed ABB CE Report TR-PENG-042, Rev. 00 provides additional details.
The autoclave facility is limited to a heat up rate of about 150'F/hr. The mass of the seal is so small that heat up and cooldown rates are less important than the temperature range. The 150*F/hr rate is between the 200 F/hr maximum heat up rate for the pressurizer and the 100 F/hr maximum heat up rate for the piping. The difference is not significant for the seal.
The special effects test mentioned in the submittal is the hydrostatic test that was performed in conjunction with the thermal cycle test on a prototype hfNSA for both the RTD and pressurizer lower level nozzle configuration. The hydrostatic test simulated the installation of the hiNSA on a nozzle. The nozzle was provided with a free end, which simulated a 360 crack. As the pressure was increased prior to reaching the maximum test pressure of 3,175 psig, the nozzle slipped and contacted the top plate of the h1NSA. No leakage was observed following the hydrostatic tests. Results of the hydrostmic tests m found in the following enclosed test report:
e TR-PENG-042 Rev 00, Test Report for hiNSA Hydrostatic and Thermal Cycle Tests (SO23 411-57-16) i A note on page B6, of TR-PENG-042, referring to the bottom pressurizer nozzle indicates the l
nozzle moved at 2,200 psi.
A note on page A3, of TR-PENG-042, referring to the RTD nozzle, indicates the nozzle also moved at 900 psi.
l 3. Seismic Tut Seal integrity cannot be demonstrated by seismic analy sis. Therefore, seismic qualification testing l was performed to demonstrate phy sical and functional integrity of the seal assemblies. Seismic test
! requirements were based on response spectra provided by SCE, and the specimens were tested at j hydrostatic test pressure to rimulate design conditions with a margin for the test. Test fixtures were
- constructed to simulate the actual configuration.
'lhe nozzle was only partially attached to the base plate ta simulate a 180 crack. Utificial seismic time histories were synthesized to envelope the stipulated OBE and SSE resp , spectra curves.
The seisade test con.;isted of several shaker tabln tests simulating five operating, basis earthquakes
12 Mechanical Nozzle Seal Assembly (MNSA) Enclosure and one design basis earthquake, representing 100 percent of the plant specific design requirements.
The carthquake loads represent the most significant mechanical cyclic load imposed on the clamp throughout the life of the plant, and therefore, can be considered in lieu of a fatigue test. Internal pressure was monitored during the test, he tested assemblics were monitored for leakage after each of the applied seismic events. in addition to the seismic qualification testing, a sine wave sweep was j conducted to determine the resonance frequencies of the nozzle assemblics. Resonance frequencies outside the range I to 33 llz were considered acceptable.
For the OBE/SSE tests, there were a minimum of 24 event cycles, each of 32 see or longer.
Assuming a minimum resonance frequency of 30 liz, and assuming significant response during 50%
of each event period, the total number of cycles therefore was at least 24x32x30x.5 = 11,500 cycles.
The test specimens were equipped with weights that simulated the RTD heads and the valve existing in the field instalbtion. It was discovered that the resonant frequencies of the shaker table and the test assembly nearly matched when the test registered seismic values of up to 10 g's. Actual input at the lowest nozzle resonance exceeded the required input by a fr.ctor of greater than 5.
Seismic shake tests were performed at ambient temperature. Results are consider:d applicable in the hot conditions based on:
- There is no increase in seismic loads in the hot condition, e There is no loss of fatigue strength in the temperature range existing during operation.
- The mechanical seal was hydrostatic tested in the hot condition, and acceptance criteria were met.
A detailco description of the test specimens and mounting is provided in Section 4,1 of ABB CE Report TR-PENG-033, Rev. 00. The test procedure and a discussion of the results are provided in Section 5.0 of the report.
Results Summarv e No leakage, loss ofintemal pressure, or sign of damage was observed at any time during or after each simulated seismic event, e Both test MNSA structures were rigid within the seismic frequency range of 1 to 33 liz.
The pressurizer bottom instrument MNS A resonant frequency was 38 IIz. The hot leg test assembly was not resonant within the sweep frequency range of 1-50 IIz.
l SCE believes that the above tests envelope all expected operating and transient conditions for the life of the plant.
13 Mechanical Nozzle Seal Assembly (MNSA) Enclosure 8.NRC Request: (DEGRADATION MECilANISMS c.)
If therc is a cruk in the "J" weld, the annulus will be filled with water and the seal will then be continuously in contact with the water. State why the galvanic corrosion will stop in the presence of this water. Provide test data to support this assertion. Also, provide test data showing performance of the seal material considering thermal cycling and environmental efTects.
- 8. SCE Response:
If the "J" weld cracks from PWSCC, water will fill the annulus created between the Inconel nozzle and law alloy steel component (pressurizer, RCS pipe, or steam generator). Galvanic corrosion of low alloy steel can occur as a result of an electrochemical reaction between the low alloy steel and the graphite (GRAFOIL) or other corro: ion resistant materials (stainless steel or Inconel) in the presence of an electrically conductive fluid such as borated water. The GRAFOIL serves as the cathodic element in the reaction and the low alloy steel is the anodic element which is corrosively attacked. Even without the GRAFOIL, low alloy steel in the presence of high temperature borated water will experience minor general corrosion.
The GRAFOIL seal maintains its integrity (no loss of mechanical preload) after a crack in the "J" weld, since primary coolant does not leak past the seal. Without a continuous flow situation, the crack in the "J" weld will not be sufficient to replenish the borated water and the wter will become stagnant. Corrosion products and boric acid crystals form in the annulus during thi9 initial corrosion period that will effectively seal off the annulus and prevent further fluid from entering. These corrosion products in the annulus will prevent the GRAFOIL seal from being wetted further. This effectively removes the anode (GRAFOIL) from the reaction since its contact with the water is reduced or climinated and the galvanic corrosion process efTectively stops.
Furthermore, when the borated water is not replenished, the pH of the stagnant water will rise as the boric acid is consumed. The rising pH level will slow, and eventually stop, the corrosion of the low alloy steel due to the borated water.
With the corrosion processes effectively halted, the seal will maintain its integrity.
No specific tests were conducted for the purpose of assessing galvanic corrosion.
Test data showing the performance of the GRAFOIL material under simulated thermal cycling and environmental effects is contained in the enclosed Test Report, TR-PENG-042, for the RTD and the pressurizer MNSA.
- 9. NRC Reque.t: (DEGRADATION MECHANISMS d.)
l The fasteners in this application are made from SA-453, Grade 660 (A-286). Since the fasteners may be wetted by leakage past the seal, provide test data on the performance of this alloy from a corrosion perspective. Address the effects of oxygen and chloride concentration on stress corrosion t
14 Mechanleal Nor2ic Seal Assembly (MNSA) Enclosure cracking performance. Define the expected environmental conditions, including measures at the plant to limit chloride contamination.
- 9. SCE Response:
SA-453 Grade 660 (alloy A 286)is a precipitation hardened austenitic nickel chromium stainless steel. Grade 660 possesses the general corrosion resistance comparable to that of other austenitic stainless steels.
Grade 660 has been shown to be susceptible to stress corrosion cracking (SCC) when highly stiessed and exposed to either pressurized water reactor (PWR) or boiling water reactor (BWR) primary water operating conditions. SCC related failures of Grade 660 fasteners have been attributed to high stress levels in nominal BWR or PWR water environments. SCC testing of Grade 660 materials (for instance," Stress Corrosion Cracking of A-286 Stainless Steel," CE-NPSD-305, enclosed) has shown that stress is the principal contributing factor to failure of this material in high temperature water environments. The metallurgical condition and processing history also influences the SCC susceptibility. If the stress level is maintained low enough, the potential for SCC failures is significantly reduced or eliminated.
Failures of Grade 660 due to SCC have not been attributed specifically to high chloride or oxygen levels. EPRI primay water chemistry guidelines for pressurized water reactor nuclear power plants chemistry limit chloride contents to a maximum of 50 parts per billion (ppb) and oxygen to less than 5 ppb under nonnal operating conditions. These chemistry limits are sufficient to minimize die potential for SCC of most austenitic materials in high temperature water providing the tensile stresses are kept to suitably low levels.
The stresses in the A-286 bolts used to attach a MNSA to the vessel or piping are created by the torquing preload. The specific stress values indicated below are from the hot leg RTD MNSA design, but are typical for all MNSA locations. The stress in these bolts is 22,500 psi, which is low compared to the minimum yield strength of 85,000 psi for this material. The other A-286 components are the tie rods. The tie rods are not stressed under normal conditions. Their function is to act as an anti-blowout device should the nozzle weld fail completely. Conservatively assuming an instantaneous weld failure, the impact load creates a stress of 32,000 psi; this reduces to 6,300 psi after the impact, when the tie rods are only reacting against the normal pressure load. Again, these stresses are low compared to the yield strength of the material.
Leakage past the MNSA seal is not likely to be able to maintain a liquid environment around the bolts or tic rods. Dry boric acid accumulation as a result ofleakage which turns to steam will not adversely affect the bolts or tie rods because Grade 660 is not susceptible to BAC. Any accumulation of boric acid crystals on or around the MNSA will require a disassembly and inspection of the MNSA when it is observed. Other impurities in the primay water would also tend to concentrate as a result ofleakage which turns to steam. The presence of increased levels of oxygen or chlorides will reduce the threshold for SCC in Grade 660. However, stresses near the yield strength of the material would still be needed to induce cracking in primary water with 6000
15 Mechanical Nozzle Seal Assembly (MNSA) Enclosure I l
ppm boric acid and 25 ppm chlorides. This chloride level is 500 times higher than the limit j permitted by the EPRI primary water chemistry guidelines. These EPRI guidelines are maintained !
at San Onofre.
- 10. NRC Request: (DESIGN AND FABRICATION)
Provide the ABB CE design report for the mechanical nozzle seal assembly (MNSA) showing ,
compliance with the ASME Section til Code.
- 10. SCE Response:
The enclosed ABB CE reports S PENG-DR-002 / Design Report, Addendum to the Pressurizer Analytical Stress Report for SONGS 2 and 3 and S PENG DR-003 / Design Report, Addendum to the Piping Analytical Stress Report for SONGS 2 and 3 demonstrates that all stresses are satisfactory-and meet the appropriate allowable limits set forth in Section 111 of the ASME Code. The design reports, stress analysis reports, and qualification reports for the steam generator PDT MNSA, hot leg PDT MNSA, and hot leg sample MNSA are not scheduled to be submitted to SCE until December 22,1997. Since the elesign is similar, and the reports use similar methodology, it is expected that all stresses and cumulative fatigue usage factors will be satisfactory, and meet the appropriate requiremerts from the ASME Code, Section 111.
I1. NRC Request: (TEST PROGRAM 2.)
Provide me design thermal and mechanical operating and transient cycling loading conditions that the MNSAs are expected to experience during the life of the plant.
I1. SCE Response:
San Onofre Units 2 and 3 Updated Final Safety Analysis Report, Table 3.9-1 (enclosed) summarizes the design transients.
- 12. NRC Request: (TEST PROGRAM 2.1)
State what steps will be taken to prevent the potential ejection of a nozzle while installing a MNSA without complete depre:surization or draining below the nozzle in question.
- 12. SCE Response:
The RCS will be depressurized, with an open vent path established, before MNSA installation commences. Draining of the loops is not anticipated, as the remaining head pressure, with the vent to containment atmosphere established, does not pose a hazard to personnel during the installation of a MNSA.-
16 1 Mechanical Nonle Scal Assembly (MNSA) Enclosure j i'
- 13. NRC Reauest: (TEST PROGRAM 3.2)
Provide the analysis of the MNSA showing that all components are within allowable stresses.
- 13. SCE Response:
The enclosed n ports demonstrate the joint safety under service conditions and that all components -
are within allowable stresses. ;
Reactor Coolant System Piping e S-PENG-DR-003 Rev. 01/ Design Report, Addendum to the Piping Analytical Stress Report for SONGS 2 and 3.
- A-SONGS-9416-1175 Rev. 01/ Evaluation of attachment locations for MNSA on SONGS -
Unit 3 RTD nozzle locations in the Hot Leg Piping Pressurizer e S-PENG-DR-002 Rev. 01/ Design Report, Addendum to the Pressurizer Analytical Stress Report for SONGS 2 and 3 e A-SONGS-9416-1170 Rev. 00 / Evaluation of attachments location for MNSA on Unit 3 Pressurizer shell bottom head instrument nozzles The enclosed ABB CE letter from J. T. McGarry to M. Ogawa dated December 12,1997, provides clarification of analytical methods used for these reports.
- 14. NRC Request: (TEST PROGRAM 4.2.1)
Provide ABB Document Number S-PENG-DR-003, " Addendum to the Piping Analytical Stress Report for SCE SONGS Units 2 and 3."
- 14. SCE Response:
S-PENG DR-003 / Design Report, Addendum to the Piping Analytical Stress Report for SONGS 2 and 3; Edison report SO23-411-57-18, is enclosed.
- 15. NRC Request: (TEST PROGRAM 4.2.2)
Provide ABB Document Number S-PENG-DR 002," Addendum to the Pressurizer Analytical Stress Report for SCE SONGS Units 2 and 3."
_=
17 Mechanical Nozzle Seal Assembly (MNSA) Enclosure -
--15. SCE Response:
S-PENG DR-002 / Design Report, Addendum to the Pressurizer Analytical Stress Report for SONGS 2 and 3; Edison report 5023-411-57-3, is enclosed.
-16. NRC Hequest: (TEST PROGRAM 4.2.3)
Provide the results of the ASME Section Ill NB-3200 fatigue analysis of the pressurizer wall and the hot leg wall, with the bolt holes.
- 16. SCE Response:
The structural integrity of the pressurizer and hot leg walls with the bolt holes has been evaluated to be in compliance with ASME Code stress and fatigue requirements as well as area reinforcement requirements for metal removed. The results are shown in reports:
- A SONGS 9416-il70 Rev. 00 / Evaluation of attachments location for MNSA on Unit 3 Pressurizer shell bottom head instrument nozzles e A-SONGS-9416-1175 Rev. 01/ Evaluation of attachment iocations for MNSA on SONGS Unit 3 RTD nozzle locations in the 110t Leg Piping
- 17. NRC Request: (INSPECTION 1.)
Describe the procedure to inspect for leakage from a through wall crack in the nozzle.
I
- 17. SCE Response:
SCE inspects for leakage from a crack in the nozzle using three separate but similar programs.
- 1) Leakage tests are perfonned under the ISI program, with VT-2 examinations,2) boric acid leakage is found under our prograr.. 'o minimize BAC, and 3) the nozzles are inspected specifically for leakage under our alloy 600 program. Each program is discussed briefly below.
- 1) ASME Section XI has the following ISI requirements for MNSA:
- A) System leakage test, Table IWB-2500-1 Examination Category B-P, All pressure retaining components.
l
, SONGS procedure SO23-XVII-3.1.1, "P.crueling Outage InMrval Examination of the Reactor j Coolant Pressure Boundary to Detect Leakage," will be useo to meet the above requhement.
Procedure SO23-XVII-3.1.1 complies with the ASME Section XI requirements. Only lI i
18 Mechanical Nozzle Seal Assembly (MNSA) Enclosure persons certified to at least Level ll VT-2 shall perform or witness examinations. Visual examinations are conductul for evidence of component pressure boundary leakage, distress, or corrosion prior to returning the unit to service following a refueling outage. The results ofexaminations shall be documented on VT-2 system pressure test reports.Section XI does not require disassembly of the MNSA for this examination. Leakage originating from the wcld or through wall leakage from any pressure retaining component is unacceptable and requires corrective action prior to returning the unit to service. The inspection summary report shall be sent to the regulatory r .thorities in accordance with Section XI, IWA-6230. 1 B) Visual, VT-1, Table IWB-2500-1, Examination Category B-G 2, Pressure retaining bolting,2 inch and less in diameter, in accordance with Table IWB-2500-1, footnote (1), Bolting may be examined (a) in place I under tension, (b) when connection is disassembled, and c) when bolting is removed. l 1
SONGS procedure SO23 XXVll 20.49," Visual Examination Procedure to Determine the l Condition of Nuclear Parts, Components or Surfaces," will be used to meet the above requirement.
SONGS procedure SO23-XXVII-20.49 is in compliance with all the applicable Section XI requirements. VT-1 visual examinations are used to determine the condition of the part, I components, or surface examined, including such conditions as cracks, wear, corrosion, erosion, or physical damage on the surface of the part or components. The examiner shall be qualified and certified to the requirements of ASME section XI. All results shall be documented in the examination report. Components detected, by visual examination, to have the relevant conditions described in IWB-3517-1 are unacceptable, and need corrective action in accordance with the Code. The Inspection summary report shall be sent to regulatory authorities in accordance with Section XI, IWA-6230.
- 2) Boric Acid Leak Inspection Program The SONGS Boric Acid Reduction Program conducts containment walkdown (visual) inspections when containment is accessible. The purpose of these inspections is to locate any emergent leakage from principal leak locations including MNSAs. The inspections are conducted by personnel who are knowledgeable of the principal leak locations. This inspection is conducted under procedure SO123-V-8.15," MODE 3 Boric Acid Leak Inspection."
- 3) Reactor Coolant System inconel Nozzle Inspection Program The SONGS Reactor Coolant System Inconel Nozzle Inspection program inspects for leakage from a through-wall crack in the nozzle and/or the "J" weld., This program identifies each nozzle location and directs the inspector to record his findings. It is conducted during MODES 4 or 5 operations, when insulation has beca removed from the nozzles and scr.ffolding provides access to the nozzlet where necessary. This procedure is also used on the return to MODE 1, to identify any emergent
19 Mechanleal Nor21e Seal Assembly (MNSA) Enc 40sure RCS boundary leakage from the nozzles before return of the Unit to service. This inspection is conducted under procedure SO23-V-8.16, " Reactor Coolant System inconel Nozzle Inspection."
With the four procedures discussed above, SCE has developed a thorough plan to locate and repair leaking nozzles. Copics of these four procedures ere enclosed.
- 18. NRC Request: (INSPECTION 2.)
Provide the inspectica method to verify that no loss of preload will occur. -
- 18. SCE Response:
Additional inspection to verify that no loss of preload will occur on the MNSA bolts is not required for the following reasons : (1) The bolts are provided with retainer washers to prevent loosening of the bolts. He retaining washers are serrated, and after torquing of all tolts is complete, one or more tab (s) of the retaining washer is bent against the flat of the corresponding bolt and one or more tab (s) is beat against the upper flange or top plate, and (2) The bolt holes that oppose the gasket compression preload are located well outboard of the seal region and do not become wetted.
Therefore, loss of preload of the gasket due to corrosion of carbon or low alloy _ steel tapped holes cannot take place. All components of the MNSA itself are fabricated from corrosion resistant stainless steel.
- 19. NRC Request: (INSPECTION 3.) -
Since the fasteners are susceptible to stress corrosion cracking, provide your rationale for not performing volumetric examination of the fasteners.
- 19. SCE Response:
As stated in the Response to Question 9 above, the stress loading of the isteners is very low, which mitigates the susceptibility of the fasteners in this application to SCC. Furthermore, the fasteners lie sufficiently outboard of the GRAFOIL seals so that the probability of an SCC environment being established is low. Without high stress and a conducive environment SCC will not be credible.
Therefore, no volumetric examination is planned. However, in the event fcakage is discovered the entire MNSA will be disassembled, inspected, and new fasteners will be installed, if needed, prior to retum to sersice, i
I
20 -l Meehanical Nozzle Seal Assembly (MNSA) Enclosure l
.l
- LISTING OF ENCLOSED DOCUMENTS - !
r Drawinyn
. SCE Document No. Vendor Document No./ Title
'S O23-411-57-4 E MNSA 228 001/ Bottom Pressurizer MNSA S O 23-411-57 5 E-MNSA-228-002 / Side Picssurizer RTD MNSA -
S O23-411-57-6 E-MNSA-228 003 / RCS Hot Leg RTD MNSA' S O23-411-57_ E-MNSA-228-004, Sht. I of 4 / MNSA Details S O23-411-57-8 E-MNSA-228 004, Sht. 2 of 4 / MNS A Details S O23-411 57 9 E-MNSA-228 004, Sht. 3 of 4 / MNSA Details S O23-411-57-10 E-MNSA 228-004, Sht. 4 of 4 / MNSA Details
, (S O23-411-57-20 E-MNSA 228-014, Sht ' of 2 / Steam Generator PDT MNSA S O23-411 57-21 E-MNSA 228-014, Sht. 2 of 2 / Steam Generator PDT MNSA S O23-411-57-25 E-MNSA-228-015, Hot Leg PDT MNSA S O23-411 57-26 E-MNSA-228-016, Hot Leg Sample MNSA S O23-411-57-22 E-MNSA-228-013, Sht.1 of 3 / MNSA Details S O23-411-57 23 E-MNSA-228-013, Sht. 2 of 3 / MNSA Details S O23-411-57 24 E-MNSA 228-013, Sht. 3 of 3 / MNSA Details SO23-919-12 Vessel Welding & Machining for San Onofre Il %" 1.D. Pressurizer SO23-919-13 Bottom Head Welding & Machining for San Onofre 1196" 1.D. Pressurizer SO23-919-16 Nozzle Details for San Onofre 1196" 1.D. Pressurizer SO23-919-19 Nozzle Details for San Onofre 11 Piping Renorts and Evaluations San Onofre Units 2 and 3 Updated Final Safety Analysis Report, Table 3.9-1 SCE Document No. Vendor Document No. / Title S O23-411-57-3 S-PENG-DR-002 Rev. 01/ Design Report, Addendum to the Pressurizer Analytical Stress Report for SONGS 2 and 3 S O23-411-57-14 A-SONGS-9416-1170 Rev. 00 / Evaluation of attachments location for ,
MNSA on Unit 3 Pressurizer shell bottom head insunent nozzles S O23-411-57-15 TR-PENG-033 Rev. 00 / Test Report, Seismic Qualification of SONGS 2 &
3 MNSA clamps for Pressurizer instrument nozzles and RTD Hot Leg Nozzles S O23-411-57-16 TR-PENG-042 Rev. 00 / Test Report for MNSA Hydrostatic and Thermal Cycle Tests 1
S O23-411-57-18' S-PENG-DR-003 Rev. 01 / Design Repod, Addendum to the Piping
21 Mechanical Nonle Seal Assembly (MNSA) Enclosure l t
Analytical Stress Report for SONGS 2 and 3.
S O23-411-57-19 A SONGS-9416-1175 Rev. 01-/ Evaluation of attachment locations for MN'sA on SONGS Unit 3 RTn nozzle locations in the liot Leg Piping CDCC# 60752 ~ CE-NPSD-305 / Stress Corrosion Cracking of A-286 Stainless Steel. -
l i
Root Canae Evaluations ,
SCE internal memorandum from C. Chiu to J. T. Reilly,' dated November 5,1986;
Subject:
forrective Action for the Unit 3 Pressurizer Nozzle Failure in March 1986
'.SCE internal memorendum from C. Chiu to M. P. Short and L. O. Cash, dated June 30,1992; ,
Subject:
Root Cause Evaluation 92-019, SONGS 3 Pressurizer Level Instrument Nozzle Leakage
~
Root Cause Asse_ssment page from Action Request (AR) 970300092-02;
Subject:
Temperature Element 2TE0101 Procedures SO123-V-8.15 MOLE 3 Boric Acid Leak Inspection SO23-V-8.16 Reactor Ceolant System inconel Nozzle Inspection SO23 XXVil-20.49 Visual Examination Procedure to Determine the Condition of Nuclear Parts, Components or Surfaces SO23-XVil-3.1.1 Refueling Outage Interval Examination of the Reactor Coolant Pressure Boundary to Detect Leakage Correspondence i -- . .
Letter, dated December 12,1997, McGarry, J. T. (ABB-CENO) to Ogawa, M. " Fatigue Analysis of
' SCE Pressurizer and Piping MNSAs" c '
(
a J
4
--.n - - ++, +.. , ,n- u - ,
i San On:fra 2&3 FSAR l
a . Updat:d MECllANICAL SYSTEMS AND COMP 02 NTS (z
Table 3.9-1 TRANSIENTS USED IN STRESS ANALYSIS OF CODE CLASS 1 COMPONENTS (Sheet 1)
Occurrence Conditions Hestup and 500 heatup and cooldown cycles during the design life of cooldown cycles the components in the system. The rate of heating and cooling is 100F/h between 70F and 545F except for the pressurizer which has a rate of 200F/h between 70F and 653F. The heatup and cooldown rate of the system is administrative 1y limited to assure that these limits will not be exceeded. This condition is based on a nor-mal plant cycle of one heatup and cooldown per month rounded up to the next highest hundred.
Power changes 15,000 power change cycles over the range of 15% to 100%
of full load at a rate of 5% of full load per minute either increasing or decreasing.
Normal cyclic 108 step changes of 1100 lb/in.2 and 110F (120F for variations surge line) when at operating conditions. This condi-tion is selected based on 1 million-cycles approximating an infinite number of cycles so that the limiting stress is the endu.ence limit. Grouped together in these cycles are pressure variations associated with fluctua-tion in pressurizer pressure between the setpoint for actuation of the backup heaters and the opening of the spray valves; temperature variations associated with the CEA controller deadband; and 2,000-step power changes of 110% of full load assuming I cycle per week for 50 weeks of the year. (Two wecks of the year are assumed required for refueling.)
Reactor t rip, 480 cycles to include any combination. Thisincludesk turbire trip, reactor trips due to operator error, equipment malfunc-loss of reactor tion, and a total loss of reactor coolant flow (i.e.,
coolant flow a total loss of reactor coolant pump electrical power).
This is based on one occurrence per month for the life of the plant.
OBE condition See paragraph 3.7.3.2.2 and appendix 3.7B for the procedures used to determine the number of earthquake cycles during the seismic event.
3.9-3
S3 Onofra 2&3 FSAR l
o Updatcd MECHANICAL SYSTEMS AND COMPONENTS
' Table 3.9-1 TRANSIENTS USED IN STRESS ANALYSIS OF
, CODE CLASS 1 COMPONENTS (Sheet 2 of 2)
Faulted Condition
- 1. The concurrent loadings produced by normal operation at full power, plus the design basis earthquake, plus loss-of-coolant accident (pipe rupture) are used to determine the faulted plant loeding condition.
2l 2. Loss of Secondary Pressure: one cycle postulated loss of secondary pressure due to (one each) a complete double ended severance of one steam generator steam or feedwater nozzle, but not simultaneously. These are not considered credible events in forming the design basis of the reactor coolant system. How-ever, they are included to demonstrate that the reactor coolant system components will not fail structurally in the unlikely event that these events do occur. The number of occurrences is an aribitrary number.
Test Condition Occurrence Conditions Primary system 10 primary side cycles from 15 lb/in.2 to 3,125 ,
hydrostatic lb/in.2a at a temperature between 100F to 400F. These cycles are based on one initial hydrostatic test plus a major repair every 4 years for 36 years which includes equipment failure and normal plant cycles. The secondary side of the steam generator is at atmospheric pressure during this test.
Primary system 200 cycles from 15 lb/in za to 2250 lb/in.2, og .
leak temperature between 100F to 400F, These cycles are based on a normal plant maintenance operation involving 5 shutdowns per year for 40 years.
3.9.1.2 Computer Programs Used in Stress Analyses 3.9.1.2.1 Non-NSSS Systers and Components Analysis of piping systems other than RCS main 1. oops is performed by use of the following proprietary computer programs: ME 210, ME 101, ME 632, ME 643-1/643-2/643-3, ME 913, ME 916, and ME 199.
t 2/86 3.9-4 Revision 2
______-_a._-_-.___._. _ _ _ _ _ _ . - _ _ _ _ _ _ _