ML20203F679
| ML20203F679 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 10/03/1997 |
| From: | Nunn D SOUTHERN CALIFORNIA EDISON CO. |
| To: | Merschoff E NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| Shared Package | |
| ML20203D220 | List: |
| References | |
| 50-361-97-15, 50-362-97-15, NUDOCS 9712170442 | |
| Download: ML20203F679 (16) | |
Text
_-
.O g
$OUlHUtN (.AlllORNI A EDISON l'20""
=
u_,m m mm.,.,
.? z October 3,1997 Idb F
OiH G i.Ji J
ij fi E. W. Merschoff "1
Y
^
~
Regional Administrator U. S. Nuclear Regulatory Commission, Region IV 611 Ryan Plaza Drive, Suite 400 Arlington, Texas 76011-8064 Dear Mr. Merschoff;
Reference:
SCE Alloy 600 RCS Penetration Nozzle Pre-Decisional Enforcement Conference, September 30,1997 As discussed during our referenced meeting, SCE agreed to provide additional information regarding A!!oy 600 RCS penetration nozzles. The enclosures to this letter provide the requested information.
The graphical information on Alloy 600 nozzle Primary Water Stress Corrosion Cracking (PWSCC) history, which was provided on pages 14 and 15 of our handout, represent actual through-wall PWSCC and non through-wall PWSCC indications, and did not include nozzles which were pro-actively replaced (refer to Enclosure 1). We have also included available information on the size and orientation of PWSCC indications. Please replace pages 14 15, and 16 from our handout with those provided in Enclosure 1.
To addrecs your question regarding how many heats, both at San Onofre and in the industry, constitute the total population of Alloy 600 RCS hot leg and pressurizer penetration heats, we reviewed industry information and have determined that there are: 31 total heats of Alloy 600 used in the industry (11 of which had detected PWSCC by 1995); and 11 total heats of Alloy 600 used at San Onofre (6 of which had detected PWSCC by 1995). This information is further represented in Enclosure 1.
i To address your question regarding visual examinations, Code related VT-2 visual examinations are performed on the RCS (including instrument nozzles) each refueling outage in accordance with the San Onofre In Service inspection (ISI) program, in addition, augmented inspections of instrument nozzles are performed in a manner equivalent to a VT-1 inspection.
i l
Additionally, we have reviewed the dose exposure information which was provided on L
page 30 of our handout. The 2.3 person-Rem per hot leg nozzle and 4.4 person-Rem per pressurizer nozzle exposure were derived from the actual dosimetry records for c u iu i28 e
CA 92674 0128 P
^=ck **gt rooim.im
E. W. Merschoff 2
October 3,1997 work performed at San Onofre to replace those nozzles. Hot leg nozzles are installed using manual welding. Although using remote welding would reduce exposure during welding, the additional time to set up the remote welding apparatus offsets this benefit.
SCE estimates use of remote welding would actually increase exposure. Portions of the pressurizer nozzle installation are performed using remote welding apparatus.
Consequently, SCE does not believe the exposure estimates would be improved by using remote welding.
We also indicated during the meeting, that SCE had identified some errors in the Inspection Report. These items are provided in Enclosure 2.
We appreciated the opportunity to discuss with you the Alloy 600 RCS penetration nozzle PWSCC situation at San Onotro, including the following points:
The apparent violation for four cases of NX7630 through wall PWSCC, was in error in that there were only two confirmed and one suspected NX7630 through-wall PWSCC.
Industry and SCE data demonstrate that NX7630 is no more susceptible to e
PWSCC than many other heats. There is no reliable way to predict if or when a particular nozzle will experience PWSCC. Accordingly, there was no reason for special replacement activities on Heat NX7630; in fact, such a program would have had significant ALARA impact. SCE can not identify corrective action in response to the apparent violation, other than replacement of all Alloy 600 RCS penetrations, which would preclude recurrence.
SCE has an aggressive PWSCC inspection program, which is believed to be the most comprehensive in the industry. Where SCE can anticipate a nozzle as being likely to leak during the next cycle, SCE will replace the nozzle. SCE will never operate the plant anticipating any specific pressure boundary component to le9k. SCE has and will continue to operate in full compliance with the 1
Technical Specifications.
SCE did not violate 10 CFR 50, Appendix B, Criterion XVI. SCE has responded to Alloy 600 PWSCC consistent with the NRC's historical positions that: Alloy i
600 RCS penetration PWSCC has a low safety significance; PWSCC leaks will not propagate rapidly; and any PWSCC leakage will be detected by inspections and repairs effected by licensees.
Finally, SCE has been, and will continue to be, responsive to any changes in the NRC's expectations for the industry in this regard, l
E. W. Merschoff 3
October 3,1997 i
N you have any additional questions, please call me.
1 Sincerely,-
j q
J
\\
k k\\
i k
Enclosures:
As stated
- cc:
K. E. Perkins, Jr., Director, Walnut Creek Field Office, NRC Region IV M. B. Fields, NRR Project Manager, San Onofre Units 2 & 3 J. A. Sloan, NRC Senior Resident inspector, Units 1,2, & 3 NRC Document Control Desk T
L
\\ '
sy
-.-r9 bq n.L
.-e
-g r.--->
_.gwr e
--m-ew.-
3r----
-r---
U
^
o
---,,a n-
,.4-
--sa..wM.
6 4__
.-u
_p----.,-
e,---,,--D,-.A.e
-u4 s.m..
a a- > -
e--,
.aA-a u
a z
--na.
,o 8
e 5
8 4
i e
1 ENCLOSURE 1
?
i ENCLOSURE 1 This enclosure provides updated versions of pages 14-16 from our handout. All data is current through 1995.
The updated pages 14 and 15 lists each heat of material in the axis of the chart for improved clarity. No changes to the data have been made. Our conclusion remains that many heats of Alloy 600 have experienced PWSCC. The performance of heat NX7630 is well within the performance of many other heats.
Two additional tables are provided. Table 1 lists all heats of Alloy 600 used in ABB/CE units, what population of nozzles are from that heat, and the distribution in pressurizer vercus hot leg service. The table shows how the higher temperature environment of the pressurizer influences whether a heat has exhibited PWSCC. Of the heats not used in pressurizer service,2 of 17 (12 percent) have experienced PWSCC, while 9 of 14 (64 percent) of those in pressurizer service have experienced PWSCC.
As expected, service temperature is a strong indicator of the potential for PWSCC.
Overall 31 heats of Alloy 600 have been used in ABB/CE units. As of 1995, of the 31, 11 (35 percent) have shown PWSCC. Since 1995, two additional heats have experienced PWSCC.
Table 2 lists the known information concerning PWSCC size and orientation for San Onofre and the Industry. From the table it can be seen San Onofre's experience is consistent with that of the industry. This information is provided in response to your l
request at the conference.
The updated page 16 reflects some improvements we have made in the Weibull l
analysis since the conference. One additional NX7630 nozzle at another plant, that has not been identified with PWSCC, has been added to the database. The analysis
(
has been refined by incorporating industry experience with Alloy 600 replacement nozzles and adjusting the time of discovery of PWSCC found by NDE to more properly account for the delay in detection when compared with detection by leakage. Finally, PWSCC of weld filler material used at three units for replacement Alloy 690 nozzles has been deleted from the industry data. Since the mode of PWSCC with these
. nozzles is within the weld filler, it is appropriate to " suspend" or exclude this PWSCC from the industry database.
i In summary, our conclusions from these updated handout pages and the additional tables remain unenanged. Heat NX7630 exhibits a PWSCC pattern consistent with other heats in service within the industry.
a am.a
-_t,_m s
._e
.w-+.4._u..
ra.
a e
.a-m-asa-an.
,,ai.-,.an..aem,,,w.-__-
-.s e-+a m.+b ar A
---we.u.-
im.u-=-re+-
8 I
9 Percentage e.,
k 1
2 y
3 i
5 5
s-8.
a e
O B.
Nm71i.'
\\
\\
o,,
Q NX8297
- c
[
\\
j e294
. \\~___
___ x_ __
n a
g gy
-(y
\\g po g
m,.
y g
7,,,.2 c
4 3
a
'g g
- Nxoi,
\\'
5,_ _ \\- \\_
a N
4 m oi N
- F i
.m.
~'
G i!!
2 72 t
u.__x_2 -
c E
@ NX7630 I.
I I
y i
o
/-- f_._ -.-
(
,1 l
?r m
5 l
j' C
f j___- j' O
h
- I o
i j
-j l~T
[ Z o
i l
g m-l l~g l
{
~~~,
a 5
I - - -_ _
eo i
g f
c e
i I
Q 0
o O
O
[
g h
3 8
h I
I i
-m
r
- l Alloy 600 Nozzle PWSCC History l
l Indications of PWSCC, by heat, as a percentage of total population for that heat through l
l 1995) l i
p w
l t
l 8m i
k E
E t
m.
i l
i
=
L f
I o a
~
5 s
g 5
j2n l
l n. Q. E j
9015297 9294 N C530 54318 LWO 73R7-2 19 4411 4t'iOt 9a7"3 24Er72 NM2571 Heat t
When compared to other heats of Alloy 600, heat NX 7630 does not exhibit an unusual rate of PW3CC l
i Page 15 i
i i
l 2
Incomel 600 Nozzle PWSCC Indications CE Flames (T.> 64te*F) ss Hest MX7630 j
(Data available through 1995) i 90-
/
i 80
- --[i7rr/insp/.c-raios9p m----+Y 1>.,- -
-y-l 70-w- - - -
60
~~~ ~~ ~r
~~' T^rd -- -
- ~i=^
~ = '= 'T: ' T E~~r 50
- - - 4 * -- ~
---t~
40 2--
L-~
--; - l-i
- - ---i" ~,
30 i ~".
- =- -
-+
^ 20
-+--F--
- I --
L'-J'e 1
t i
Qw 1
i i
,s t
g
}Q,
-.-_._-_~r--1,-~--
-l
._..._.._.~u.--
. _ -. ~..
,4-O A
a l
V 4
3 I
w.
4 i.... q 4' -. -. -
- - - - =
- a- - - - - -
wy*.%<.---4
~.-.i-*a.-
.=----.y
-+
a i
b
'[ Nx7 63 0 'l-a _ _x i ; i I I
s 4
3
]
+
4
_a__
--+-~_~~. - - _. - ~ -
- - _ - - - ~ ~,....
N.. _._. ~
- X
'N l
i
= 1
,t i
W 1_
~_g___,__
CE Plants (Thot >600 Deg ) ' r -
I l
l 7 _. m.
z' r
1
.5
~.-
.W Eta Beta r^2 n/s i
127.O484 1.81'
'361/357
.2 7 q. _.
6 _.._.
125.3549 1.593 47/43 1
4, 3
I 1
10 100 1000-Tsume to Indicaties (EFFH x lete)
Nennalised to 653'F l!!-
The Weibull plot for NX7630 indicates performance consistent with the industry data. This can be seen by the close agreement of both the Characteristic Life (Eta), and the Slope Factor (Beta) for both populations.
Page 16 y
I a,.
r,-
,--.e r
TAllLE 1
All.OY 600 llEAT POPUI.ATION SUMh1ARY In 1995 Heats with PWSCC Plessunzer Hot Leg Pressunzer & Hot Number of Indications
- Nozzles Nozzles Leg Nozzles indications Combined
~
LO90 0
8 8
3 NX7630 8
39 47 4
9294 0
14 14 1
64318 10 5
15 3
NX8297 37 0
37 1
7387 2 8
0 8
3 94758 2
0 2
1 NX0571 1
0 1
1 NX4411 5
0 5
2 41501 5
6 10 4
24872 4
0 4
2 lotal 80 71 151 25 Heats without Pressunzer Hot Leg Pressunzer & Hot Number of PWSCC Indications Nozzles Nozzles Leg Nozzles Indications Combined 9915 0
33 33 0
K248 2
0 2
0 NX6895 0
10 10 0
26166 0
10 10 0
NX0566 2
0 2
0 NX9731 0
10 10 0
NX9752 0
38 38 0
NX0003 0
30 36 0
NX3918V 0
4 4
0 7760-6 0
1 1
0 82516 7
0 7
0 NX9484 0
12 12 0
7617-4 0
5 5
0 7760-4 0
1 1
0 K259 0
5 5
0 L51723 0
9 0
0 NX0379 0
19 19 0
NX3362 33 0
40 40 0
K333 4
0 4
0 NX4829 7
9 16 0
Total 22 242 264
- Includes both through wall and PWSCC indications There are 31 heats used in ABil/CE Units. Of the 31,11 or 35% have shown PWSCC through 1995.
4 TABLE 2
Size and Orientation of PWSCC San Onofrq PWSCC case Orientation I.ength 1986 U3 pressurizer Axial 5/8" 1992 U3 pressurizer Axial Nozzle 11 - less than 7/8" 1997 U2 pressurizer Axial not available linltistry
'lhe above San Onofre experience is consistent with the available U.S. Industry data.
Referring to !!PRI lleport Tit-103696, "PWSCC of Alloy 600 Materials in PWR Primary Sy. stem Penetrations", July 1994, the following PWSCC llaw evaluations have been reported:
.Ynt Plant Notes 1987 Plant W 2 of 4 pressurizer nozzles inspected had axial indications near the J groove weld 1989 Plant X An axially orientated flaw was reported to be 0.75" long starting from the inside of the pressuri7er.
1990 Plant Y An axially oriented flaw was reported near the J groove weld in a pressurizer nozzle.
1991 Plant Z 2 hot leg nozzles found with axial flaws ranging from 0.37" to 0.89" in length (length was obtained from licensee).
m,...
..a_J
.a__.
m
.h..
. _..a.___A.
.a
_.u m
J m
h i,
l I
4 1
1 ENCLOSURE 2
i ENCLOSURE 2 SCE COMMENTS ON NRC INSPECTION REPORT 50-361197-15; 50-362/97-15 The information contained herein identifies thote instances where SCE's positions are not accurstely charactenzed in the NRC's inspection Report (IR) for San Onofre Units 2 and 3 dated September 10,1997. (Bolding and underlining added for emphasis)
IR Page.
Inspection Report Statement (s)
SCE Response Paraoraoh No.
Cover Letter, "The inspection identified one violation Actually there were three leakage occurrences in 1st page,2nd pertaining to the placing of the Unit 3 reactor 1995, two in the hot legs and one in the Paragraph coolant system in a 10 CFR 50.C5(a)(2) category pressunzer.
as of July 10,1996, despite the identification in 1995 of through-wall cracking in fou_t reactor Note: This comment also applies to similar coolant system nozzle penetrations."
references made by the NRC on page 2 (third bullet, executive summary), page 5, second paragraph, page 6, last paray aph, page 7, third and fourth parayaphs, and page 18.
Executive "The licensee engineering program requirements Documertt 90022 states, in part, "As a result of
- Summary, for inconel 600 reactor coolant system nozzle the development of the exterior nozz'e repair Page 2,5th penetrations, which were contained in Document technique, there are currently no planned nozzle Paragraph 90022 Revision 1, dated February 9,1995, replacement activities? This refers to the l
stated that a progressive replacement program, qualification of the half nozzle repair eliminating based on susceptibility ranking, was the most the need to pursue activities relating to important of the defined activities. Document development of whole nozzle replacement 90022, Revision 1, also stated, however, that techniques. The pian makes no mention of the j
there were no current planned nozzle development of an in-house capability to perform replacement activities due to the external nozzle repairs.
1 development of an in-house capability to perform extemal nozzle repairs (Section El.1)."
Note: This comment also applies to similar references made by the NRC on page 15, paragraph 2, and page 17, paragraph 1.
Page1of4 4
w
.-y._.
~
IR Page, inspection Report Statement (s)
SCE Response Paraaraoh No.
IR, Page 9 Table 1, " Units 2 and 3 Nozzle Penetration Since the table includes not failed (i.e., not Table 1 Failure History as of Cycle 9" through-wa!!) nozzles it would be more appropriately titled " Units 2 and 3 Nozzle Penetration PWSCC History as of Cycle 9".
IR, Page 9, Note 3 states, " Failure caused by fatigue and The failure of nozzle 3TE0121Y2 was not caused Table 1,.
not primary water stress corrosion cracking."
by fatigue as noted in Note 3. SCE believes the Note 3 failure mechanism was PWSCC for this nozzle.
Note: This comment also applies to similar references made by the NRC on page 12 i
Table 2, Note 2, and page 17, paragraph 1.
IR, Page 10 "This nozzle [2PDT0978-1] had been This failure was the first through-wall PWSCC 3rd Paragraph manufactured fmm heat NX7630, thus, identified in Heat f4X 7630. The prior (or first) representing the second failure of this material indication was not through-wa!! and was found by heat in Units 2 and 3."
PT exam.
IR, Pcge 10, "The [ hot leg] nozz!e was replaced with a partial No weld buildup pad is used on this piping. A 3rd Paragraph inconel 690 nczzle, with the attachment weld pad is only used on the pressurizer (pressure i
made to a weld buildup pad on the piping vessel) repairs.
outside diameter surface."
Note: This comment also applies to similar references made bv the NRC on page 11, paragraph 2, and p ge 12, paraw oph 1.
4 Page 2 of4
IR Page, inspection Report Statement (s)
SCE Response Paraaraoh No.
IR, Page 10,
'During the Unit 3 Cycle 8 refueling outage in Evidence of leakage was found at only one 4th Paragraph June 1995, evidence of leakage was found in nozzle locatiort A non-through-wall indication two of the four pressurizer steam space nozzles was also identified in one other weld.
penetrations that had been installed in 1992."
IR, Page 10
" Hot leg Nozzle 3PDT0979-4 was manufactured This failure was the second through-wall PWSCC 4th Paragraph from Heat NX7630, thus, represensting the third identified in Heat NX 7630.
failure of this material heat in Units 2 and 3."
IR, Page 11,
" Nozzle 3PDT0979-3 was ascertained t y the This failure was the third through-wa!! PWSCC 2nd Paragraph inspector to have been manufactured from Heat identified in Heat NX 7630.
NX7630, thus, representing the fourth failure of this material heat in Units 2 and 3.*
Note: This comment also applies to similar references made by the NRC on page 15, paragraph 3.
IR, Page 11, "During review of licensee root-cause Root Cause Evaluation Report 97-001 states, in 2nd Paragraph evaluations (see " Licensee Actions to Determine part, "The most likely root cause is that Root Cause," below), the inspector teamed, manipulation of the RTD in the thermowell during i
however, that another cold leg nozzle the previous refueling outages created cold work l
penetration was evaluated for root cause after of the inconel at the section transition and being found to be leaking in 1996 through the initiated the crack. The crack further propagated threads of a plug that had been installed in the via fatigue as a result of flow passing across the i
nozzle (3TE9179-3)thermowell. This protruding thermowell resulting in the final evaluation determined that the thermowell fracture."
had failed by fatigue, with the failure related to improper installation of the thermowell in the nozzle during vessel fabrication."
Page 3 of4
~
l l
1 IR Page, inspection Report Statement (s)
SCE Response Paraaraoh No.
IR Page 12, For Heat no. 94758, the No. Replaced at Unit 3 The actual numberis 2.
Table 2 is listed as 4.
IR Page 12, For Heat NX 7630, Failure Oty. at Unit 3 is listed The actual number is 2.
Table 2 as 3.
IR Page 12, For Heat NX 7630, the No. Replaced at Unit 2 is The actua! number is 2.
Table 2 listed as 1.
IR Page 12, For Heat K248, the No. Remaining at Unit 3 is The actual number is 1.
Table 2 listed as 0.
IR Page 14 "The inspector considered that the presence of SCE believes that while Heat NX7630 1st Paragraph shallow intergranular cracking (remote from the microstructure may increase susceptibility to more highly stressed material adjacent to the J-PWSCC, other factors (e.g., temperature, yield weld) indicated that the Heat NX7630 material strength Crreironment, residual stresses, etc )
was highly susceptible to primary water result in Heat NX7630 being no more susceptible stress corrosion cracking. The laboratory to PWSCC than other heats.
results were, thus, considered to prov;de some expianation for the subsequent failure history of this heat."
IR Page 15, "The inspector reviewed Document 90022, The program plan orovided recGruiser4ations, 1st Paragraph
' Susceptibility of Reactor Coolant System Alloy not requirements.
600 Nozzles to Primary Water S' Jess Corrosion Cracking and Replacement Prejram Plan,'
Revision 1, to ascertain the licensee's program requirements for addressing primary water stress corrosion cracking in inconal 600 components."
Page 4 of4
=
Ama a4e _-
- sja_.4 mm AAA,4_A A_ b.6 maaJ....;.m__
..pa,,Ame,J_a_._,_
_m.L.A_,,m_J,m._da..AJ4,,,,
4sWapm f q _ h.h,A
^'A-me S 4 C 5_5 4 %_m 44s-W,_ 4 A,4 J h 4 km*44hheA-h4-me-t
' I h
I e
R.NCLOSURE 4 4
i 1-l ii k
P i
CONDENSkD TRANSCRIPT OF PREDECISIONAL ENFORCEMENT CONFERENCE l
l I
f l
e t
' I 6
)
r i
i I
?
6 b
i I
(
t l'
s
..