ML20042C038
| ML20042C038 | |
| Person / Time | |
|---|---|
| Site: | Millstone  |
| Issue date: | 03/05/1982 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20042C037 | List: |
| References | |
| TAC-47577, NUDOCS 8203300104 | |
| Download: ML20042C038 (8) | |
Text
.
pa u%IC UNITED ST ATES g % 5,(* g NUCLEAR REGULATORY COMMISSION g
C WASHINGTCN, D. C. 20555 S
a4d' l
%, %.....f
' SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION SUPPORTING AMENDMENT NO. 7 3 TO FACILITY OPERATING LICENSE NO. DPR-65 NORTHEAST NUCLEAR ENERGY COMPANY, ET AL.
fi!LLSTONE NUCLEAR POWER STATION, UNIT NO. 2 DOCKET NO. 50-336 Introduction By letters dated January 12 and flarch a,1982.as supplemented by letters of January 22 and February 5,12 and March 1,1982, Ncrtneast Nuclear Energy (Company (NNECO or the licensee) proposed changes to the Technical Specifications TS) to allow less than 100 percent inspection of the _tutes in each steam generator (SG) and to nake additional reports on primary to secondary leakage rates during Cycle 5 operations of Millstone Nuclear Powqr Sta.tton, pnit No. 2 (Millstone-2).
The reduced inspection of SG tubes is to be based on the results of sludge analysis and eddy-current testing (ECT) results.
Background
fiillstone-2 is currently in the Cycle 5 reload outace. TS 4.4.5.1.2 requires at least a 3*; sanple of the total number of tubes in all SGs be ECT inspected.
NNECO performed ECT inspection of about 1100 and 800 tubes in SG-1 and $3-2, respectively. Entry for these inspections was from the hot leg side taking the probe around the top bend to check for denting indication at the uo;er support clates or any other tube degradations. Earlier inspection results sho ced i-ci:ati0ns (defined in TS as > 205 tube wall decradation) in s me of the central tubes between the tube shee and the first su; port plate.
. The nu*ber and severity of indications was such as to require a 100 percent inspection of both SG.
To data, the licensee has performed a 100 cercent tube inspection within a
- .9 o's re2 on of each SG uhich they believe ennservatively bound the area of i
..ga.:. -
. n. a.. : ~.,.... :.,.., :.
- .
- . 7 : 7,:, ;.
7..,-
generator No.1.
NhECO's recuest is to linit the required iOC percent inspec-tion to the area of the SG in which sludge pile maaping and the additional insnection indicatas the degra:ation has Occurred.
Discussion Millstone-2 has experienced a moderate amount of denting, particularly at the drilled support plates which has led to the plugging of approxi-mately 800 tubes. Corrective actions taken in 1977 appear to have substantially reduced the rate of denting. These actions included retubing the main turbine condenser with 90-10 cupro-nickle (Cu-Ni),
installation of a full flow condensate polishing systen, elimination of 82033CK)104 820305 PDR ADOCK 05000336 P
_ hardspot areas ("rimcutting")-in the drilled suppo'rt plates, and improved secondary water chemistry control. Eddy current inspections of the drilled support tube intersections performed during the current outage have indicated scme progression of denting since the previous inspection (average progression of 2 mils), but no tubes restricted r.ssage of the standard size eddy current probe at the drilled supports. Hows
, four-tubes restr.icted passage of the eddy current probe at the first egg crate support on the cold leg side and were preventively plugged. Profilomentary techniques rather than eddy current testing is the most accurate monitoring method for denting at egg crate supports, due to the tendency of eddy current to underestimate ovalized dents which is the characteristic dent geometry at the egg crate supports.
Profilometer inspections have been performed for approximately 800 tubes during the current outage, but the results are still under evaluation by t"e licensee. Mavicun dents easured at the egg crate suppcets during the previcus inspection we e approximately 40 mils.
Initial eddy current test sampling revealed a large nunber of indications in the central bundle regions of both steam generators where sludge has accumulated on the tubesheet. Based on these findings,- the licer.see defined an inspection in each steam generator encompassing the locations where the initial indications were found, and performed an ecdy current inspection of each tube within this boundary (with exception of 54 tubes as discussed below). All. indications found as a result of these inspections were located well within the inspection boundary. The inspection boundary bounded the Outer 90st tubes containing indications by a minicum of five tubes.
The licensee -eported in its letter' dated February 12,1982 that 33 and 21 tubes in steam generators 1 and 2, respectively, which are located within the iaspection boundary defined in the January 12, 1982 letter, were not iassscted. These tubes were recortedly inaccessible to the eddy curren aqui: en withcut e<tet.sive equiprent dedification cr personnel radiation e<;csure.
Tne licensee has justified not inspecting these tubes on the
- Basis tnat no defective tubes have been identified within at least 5 tubes a# any cf the uninspected tubes.
Our 'ev'e.. indicated tPat the exclusion of over half of these tubes accunts
--'mr adjust eat c' :Pe i scection touMary consistnt tith ret % d:':gy
_c4 c -
C :; :..e.cs :e.i:i c;:c.
.e T Si i. ; :.e % -;,.
tubes are locateo along rca 1 on the hot leg side Of both stean generators, ir ediately adjacent to the open lane between the not and cold legs.
Tubes llocated outside the inspection boundary were confined to the peripheral areas of the tube bundle.
Sludge mapping indicated the peripheral region of the tube bundle to be essentially free of sludge. However, several hundred
.e tubes were inspected in the peripheral areas to co'nfirm the absence of pitting in these areas and to monitor the progression of denting. The following table shows.the number of tubes inspected and the results.
SG-1 SG-2 Hot Leg / Cold Leg Hot Leg / Cold Leg ECT ' Inspected 7200/6900 6800/6500 Indications (>20%)
175/575 17/352 Defects (>40%)
424 280 All tubes with 40% or greater indications have_ been plugged in accordance wi th T3 4.4.5.1.4.
This included the 424 tubes in SG-1 and 280 tubes in SG-2, a total of 700 tubes plugged for the pitting problem this outage.
This leaves only 151 and 72 tubes, in SG-1 and SG-2, respectively, with known degradation at the beginning of Cycle 5 operation.
The type of plug used is the 'destinghouse designed renovable plug used previously in other Westinghouse steam generators. NMECO is considering the return of these tubes to service during the next refueling outage by sleeving the defective tubes. This will~ be reviewed by the staff at a later ti:"e.
The eddy current indications occurred at discrete elevations, usually within a narrow (1/c inen) band for any given tube betaeen 0 and 10 inches above the tcp of the tubesheet. Sludge napping established that the indications Sccurred within the sludge area.
ost of the indications occurred on tne col d leg side of each stean generator.
Three tubes centaining eddy current indications were removed from SG-1 for laboratory study. Examination of these tubes shows snall (< 60 mils) dianeter cits as the source of the indications.
Similar pitting attick has recently bean observed at Indian Point-3 although the magnitude of such defects e ss-s less sevs e in the "ilistene-2 stein genera y.
Evaluation of tne eddy current data was cenplicatec by the presence of interference signals originating fron cooper deposits on the tubes. The eddy current test progran enployed a conventional coccin probe using multi-frequency techniques (400 and 200 kilo-hertzs (kHz) differential,.100 kHz absolute). To verify acceptable sensitivity to these small volune pits,
^
the licensee reinspected twenty tubes using a nodified bobbin probe (reduced coil spacing) and a 500 kHz test frequency to ninimize signals originating from the outer surface of the tube (e.g. copper signals). Good correlation i
(i.e. within + 15%) be. tween the conventional' probe data and the nodified probe data was obtained except for one tube with a 29% difference, the nodified probe predicting the larger pit depth of SM.
J
.. IGood correlation was also observed between the con'ventional probe data (i.e.
field data) and what was actually observed in the laboratory fo.r one of the pulled tube specimens (81 percent in the field eddy current test vs. 77 percent wastage observed in the lab). Comparative data is not yet available for the other twn tube specimens which exhibited field indications of 83%
and 28% respectively.
Indications of the type observed during the current outage had not been observed previously. Of these tubes found to contain indications, only 6 tubes had been inspected previously. Eddy current data tapes fron previous inspections were reviewed for these tubes with the follcwing results.
SG Tube-August 1980 December 1981 Mu :er Outage Cutage 2-137-43 none 84%
2-137-45 none 99%
l-122-24
<20%
21%
- ~ ~ ~
?3%
l-122-48 none l-114 44 none 24%
l-84-100
<20%
30%-
The first two tubes were inspected in 1980 by multifrequency while the other 4 tubes were inspected previously by single frequency. These conparisons were chtained by relooking at the 1980 ECT tapes.
The licensee has conducted a test program, cescribed in our meeting on Feb-uary 15,1982 and documented by letter dated " arch 1,1982, to evaluate the Cotential effects of sitting on tube integrity. These tests were per-
- or ed on tube specimens containing simulated cits. The tests included speci ens with individual pits ana other stecinens with nuitiple arrays of cits, both in a circumferential and axial orientation. These tests includeo
- pit 'dianeters ranging to.187 inches which is conservative for pits actually cbserved at '4111 stone-2 or Indian Point-3.
2 assure tests 'y the licensee demonstrate that pits ranging in depth to 925 t"e' o-i;i al e0' :"4:t: ness cai sustain censideribly ir, excess of the
- re3 s.-2 ';ed ng as sa
- nec..i ta a a stul a*.ed m e n s;e> line craai u i.
'",SL5) esith no. leakage. " hen tested to failure, the failures are generally local to the pit (i.e. within the pit) rather than a general or gross -failure
.<hicn could cause large leakage. For severely pitted tubes, the nultiplicity oof closely spaced-pits may degrade tube integrity more than would be indicated by pressure tests of. tubes with individual pits since the failure could involve more than one pit. However, even assuning an axial array of four.187 inch l
~
J
diame' er pits, and penetrating 88 percent through' wall, with a 10 mil t
ligament spacing between p'ts, pressure tests by the licensee have demon-strated burst pressures in excess of postulated MSLB pressures. The licensee also performed a pressure test on one of the pulled tube specimens which exhibited an 83% field ECT indicat' n.
This tube exhibited a burst pressure in excess of 9000 psi, which is again indicative of the,high residual stre,ngth associated with even severely pitted tubes.
Visual inspection of the secondary sides was' performed by NNECO from both the top of the bundles and from the lower hanu holes. The gap between the upper support plates and the shroud, established by the rim cutting operation in January 1978, was reported to be apparently unchanged since previous.
inspection. No plate to shroud contact was reported. The licensee inspected fa* the presence of locse parts such as were found in previous years as a result of cracking of the tube support plates where the rim cut operation was performed. However,'no such pieces were found during this inspection.
Evaluation L'e have reviewed the licensee's inspection program and results. We find that the licensee has performed adequate inspections to define the areas of the tube bundles affected by the pitting phenomenon. The licensee has inspected essentially all tubes in regions which we believe adequately bound the areas of concern.
- 1though there were 22 tubes unich were not inscected and are located within the inspection boundary and in what can be generally described as the affected area of tne steam generator and an area of some sludge accumulation, we believe that it is unlikely that these tubes contain retectable pits based ucon our
-eview of spat al distribution of eddy current indications and the f act that 4
tqere are no indications..ithin a minimun of five tube distance from these uninspected tubes.
In addition, considering also tne randonness by whicn pits in this general area tend to occur, the presence of detectable pitting in more than one or two of these tubes is even more unlikely. Finally, even if pitting is present in these tubes, the pitting would be expected to result in only a small leak before tube integrity would become sufficiently degraded
-- resu't in a g :ss tute failure er large '22 Rage 'as discussed 'urthe-The t05-plugging limit implemented during this inscection was developed'to ensure adequate tube integrity for generali:ed forms of corrosion such as uniform thinning of the tube wall or linear defects with uniform wall pene-
- tration, in addition, tubes must be inspected at frequent enough intervals
.to ensure that further progression of corrosion beyond the plugging limit will not excessively degrade tube integrity.
o k
n
For the case of Millstone-2, we are less than confident that substantial progression of pitting will not occur during the next operating interval.
l For generalized corrosion involving long linear defects of uniform wall penetration, the allowable wall penetration to ensure no gross failure under postulated MSLB conditions is 78%. To provide a margin of three against i
burst under nornal operating ccnditions (per F.egulatory Guide 1.121), the allowable wall penetration (again a:suming generalized corrosion) is 63%.
However, the corrosion problem at Millstone-2 involves small diameter pits, which have a substantially smaller effect on tube, integrity than nore general.ized forms of corrosion as has been substantiated by tests. Tube examinations at both Millstone-2 and Indian Point-3 indicate that the pitted tube surface generally would not be expected to exhibit continuous linear flaws of any significant length (say >.75 inches) involving significant I
average wall panetrations (say > 78%). We believe that the burs,t test tube
)
specimens which contained a linear array of individual pits, spaced 10 mils apart, over a 3/4 inch length provides a reasonable conservative bound on the potential for individual pits to interact. Fpr.this_ case, the test results I
submitted by the licensee demonstrate that each of 'the pits in the array l
would have to penetrate in excess of 88% through wall to reduce the burst strength of the tubing to less than the MSLB pressure differential. Therefore, considering that the amount of through wall penetration for individual pits will vary randomly, we believe that the further progression of pitting can be expected to result in snall leaks before tube integrity becomes sufficiently degraded to result in a gross tube failure during normal or postulated accident conditions.
Similarly, leaks would be expected before large numbers of indivi-dual pits have penetrated through wall (> 90%) such that significant leakage might result during postulated accident conditions. Allowable limits on
- rinary to secondary leakage (i.e., 0.5 gpm) are specified in the TS 3.a.6.2 beyoni which the plant must shut down for steam generator inspections and re; airs.
' The licensee olans to operate for a full cycle (approximately 3a0 effective full power days (EFPD)) orior to performing the next steam generator inspec-tion. We io not find data at this tine (cending the next steam generator i9stection)
- ate any meaningful estinate of the rate of pitting progression c -
.-- S i c s n;n59 -s 3.ir7 c.,;;,
- 3..
.y tunes af fectec by pitting, and the potential for significant progression of pitting during the upcoming cycle, we find it necessary to keep ourselves inf ened of the minimun reason?.hle increase in primary to secondary leakage.
To accomplish this goal, m:ECO committed to determine the secondary gross activity and report to the assigned Project Manager the baseline leakage of
each steam generator (determined after two weeks of equil'ibrium operation) and any subsequent incremental increase of '. 05 or more within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />; followed by a written report within 14 working days. The written report will contain NNECO's evaluation of the increased leakage and justification for continued operation. The staff may, after review of,,these increased leak 3ge recorts,.take action as necessary, including ordering a reactor shutdown, to protect the pricary system integrity. NNEC0's above commitment has been documented by applicat on for TS changes, dated March a,1982.
i The af fected TS is Page 3/4 7-8.
These additional reporting requirements apply only to primary to secondary leakage in excess of the baseline leakage to be determined by the end of the
- cc-d week following restart fren the current outage. We believe that any l
primary to secondary leakage'immediately following the current cutage would likely be caused by weepi'ng plugs such as has occurred recently at this facility.
The staff notes that the above restrictions are in-addition to the current l
TS requirements which specifies an immediate plant shutdown in the event o' primary to secondary leakage exceeding 0,5 gpm Based on the above evaluation and the new sampling and reporting requirenents, we conclude that the inspection program implemented during the Cycle 5 refueling In addition, cutage, and thus the requested changes to the TS, are acceptable.
we conclude that operation of the Millstone-2' steam generators during the next operating cycle, subject to the added coerating co mitments does not represent an undue risk to public health or safety. The staff did r.equest,.however, that the licensee cennit to submitting the results of the profilcmentary inspec-tions at the egg crite supports, and their assessnent of the results within In addition, 60 days following restart fron the current Cycle 5 refueling outage.
we believe 'WECO shoulo inspect the 22 tubes identified in their February 12, 1E:2 lette
.iich were not inspe:ted daring -he current inspection, but which are located in areas affected by pitting. NNECO nas agreed to these two
- additional requests as docunented in their " arch 4,1952 letter.
E " rec e-tal Ocnsideration
- -r/e Mter-i ed t"It the a.end ent noes *ct aut Orize a change in effluent
..a
.. a c 3 r i n; a: n
- ...e 's.i.
m.,il' 5sf ;
in any significant environmental impact. Having made this determination, we have further cencluded that the amendment involves an action which is insignificant from the standpoint of environnental irpact and, pursuant to 10 CFR 551.5(d)(4), that an environnental impact statement or negative declaration and environmental impact appraisal need not be prepared in connection with the issuance of this amendment.
4
.]
t
_ 8_
Conclusion We have concluded, based on the considerations discussed above, that:
(1) because the amendment does not involve a significant increase in the probability or consequences of accidents previously considered and does not involve a significant decrease in a safety margin, the amendment does not involve a significant hazards consideration, (2) there i's' reasonable assurance that the health and safety of the public will not be endangered by operation in the proposed manner, and (3) such activities will be conducted in compliance with the Commission's regulations and the issuance of this ~ amendment will not be ininical to the common defense and security. or to the health and safety of the public.
Date:'?3R 5 '952 Principal' Contributors:
Conrad McCracken Enmit Murphy
' lor:te Conner t
G t
l
~
m