ML20127L398
ML20127L398 | |
Person / Time | |
---|---|
Issue date: | 11/10/1992 |
From: | Butler W Office of Nuclear Reactor Regulation |
To: | |
Shared Package | |
ML20127L395 | List: |
References | |
NUDOCS 9211240163 | |
Download: ML20127L398 (16) | |
Text
_ _
e 1r -
7590-01 UNITED STATES NUCLEAR REGULATORY COMMISSION PROPOSED GENERIC COMMUNICATION AGEhcY: United States Nuclear Regulatory Commission.
ACTION: Notice of opportunity for public comment. #
SUMMARY
- The Nuclear Regulatory Commission (NRC) is proposing to issue a Generic Letter. A generic letter is an NRC document that: 1) transmits <
information to and requests that analyses or descriptions of proposed corrective actions or both be submitted by the addressees regarding matters of safety, safeguards, or environmental significance; 2) informs addressees of changes in NRC policy and requirements approved by the Committee to Review Generic Requirements (CRGR), the issuance of a topical report evaluation or a NUREC type document of interest to them, or changes in NRC administrative procedures; or 3) requests the cddressees to submit revised technical specifications or other technical or administrative information which does not involve any physical changes to the facility, but which NRC needs to properly perform its function.
This draft generic letter recommends actions to be taken by operating plants with Mark I and Mark 11 steel containments to implement certain inservice inspection procedures that would prevent inadvertent loss of containment integrity and maintain continued conformity w'th their licensing bases.
The NRC is seeking comment from interested parties regarding both the technical and regulatory aspects of the proposed generic letter presented 9211240163 921110 PDR ORG NRRB PDR u . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
a under the Supplementary Information heading. The NRC will consider comments received from interested parties in the final evaluation of the proposed generic letter. Should this generic letter be issued by the NRC, it will
, become available for public inspection in the Public Document Roots. ,
DATES: Comment period expires . Comments submitted after !
this date will be censidered if it is practical to do so, but assurance of ,
consideration cannot be given except for comments received on or before this date.
ADDRESSES: Submit written comments to Chief, Rules and Directivos Review Branch, U.S. Nuclear Regulatory Commission Washington, DC 20555. Written comments may also be delivered to Room P-223, Phillips Building, 7920 Norfolk Avenue, Bethesda, Maryland, from 7:30 am to 4:15 pm, Federal workdays. Copies of written comments received may be examined at the NRC Public Document Room, 2120 L Street, NW. (Lower level), Washington, DC.
FOR FURTHER INFORMATION CONTACT. Ronald Eaton, (301) 504-3041.
SUPPLEMENTARY INFORMATION: The proposed generic letter text is given below:
AUGMENTED INSERVICE INSPECTION REQU STEEL CONTAINMENTS, REFUELING CAVITIES
, AND ASSOCIATED {REMENTS DRAINAGE SYSTEMS FOR MARK- I AND M !
t PURPOSE The staff of the U.S. Nuclear Regulatory Commission (NRC) is issuing ;
this generic letter to request that licensees of operating plants with Mark 1 and Mark 11 steel containments adopt certain inservice inspection procedures that would prevent inadvertent loss of conta' rant integrity and maintain ,
continued cunformity with their licensing bases.
' Includes the refueling cavity as well as the adjacent equipment pool and the spent fuel pool.
i
.,--.-m-w.,_,,. r,-,m,m._vr,.-g e,-r-m. ,.e -,,w,.. , ,->,,.r-,,,,,,,-, awa,.o---,..n-n - .a-.--n,., , ,-v-. - -- - ~
c s
i The staff, with industry assistance, has devised an inspection program 4
which should produce the necessary information. This inspection program consists of the inservice inspection (ISI) of all Mark I and Mark 11 steel containments, refueling cavities, pools, and associated drainage systems.
Each licensee should indicate whether it will adopt the staff's inspection program or an alternate equally effective inspection program. Each licensee should inform the staff in detail regarding the inspection program that it will adopt as part of the plant's inservice inspection program to address the corrosion considerations discussed in this letter. If a licensee adopts an alternate inspection program, which it asserts is equally effective but which
/
deviates from the staff's recommended program, the licensee should identify all deviations and provide the bases for such deviations su that the staff can make a determination of the equal effectiveness of the alternate inspection program.
BACKGROUND Corrosion was-first discovered in the sand cushion region on the outside face of the steel drywell of the Mark I containment of the Oyster Creek Generating Station, and later on the outside face of the upper regions of the drywell. The Mark 11 steel containment has a different phy;ical arrangement but with construction detalis which could lead to corrosion as in the Mark.I drywell. Corrosion and coating degradation have been discovered on the inside face of Mark I suppression pool tori and may occur on the inside face of the suppression pool portion of the Mark 11 containment. Because of these discoveries, the inservice inspections recommended for Mark I drywell and torus also apply to Mark 11 containment l
-a
DRYWELL
, On March 12, 1987, after completing it' review and evaluation of-the licensee's detailed investigation of the steel drywell shell corrosion event at the Oyster Creek Nuclear Power Plant, the NRC staff issued Generic Letter (GL) 87-05 to all licensees of operating reactors with Mark I containments.
In the generic letter, the NRC requested licensees to provide infoemation on the following items:
(1) Drainage of the sand cushion (2) Preventive maintenance and inspection activities to minimize any possible leakage from the refueling pool (3) Plans t * . ltrasonic thickness measurements for those drywell shells with open sand cushions (4) Confirmation of information as listed in T, ale 1 of GL 87-05 Having performed its review .nd evaluated the response provided by the licensee, the NRC staff concluded that the extensive corrosion at Oyster Creek may have been caused by the water leaking through the flexible seal in the refteling pool, and that similar containment degradation may exist at other plants.
TORUS (SUPPRESSION P0OL)
On October 14, 1988, the NRC staff issued Information Notice 88-82,
" Torus Chc1'c with Degraded Coatings in BWR Containments," alerting licensees F
of BWR plants to the discovery of corrosion of the torus at Nine Mile Point 1 and of degraded coatings of the tori of some BWR plants in Region I. The torus at Nine Mile Point I was designed and constructed without a coating.
l l .
4 m~ - ~ ,e m n,-: n + --,n= . - - - - - ,
1 l
l Corrosion of the torus shall and degradation of the torus coating that could l
lead to corrosion of the shell base metal may jeopardize the integrity of the torus suppression pool, wh;ch is critical to the BWR containment as a whole. l The suppression pool boundary in the Mark 11 containment may also become corroded.
DISCU$SION ,
~~
DRYWELL The information collected abe9t the drywell shell corrosion at Oyster Creek indicates that the corrosion resulted from water that had leaked into the sand cushion from the refueling pool. However, the design of the ,
refueling pool and the provision of various drain lines should prevent the water from entering the sand cushion if the seal and the drainage system have been properly constructed and frequently inspected and maintained.
Consequently, the NRC issued GL 87-05 to all licensees with Mark I or Mark II ,
steel containments.
While reviewing the. licensees' responses, the NRC staff has made the following observations:
(1) Functionality Check of the Drains Because most of the draint from the sand cushion are filled with sand and fitted with traps or screens at the ends, it is difficult to determine whether or not such drains are functional. As a result, some of tne licensees have visoally inspected the drain ends, while othtts have not performed any
- inspection. Water that is observed coming from these drains can provide an indication that the drains are unplugged. At one plant, the. source of.the leakage was from the refueling cavity. Although small amounts of moisture P
I k -
..J._ . . . . __..~4._..-,_~._._._.. . _ . - . __,.______,m, , . , . . , . , , . . - , ..J. . . ,v -
were present at three other plants, no indications of corrosion products or ccntamination were observed. This led tt.c staff to conclude that the moisture
- resulted from condensation rather than from leakage. However, when no water is coming from the drains, visual inspections will not indicate whether the drains are plugged or whether the sand cushion does not contain water or ooisture.
(2) Leiiahtness c' the Refuelino Cavity and Other Pools Above the Reactor With the exception of the licensee for Oyster Creek, all licensees indicated that the connections in the reactor c;vity seal drain are welded, and no nonmetallic gaskets aie used. Therefore, the reactor head cavity seal-should not leak. Any leakage will be directed to a drain that is equipped with leak-detection instrumentation and an alarm system to notify the operator about the leak. Licensees are allowed to set these leak-detection devices to sense different rates of Icakage. For instance, in one plant the flow indication switch is set to trip the alarm at 5 gallons per minute (gpm), in another plant the switch trips at 0.1 gpm, in a third plant the switch trips at 10 gpm, and in other responses, licensees did not mention settings.
However, if the leakage is continuous during refueling, even at a level below the lowest possible setting, the potential for moisture in the sand cushion to cause corrosion still exists.
Besides finding the leakage through the connections in the reactor cavity seal drains, the GPU Nuclear Corporatic.. the licensee of Oyster Creek, has further found that the stainless steel liners in the refueling cavity and the equipment pool developed cracks along the perimeters of the liner plates where they were welded to embedded channels.
-- - r-. . - _c--,, .- , , . ,. . -,,.--,_e - , + ~ .- - - - + - . , , , , . . - - . - , , . _ , , 4.-r- r
- t ..
^1 To ensure that no leakage occurs when the reactor cavity and the equipment pool are flooded, liner cracks are sealed with adhesive stainless steel tape and a strippable coating is applied on the liner before the flooding and is removed after this water is drained, it appears such a procedure has been effective in stopping the leakage.
(3) Ultraton'c Thickness Measurements of the Drywell Shell In most plants, the thickness of the steel drywell shell in the area of concero cannot be measured by ultrasonic testing (UT) devices because it is sandwiched between the concrete floor or, the inside and the sand cushion on the outside (see enclosed figure), in some plants, licensees conducted UT measurements from inside the drywell for areas immediately above the sand cushion and did not find either general or local corrosion. In addition to occurring in the sand cushion area, moisture may also accumulate at locations where the gap-forming material has not been removed (for exsmple, between the drywell and the shield building). The licensees should identify such areas.
I Even though GL 87-05 proposed that licensees need to perform UT thickness measurements of only those drywells with open sand cushions, drywells with closed sand cushions should be remotely examined with a visual device to ensure that the cover plate sealing joint is not degraded. If the sealing joint shows evident s of degradation, water leaking througn the cover plate sealing joint into the sand cushion could cause corrosion of the drywell.
TORUS (EUPPRESSION P0OL)
The specified thickness ;f the Nine Mile Point 1 torus shell inc.:r.. orated a 1/16-inch allowance for corrosion based on an astimated rate of corrosion. From tha measured shell thickness, the actual rate of corrosion L.
. _. - .. ~ . . - - . . . - . . - _ - . . . -
j
)
appears to be greater than the original estimate. As a result, the thickness ;
of the shell may not meet the requirements of the American Society of -
t Mechartical Engineers (ASME) Code for the designed life of the containment.
The Nine Mile Point I torus internal surface was not designed to isave,-and does not have a protective coating.
The degradation of coatings in tori may lead to the cc'rosion of the ,
steel shell. The migration of flakes of the coating material to the pump strainers may also degrade the performance of safety related post-accident fluid systems. This could also be the case for a suppression pool with a coated steel boundary as in the case of Mark II steel containments.
RECOMME9 LD ACTIONS TO BE TAKEN BY ADDRESSEES Considering the importance of the containment to the health and safety of the public, each licensee should implement a program of inspection to augment its plant inservice inspection program to ensure that it conforms to Appendix A to 10 CFR Part 50 general design criteria (GDC) 16, 50, 51 and 53.
2 DRYWELL The licensee should inspect the refueling cavity equipment and spent !
fuel pools, drain lines from pools above the drywell, the drains from the sand '
cushion, and the drywell as follows: ,
i (1) Before next refueling, plant persennel should perform inspections and tests for leakage of the joints and seals of the refueling cavity-i l
l l
2 The licensee of Oyster Creek Generating Station which has a unique t ongoing inspection program for the drywell is not requested to take actions recommended herein.
l t
.~ - . _ _ _ _ . . . _ . _ _ . . , _ _ . . . . _ _ _ . ... _ _ ___.._. , _ . - - . . . _ , .
equipment pool and spent fuel pool that could leak water into the air gap. Because the level switch, the flow indicators, and the flow switches cannot detect small leaks, they should not be relied on for leakage detection. Personnel should repeat the inspection when there is leakage as evidenced from the inspection of the sand cushion drains, observed flow of water through sand cushion drains or detected corrosion
'~
of the carbon steel specimens inserted therein. However, in the absence of above indications this inspection should be performed at least every 10 years.
(2) Before the next refueling, drain lines that drain the leakage from -
the pools should be checked to ensure that these lines do not contain restrictions that would inhibit the flow, and to ensure that the water is not directed into the drywell air gap. To per#rrm this test, plant personnel can use compressed air, a boroscope, or other appropriate means. Further inspections of these drains should L performed during outages of opportunity when water is found in the sand cushion drains. These inspections should be performed at a frequeny not to exceed 10 years. The drain lines above the closed sand cushions should be checked in 3
the same manner, (3) The sand-filled drains leading from the sand cushion should be checked for functionality by testing with compressed air or other mea..s before each refueling and by collecting send samples to test for the presence of moisture before and after each refueling. To 5
w -
4 ,,w--- ,--,.-,----,o., . , . , . , - - - - - - < = . n-.. - - - - . + .4~ -
.- . - -. - ~ ,--.--....e.
-m,-- - - . - . , , , - . , ,. . - , . , . - - -
. 7 t
10 - l i
determine the effect of the moi;ture on drywell corrosion, carbon steel specimens should be inserted into the sand cushion through the drains !
i and withdrawn every six months to check for any indication of corrosion. !
(4) Plant personnel should perform UT thickness measurements of the drywell i
shell if water is detected in the sand cushion nr if the steel specimen l
is corroded. At facilities in which the sand cushion is sealed f with a plate and has no drains, these inspections should be --
performed if the sealing joi.it of the cover plate has evidence of 1
degradation. The measurements should include not only the sand ,
cushion region, but also that portion of the drywell shell- from which the gap-forming material was not removed. In most plants, f l the drywell shell area adjacent to the sand cushion is sandwirned i
l between concrete on the inside and the sand cushion on the outside. At these plants, some modifications in the concrete floor construction may be necessary to permit periodic transducer access to selected portions of the drywell shell. The frequency of the UT thickness measurements should be established from the results of the UT thickness measurements performed during the first two refueling outages and from the extent and nature of the corrosion. However, this
- frequency should not be less than the frequency of containment inspections performed before the ILR1 ss stipulated in 10 CFR Part 50, Appendix J. - - -
i I
, , .-- .._,..~--,,--_a,-. ,.+_,,c'---.n., .',,..-, . . ~ . , - . , . , - , - , . ,,_.,n,-, ,n. , ,nn ,~,.--,n.
3 TORUS (SUPPRESSION POOL)
During the next refueling outage, plant personnel should inspect the
- inside face of the torus (suppression pool) which is coated and forms the boundary of the containment as follows:
- 1. The inside face both above and below the waterline should be visually examinedtoidentifyareasofapparentdegradationanddeposj,tsonthe surface of the steel shell. The use of underwater examination methods +
and techniques may be required.
- 2. UT measurements should be performed on areas identified for potential corrosion due to degradation of coating and deposits on the surface.
- 3. The same visual examination and the associated UT measurements should be performed during subsequent refueling outages.
- 4. If the same areas of corrosion are identified in con:ecutive
- I inspections, the frequency of inspection should be revised on the basis of the rate of corrosion determined from the consecutive UT measurements and from the sample coupons (see 5 and 6 below), and on the past experience considered by the licensee to be relevant to the degradation of the coating. The inspection of such areas should_be continued until there is no further degradation to ensure that the design limit of tbo ,
steel shall is not reached.
- 5. To correlate the extent of degradation of the coating and the rate of corrosion of the shell base metal, representative sample coupons which are of the same' caterials (coating and steel base metal) as those of the l
3 The licensee of Nine Mile Point 1 (which has a unique ongoing. inspection program for the uncocted torus) is not requested to take the actions !
recommended herein.
l
. t shell should be placed at the waterline at least one in each bay (t.
- ween two successive saddle supports) or in each 20-degree sector of the pool.
- 6. Should the rates of corrosion obtained from items 4 and 5 above be different, the frequency of inspection should be determined on the basis of the highest rate of corrosion. The frequency of inspection thus
~
determined should not be less ihan that of refueling outages [
- 7. Unless the i<ientified degraded coating and/or the surface depcsit is removed to perform the inspection, the UT measurement should be qualified by including the coating and surface deposit. A statistically significant number of data points should be used to establish the accuracy of the JT acasurements. The scanning system whether manual or remote, should be specified.
Licensees of plants within less than 6 months to the next refueling
, outage may request NRC approval to defer the augmented inspections for one refueling cycle.
BACKFIT DISCUSSION This generic letter expresses new staff positions which are considered to be backfits justified under the criteria of 10 CFR 50.109(a)(4)(i)
- [ compliance exception backfit]; in addition, the requirement for a response is considered to be justified under the criteria of 10 CF. 59.54(f) (information request].
-. - - , - v - = ,...._.-.Im-._v m . . ._,. -- --e. ---....wv . , , ,,,,w r ,- . ,y,, -m 7:A =1- v- w - w --r -m e-ei- * * * * - -t
In Appendix A to 10 CFR Part 50, GDC for nuclear power plants, GDC 16, ,
50, and 51 require that the containment be designed with sufficient margin
- to ensure that the design conditions important to safety are not exceeded for as long as postulated accident conditions require. The extensive hidden corrosion discovered on the Mark I drywell steel shell at the Oyster Creek Nuclear Generating Station could have reduced the margin and, if not detected in time, could have breached the antainment boundary. This discovery raised concerns regarding the assurance of the margin and the structural integrity of steel containment shells of similar design and construction in other boiling water reactor (BWR) plants. The staff issued GL 87-05 to obtain information on this type of containment. The staff assessed the information obtained in response to the generic letter and, is proposing reccmmendations that, if incorporated in the licensee's inservice inspection programs, would better-ensure maintenanca of sufficient margin as required by the GDC for cortainments and the continued structural integrity of both Mark I and Mark 11 containments that are essential to the protection of the public health and safety. ;
A documented evaluation of the type described in 10 CFR 50.109(a)(6) was prepared to state the objectives of and reasons for the modification, and the basis for invoki".9 the compliance exception. Because the generic letter also
- The term sufficient margin is used in the context of General Design Criterion -
l 50, and is used here to imply that containment functionality will be ,
i l
acceptable as long as the existing margin, based on the strength of the weakest link in the containment functionality chain, is not less than that required by the design code approved by the staff for the licensing basis of the plant.
J 4
4 requires submittal of written reports under 10 CFR 50.54(f), the document also contains the reasons for the information request in view of the potential safety significance of the problem. This document, which is a review package submitted to the Committee to Review Generic Requirements (CRGR) and the associated minutes of the meeting are available in the Public Document Room.
frPORTING RE0VIREMENTS Pursuant to Section 182a of the Atomic Energy Act and 10 CFR 50.54(f) .
addressees shall submit under oath and affirmation a letter within 120 days of receipt of this generic letter containing a statement indicating whether or not the actior in Recommended Actions to be Taken by Addressees. have been, or will be taken. If alternative actions are proposed, supporting justification shall be provided. Each addressee should provide a schedule as well as a plan for implementing the recommended or approved alternative actions.
If you have any questions about this matter, please contact the i
appropriate Project Manager in the Office of Nuclear Reactor Regulation (NRR).
This request for information is covered by the Office of Management and Budget under Clearance Number 3150-0011, which expires June 30, 1994. The :
estimated average number of burden hours is 100 person-hours per plant response, including assessment of the recommended action and preparation of the response. This estimate of the a arage number of burden hours pertains only to these response-related matters that the staff has identified and does not include the time for implementing the requested actions. Comments on the accuracy of this estimate and suggestions to reduce the burden may be directed i
to Ronald Minsk, Office of Information and Regulatory Affairs (3150-0011),
I v
. E
e
?
4
. j NE0B-3019, Office of Management and Budget, Washington, D.C. 20503, and'to the i
U.S. Nuclear Regulatory Commission, Information and Records Management Br anch, Divisior of Information Support Services, Office of Information and Resaurces Management, Washington, D.C. 20555.
k & * *
- Dated at Rockville, Maryland, this 10th day of November 1992.
FOR THE NUCLEAR REGULATORY COMMISSION i jf vb Walter R. Butler, Director Project Directorate 1-3 Division of Reactor Projects - I/II Office of Nuclear Reactor Regulation
- -
- r-Tir--*++- --ww gyg
s REFUELING i CAVITY ,
EQUIPMENT . x,.
[
,t ,
.i POOL r 9e
+ z...
c;.> _ e i
, s -
. , .. . ' at
..h.'. e
,g , ^.,. ;,. f . ,
., 'h p4 ': .43 b, Mk.dh*Oh I- J & "..,k*
F' , " '-
1 h t;rg K g' re Q - c'A
% (. cJa. .etijgN SPENT FUEL ff"&fki 3 Q H +
ygg:7 POOL h 6 ' e,g m . -mu RFACTOR O 3* dwS4s- ~
fhs:wkh.ngkyngsug#d4%W
?5fh%&. . . .
4 Bale & Maedi, n r'-
- i. .W ggtqt. . .
E i.m.m . a.mu.s .9h < a
, : , . :.' m.:
GAP
?. ~ i
- f-.
Q:s, ; - ::, ,
, 1 DRYWFIi>
, i:
.q . - SHIE1D
. BUILDING 1lp i i .
g 3p SAND ClJSHION l
- m ).'
.C>#.b , cc y; -y SAND-Fit i FD TORUS x%
1 l 7 .w.f n y, , s. , .
p .. . ,. mu ;, ,. m , u. y m -;
DRAIN
) c . .~ . .:
ag...;.a,m., c, I 9-s y.5 ce s ,,*.2e z u 4,y W _.g,,; , w%...n ,,-. e,
.;3, il ;
Jm _ c : . - -
~t .~s 9- a
- g. . .v , &. . , . +: < . . .
s,. ,
- nn w. ,
- A 9
,h . ,y' a kn,9 4 MARK I CONTAINMENT
-