ML20045E622
| ML20045E622 | |
| Person / Time | |
|---|---|
| Site: | Haddam Neck File:Connecticut Yankee Atomic Power Co icon.png |
| Issue date: | 06/29/1993 |
| From: | Opeka J CONNECTICUT YANKEE ATOMIC POWER CO., NORTHEAST UTILITIES |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| B14530, NUDOCS 9307020248 | |
| Download: ML20045E622 (10) | |
Text
$
,4 NORTHEAST UTILITIES cone,,i Omc.,. seioen si,,,i. e.,iin. Conn.ciico, I
UNE.NErN52 s
P.O. BOX 270 v$oww.Y E :1,
H ARTFORD, CONNECTICUT 06141-0270 g
E J
- vy -
(203) 665-5000 June 29,1993 Docket No. 50-213 B14530_
Re:
U.S. Nuclear Regulatory Commission Attention: Document Control Desk Washington, DC 20555 Gentlemen:
Haddam Neck Plant Reauest to Use Alternative to ASME Code Section XI The purpose of this letter is for Connecticut Yankee Atomic Power Company (CYAPCO) to propose to the NRC Staff, in accordance with 10CFR50.55a(a)(3)(i), an alternative to the 1983 Edition, Summer 1983 Addenda of the ASME Boiler and Pressure Vessel Code for the inspection of the feedwater system at tne Haddam Neck Plant.
During the current Cycle 17 refueling outage, radiographic examinations revealed indications of cracking in the feedwater nozzles on three of four steam generators of the Haddam Neck Plant. According to the ASME Code, the number of cracks identified result in the sample size of welds required to be examined to expand to include all of the feedwater piping from the steam generators all the way back to the feedwater isolation motor-operated valves (MOVs) in the turbine hall.
CYAPCO believes that it is not necessary to perform an examination of all these welds-to determine adequacy and operability of the feedwater system. Theretore, CYAPCO is hereby proposing an alternate program in accordance with 10CFR50.55a(a)(3)(i). The proposed alternate program is described in Attachment No.1 to this letter. CYAPCO believes that the proposed alternative provides an acceptable level of quality and safety.
Also, CYAPCO believes that compliance with the specified requirements (i.e., inspection of all of the welds in feedwater system piping back to the feedwater isolation MOVs) would result in hardship and unusual difficulty without a compensating increase in the level of quality and safety.
CYAPCO has concluded that the cracking experienced in the three feedwater nozzles is localized and results from known phenomer}a. Our proposed alternate program is intended to confirm this conclusion without having to perform all of the otherwise required inspections. If all of the feedwater piping weld inspections had to be performed, this 0 9 0 0 M.
9307o20249 93o629 PDR ADOCK 05o00213 h)\\
/
G PDR
I'
.o it U.S. Nuclear Regulatory Commission B14530/Page 2 June 29,1993 would significantly delay the restart of the Haddam Neck Plant for Cycle;18 and would significantly impact personnel radiation exposure and personnel safety related to asbestos removcl.
CYAPCO respectfully requests that the NRC Staff respond as quickly as possible to this request. A response by Friday, July 2,1993, would support the current schedule.for startup of the Haddam Neck Plant for Cycle 18 operation.
If you have any questions, please contact us.
Very truly yours, CONNECTICUT YANKEE ATOMIC POWER COMPANY PLb J. F. Opeka_j 3
Executive Vice President 1
Attachment cc:
T. T. Martin, Regional Administrator A. B. Wang, NRC Project Manager, Haddam Neck Plant W. J. Raymond, Senior Resident inspector, Haddam Neck Plant L
-, r.
Docket No. 50-213 B14530 Attachment No.1' Haddam Neck Plant Alternative Program of Feedu'ater Piping Weld inspections in Accordance with'10CFR50.55a(a)(3)(i) b 2
June.1993 i
m.
~
1 n
PURPOSE:
L To discuss the efforts undertaken by Haddam Neck to confirm that the root cause of the steam generator-(S/G) feedwater nozzle cracking detected during the 1993 refueling outage was thermal and chemistry conditions limited to the nozzle region only and do not extend into the remainder of.
the feedwater. system. This will-then be used as the basis to justify the use of an alternative to the Edition of the ASME Section XI Code applicable to Haddam Neck (i.e. 1983 Edition, Summer 1983 Addenda).
BACKGROUND:
The four steam generator feedwater nozzles were first~
radiographed during system installation. These nozzles.were again examined during 1979 and 1980 as required'by I&E Bulletin 79-13. The-steam generator #2 feedwater nozzle was again examined in 1986 to confirm that the cracking phenomena described in I&EB 79-13 had not initiated since the 1980 examinations. Although indications were identified by the 1
Level II inspector in some of the 1979 inspections, the final Level III review did not identify any unacceptable indications during either the 1979 or the 1980 examination.
A review of the S/G #4 nozzle 1993 radiographs revealed-unacceptable linear indications. Subsequent reviews indicated ~
that the indications detected in the 1993 radiographs were also present in the 1979 and in the 1980 radiographs, although to a lesser degree (i. e. the 1993 indications were-approximately 4-5" longer). No indications were. detected during the 1986 examinations, partially due to the fact that the " floating" thermal sleeve had extended into the area adjacent to the weld and masked the area of_ potential concern.
Steam generator feedwater nozzle cracking was first detected in 1979 at D.
C.
Cook Unit 2, and it was the subject of NRC Bulletin 79-13. Haddam Neck joined'the Westinghouse-Owner's Group program in 1979 to determine the root cause of the steam generator feedwater nozzle cracking and to develop-repair alternatives. The results of these activities, indicated that the feedwater nozzle cracking was a result of corrosion fatigue caused predominantly by thermal-stratification and striping during cold, low-flow, feedwater injections. Other significant parameters included high-oxygen.
exposure, counterbore geometry, material' yield strength and the-number of times that large: temperature differences (stratification) existed during heatup, hot standby, and low power operations.
Recent discussions with Westinghouse personnel-indicated that the stratification effects typical of the feedwater nozzle and adjacent piping only extends up.
to the first vertical section of piping and does.not affect the remainder of the system. In order to confirm this
.I 2
q
.m
a finding, Haddam Neck examined additional welds and additional supports to confirm the absence of global system stratification.
ASME SECTION XI CODE REQUIREMENTS:
The S/G #4 elbow to nozzle weld which was inspected during the current '(i.e.
1993) outage was part of the ASME Section XI, IWC-2000 weld population required per. Code Case N-408-2.
For this reason, the detection of unacceptable indications-required a sample expansion to include four additional' pipe.
welds. Haddam Neck chose to expand only to the other three S/G elbow to nozzle welds since these welds were believed to be susceptible to the same cracking phenomena.
The detection:
of additional rejectable indications in two of the.other three nozzle required further sample expansion per ASME-Section XI IWC-2430 to essentially. include-the. entire feedwater system. The Edition of the ASME Code currently committed in the ISI Program for Haddam Neck (i.e. 1983 Edition, Summer 1983 Addenda) does not-provide the flexibility to differentiate between susceptible and non-susceptible weld-locations. For this reason, Haddam Neck proposes as an alternative to the~1983 Edition, Summer 1983
. Addenda of the ASME Section XI IWC-2430 that the rules of IWC-2430 of the 1992 Edition of the ASME Section XI Code be applied. The 1992 Edition allows a reduced sample expansion whenever technically justified by a root.cause evaluation of the unacceptable conditions. As discussed below,.Haddam Neck has performed an evaluation of the nozzle cracking detected and has concluded that expansion of the inspection populat' ion to include the remainder of the feedwater system weld locations is not structurally justified and will not significantly increase plant safety.
DESCRIPTION OF DETECTED CRACKING:
The feedwater nozzle indications were first detected in the S/G #4 nozzle by radiographic examination and confirmed by ultrasonic examination using.various scanning techniques.
Although the outer surface of the nozzle did not provide an ideal inspection surface, the indications detected by RT were confirmed by UT. Upon confirmation of these indications the inspection scope was extended to include the remaining three nozzles. These additional inspections revealed: crack-like indications in'S/G's #2 and #3 along with movement of the thermal sleeve as compared to the 1979, 1980 and.1986 radiographs. Based on these findings the piping' elbows adjacent to the nozzles were removed to provide access to the inside surface of'the nozzles. Subsequent visual and-Liquid Penetrant (LP) examinations performed before and after flaw removal confirmed the length and depth of the cracks detected during the RT and UT examinations.
Upon confirmation of the cracks in nozzles #2, #3 and #4,.
3 E
Haddam Neck' developed a repair program in accordance with ASME B&PV Code Section XI rules to remove all detected cracks and weld repair the affected regions by building up the nozzle wall to its previous thickness. Grinding was'also-performed after welding to remove any surface discontinuities introduced during welding and therefore minimize the stress concentrations.
As a result of the thermal sleeve movement and the nozzle to thermal sleeve gaps, Haddam Neck also performed a secondary side visual examination to ensure that sleeve bwass flow did not result in nozzle corner cracking and the-thermal sleeve movement was not indicative of. feed ring damage.
Although the secondary side water level required'for personnel shielding minimized the accessibility to the nozzle corner, no visible cracking was detected'in this region. An.
inspection of the feed ring also revealed that the thermal sleeve movement was within.the intended design movement based on the gaps between the feed ring.and the corresponding-support brackets. Therefore it.can be concluded that based on the secondary side visual inspection, no unacceptable conditions were detected.
STRUCTURAL AND ROOT CAUSE EVALUATION:
Haddam Neck, with the assistance of Westinghouse.
personnel, has evaluated the cracking detected ~duringithe current refueling outage and has concluded, based on the-results obtained to date, that the cracking is similar to the corrosion fatigue ~ cracking described in~I&E Bulletin 79-13.
The bases for this conclusion are as follows:
(1) The cracking was detected in the bottom of'the nozzle.
inner surface step transition. This location was both the thinnest area and it contained the highest stress concentrating effects. Stress concentration from weld counterbores was identified as a primary _cause of cracking in the I&E Bulletin 79-13' evaluations.
(2) Although cracking was detected in the horizontal section of the nozzle (see figure 1) no cracking was detectedLin the first upstream, vertical pipe to elbow weld. This indicates that the loads at the elbow to nozzle weld are localized and are not transferred-through the_ elbow into-the remainder of the-piping system. Local thermal stratification and stripping was identified as al primary cause of~ cracking in the I&E Bulletin 79-13. evaluations.
(3) _A comparison of the indications detected in the 1979/1980-radiographs-and those detected in the 1993 radiographs showed no significant' crack extension over the 14 year period. Typical fatigue cracking under constant stress is characterized Dy a relatively long crack initiation-period followed by a shorter crack propagation period.
4
Therefore the relatively slow crack growth-rate observed between 1979 and 1993 indicates that the predominant crack driving potential is a result of local self-relieving stresses and not a result of global system loads.
(4) Although a definite correlation can not'be.made between-the operating characteristics of the generators and the propensity for cracking, the extent.of cracking appears to correlate well with the operating duty of.the S/G's.
Following plant trips, the RCP's in loops #1 and #3 also trip while the RCP's in loops #2 and #4 remain in operation.
Additionally, the RCP's in' loops #3 and #4-are the first_to be activated and the last to be.
deactivated during normal plant-heatup and cooldowns.
Based on this.it is concluded-that the duty cycle of S/G-
- 1 is less severe than those of S/G's #2,.#3 and!#4. This correlates with the propensity of-the cracking /G's #1.
detected-in three S/G's with no cracking detected in S (5) Although a detailed chemistry history of each S/G is not available for review, the magnitude of S/G tube damage has been more pronounced in S/G'e #2 and #4.'Since"these two S/G's also exhibited gre';6r nozzle cracking it can-be concluded that past secondary side chemistry:which has caused tube damage (i.e..high oxygen) may'have contributed to the nozzle cracking.
In. summary it is concluded that the corrosion fatigue phenomena in nozzle weld counterbores described in I&E Bulletin 79-13, is the same phenomena which caused the Haddam Neck nozzle cracking detected during the current refueling outage.
ALTERNATE PLAN:
In order to ensure that the pipe / fitting welds in the feedwater system are not susceptible to the type'of cracking-detected in the feedwater nozzle to elbow welds,'Haddam-Neck
~!
will complete the following additional inspections and evaluations prior to returning the plant to service:
(1) Expand the inspection scope to include the first two
)
welds beyond the first vertical pipe section'in lines supplying S/G's #2 and #4. These welds.were chosen based 1
on the magnitude of the cracking detected in the #2 and-
-l
- 4 nozzles and.also to include both the"long and~the short piping arrangements between the containment i
penetration and the steam generators. The-purpose of.
these inspections-is to verify that global system stratification beyond the first vertical pipe section does not exist.
(2) Perform a visual examination of the feedwater piping-5
.i
=
rigid restraints inside containment to confirm the absence of support damage resulting from global system stratification caused by system back flow. The-absence of t
support deformation / damage confirms that the-Haddam Neck feedwater system is not subjected to the' stratification-phenomena described in I&E Eulletin 88-11, in I&E Information Notice 91-38 and in Information' Notice 84-87.
(3) Examine the piping welds in the pipe which supplies S/G-
- 4 adjacent to the containment penetration. The absence of unacceptable indications in this weld further confirms-that-global stratification resulting in high-loads at;the system fixed ends, does not exist.
(4) The indications detected in the 1993. radiographs were also detected in the 1979 and the 1980. radiographs. For this reason Haddam Neck has reviewed the remainder of the I&E Bulletin 79-13 inspection results to confirm that no additional questionable indications were detected in any of the areas originally covered by:the Bulletin.
(5) Haddam Neck has reviewed _the' industry history related-to-steam generator feedwater nozzle cracking to determine if significant cracking resulting from stratification loads-has been' detected beyond the first vertical pipingsrun.
This review did not reveal any instances where cracking-related to stratification loads was detected beyond the first horizontal.section of piping adjacentito.the feedwater nozzle.
l (6) The feedwater system has been subjected to the ASME-Section XI ISI requirements which included inspection of numerous welds since 1979. A review of these inspection
+
results revealed no unacceptable crack-like indications detected during these inspections.
(7) Haddam Neck will complete a structural evaluation of the feedwater nozzle and the piping to determine'the critical =
l flaw size for the system. This critical-flaw size will; i
then be compared to the maximum flaw size which could-have been missed during inspection to confirm the absence of structurally significantly flaws.
t COMPLIANCE WITH 10CFR50.55a:
Haddam Neck believes that the above expansion plan ensures compliance with the requirements of 10CFR50.55a(a) (3) as discussed below:
1e (i)
.The proposed alternatives provide an acceptable level:of-quality and safety.
The results of the above evaluations and examinations-6 6
M 4
)
a along with-the review o,f the s/G Feedwater nozzle cracking history' indicates that the type of cracking
]
characteristic of the feedwater nozzles-does=notJextend-u beyond.the first horizontalEsection of piping.
J Therefore extending the inspection scope as required by j
the 1983 Edition, Summer 1983 Addenda of the'A3ME I
Section XI Code, subsection IWC-2430 would not resultiin.
a significant increase-in plant safety.
(ii) Compliance with the specified requirements of this-section (i.e. 10CFR50.55a) would result in hardship or unusual-difficulties without. a compensating increase fut the level of quality and safety.
The inspection-'of the remainder ofRthe entire.feodwater; system would require,.significant personnel exposure,-
significant-financial exposure as.well as personnel safety implications related to asbestos removal-and erection of scaffolding. Since there is no evidence that the' remainder of.the feedwater system is susceptible-to the cracking mechanism detected at the nozzles, further.
expansion of the inspection scope beyond the additional four welds and ten supports discussed above,'would not significantly increase plant safety and it would negatively impact personnel' safety.
m 7
l....i.,
J.
A
t
~
WO B
L E -
GN ICUD E
R 21 s
X k
41 x
N 6
O IT v
~
AC l'
OL E
/
R K
U E
C G
V A
I R
F EE C
LS \\
LAM R
EH e
T c
G N
IR D
EE F
,