ML20045E245
| ML20045E245 | |
| Person / Time | |
|---|---|
| Site: | North Anna  |
| Issue date: | 03/03/1980 |
| From: | Fitzgerald J NRC OFFICE OF THE GENERAL COUNSEL (OGC) |
| To: | |
| Shared Package | |
| ML20038A409 | List:
|
| References | |
| FOIA-92-436 SECY-A-80-028, SECY-A-80-28, NUDOCS 9307010289 | |
| Download: ML20045E245 (78) | |
Text
.
UNITE D STATES NUCLEAR REGULATORY COMMISSION WASHINGTON, D. C. 20555
~
~ ~ ~
- ~^~* ~**
March 3, 1980_-
ADEDKAME COMMISSIONER ACTION For:
The Commissioners From:
James A. Fitzgerald Assistant General Counsel
Subject:
REVIEW OF ALAB-578, VIRGINIA ELECTIRC AND POWER COMPANY Facility:
North Anna Nuclear Power Station, Units 1 and 2
Purpose:
To inform the Commission of an Appeal Board ' M ('
decision Rhich, in my opinion, Review Time Expires:
March 12, 1930 Discussion:
In ALAB-57 8, the Appeal-Boasd > addressed-__the '
- u.'
issue of the service-sater pumphouse settle-ment at North Anna 1/ and concluded that it does not pose a threat to the public health and safety.
Because of the importance of assuring close monitoring of future settle-ment, the Appeal Board required the applicant 1/
The Appeal Board undertook review on its own initiative of the Licensing Board decision to issue operating licenses for North Anna (Units 1 and 2).
It affirmed the Licensing Board decision in all respects except the pumphouse settle-ment issue, the turbine missiles issue, and the generic radon release issue.
ALAB 491, 9 NRC 245 (1978).
The Appeal Board subsequently decided that an evidentiary hear-ing was necessary, ALAB-529, 9 NRC 153 (1979), and this decision is based upon the record of that hearing.
Inter-venor Geraldine Arnold took part in the pumphouse settlement portion of the evidentiary hearing through the conduct of cross-examination.
.rdcra!ol i?. tis rt:ord v;as de'eted CONTACT:
Marian E. Moe, OGC p meWB s Fr@m of MDWO!)
nct, matas 634-3224 F01 A _
-W4 k[O f
~~M.9 930317
- n c,
em
- \\Y
i
-9 2
i to notify NRR within 10 days of any changes in its internal procedures which relate.to the reporting of monitoring results.. F1-nally, the Appeal Board-terminated its juris-diction over the settlement issue.
Since this decision is primarily technical, OPE reviewed it and after review agreed with~the Appeal Board conclusions.
Between 1975 and March 1978 the pumphouse settlement at North Anna exceeded 75% of the
" allowable limit" under the applicable Techni-cal Specifications, necessitating an engineer-ing evaluation.
VEPCO's request to increase the limit followed. The Appeal Board looked' at the safety implications of increasing this limit and addressed four questions:
(1)
The cause of past settlement and the potential for future settlement.
(2)
The level of settlement that might-threaten the integrity of the service water system.
(3)
The ef fect'5fdf a'failui':e"'ih" the service,., J. '
water sysMm "Gyoht) e [ss;fe,ty']6f})e;
[T,',,,
plant.
(4)
The technical specification conditions necessary to ensure that operation of the facility will not endanger the public health and safety.
Each of these is discussed briefly below.
1.
The cause of past settlement and potential for future settlement.
Over the years, the pumphouse_has settled in a pattern that has been difficult to corre-late with any one cause.
Various theories suggest that settlement was caused in part by construction activity, filling of-the reservoir, heavy local rains, and installation of hori-zontal drains, but the mechanisms of the
~
3 settlement are not completely understood.
Fast predictions of settlement have been so inaccurate (the original prediction was 1-1/2 inches, and settlement is now beyond-6 inches), that neither the staff nor the applicant are prepared to rely on new esti-mates of future settlement.
2.
The level of settlement threatening the service water system $ntegrity.
The Appeal Board examined the pipe stress analyses by the staff and the applicant af ter flexible expansion j oints were in-stalled in the piping.
Based on these evalua-tions the Board was satisfied that the expan-sion joints could safely accommodate the proposed increase in allowable settlement.
3 The effects of failure'in the service water system upon safety.
The Board agreed with the Staff and applicant that the postulated failures of the service water system would not adversely affect public health and safety.
This conclusion was based on three premises: '(a) arsettle "
ment-induced break in the service water system is unlikelij~(b) corrective action' could be taken before major problems arose; and (c) sufficient back-up equipment (such as the auxiliary service water system) is avail-able to accommodate any breaks.
4 The technical specification conditions necessary for safe operation.
Based on the applicant's 1978 report and analysis and its own safety evaluation, the staff set four distinct new limits on the settlement of service water system components rather than an absolute limit on pumphouse-settlement.
The Board agreed that these revised limits (which were in effect since l
1 1
.~
O t
i June :1979 subj ect to ' the-Appeal. Board'shfinal review) posed no safety. problem. 2/
In addition, the Board. ordered'the applicant to notify NRR staff at least 10 days prior ton
' instituting any changes,in the internal procedures governing the reporting of:moni.
toring~results.
While the -current : require o
ments are adequate,.the: Board. felt this additional' requirement'was-desirable toL.
ensure that no changes are made without the' opportunity for NRC evaluation.
The Board made two other suggestions.to improve the monitoring offthe settlement'.'.It urged the staff and applicant,to consider-monitoring the expansion joints directly, instead of relying on the current surveying measurements.
It-declined to'actually; order
~this change absent'a. showing that the' current-methodology is inadequate.
The Board ~also; asked the applicant.to install in the near future.a-device'to readily detect leaks in the expansion joint enclosure.unless within thirty days of the date of the order the-l applicant explains why it should-not do so.
'l:
f
+
5 :c 1
-- ~-
.O i
/
JL
\\
Japes A. Fitzgerald Msistant General Counsel J
Attachment:
ALAB-578 j
1 2/
The Board noted'that the original limit had been set ~ based on the applicant's erroneous prediction of lifetime settlement,
~
rather than on the limits of. settlement that'would assure a reasonable margin of safety.
The consequence lln this casei was an overly conservative limit that resulted in the appear-ance of technical specificationElimits being revised because the licensee was unable to meet them.
q u
i 5
Comissioners' comments should be~ provided directly to the Office of the Secretary by c.o.b. Wednesday, March 12, 1980.
Comission Staff Office comments, if any, should be submitted to the Commissioners NLT March 7, 1980, with an information copy to the Office of the Secretary.
If the paper is of such a nature that it requires additional time for analytical i
review and comment, the Commissioners and the Secretariat should be apprised of when comments may be expected.
DISTRIBUTION Comissioners Comission Staff Offices Secretariat g,
Wo e d.
s' YO
..;:. : w +:
e m
- .V..
,'{,- [/ y w,A.,.. ~ ',
Q'.a.,'.,,. - *
.'-,i.~ - > -
's W-
..a.
9' 4 ;,n9
+
< n..*
s.
- t. A.....
..r
,r..
s s
.s...,..
r..
.-T
- v.,.
,...t r.
a.
.c
.w.
c
- x..,. c w,. n.
.+
. s:
- r.,.-.,.,.
.,e,,
4.j,-h.2. 4,[....t - [, s.f'? '. 'I.[.. j a f['
'..'y L. *..J.
g-
~
e;. :,.[/ { '.%..,...., I;.).f.g$. ?..
/,,.4 MS,,a A;,& i:
4;).
,0
.f. w: y,,
N : % M %wC;,.r~.Q. y~.R.'f} f y: n m4..x("p[... y :.. yd s g..
.,.. n
.m a
7 -
.;w s a,
<=
t.
. a,;r s Q; (-Q.s L g
. t>
d.,*
Q,:<2G.Q'.o -.7 Q... g, 3_ t - }',.
- ~c
. f: *q.Q.Q.
} &. Q.
'- f. 9 i
- T'.
.
- 0.:..
- w a n u -~-
.d< 4 f. ::,'.9,.M.' @* a..d... n Q. *;"ny r *.l:' ' ?ey *+.
s as t
.-n w., :.;y w :
y'
>:- 1. s...,
t ..,: g .f.. .'jQ.. P. ' %.. R s.[A;:;,QW *t.v -..,; - v. W -
- x..
a* a... 4.:t,g n s.ls~'. W : ~
- o. -
- (,- _,..e ,4, . f.,....,..;.. 's g,, ,. -v;. t _w, ....w 7,..+.. .m,., v. .;.,~..., 4..>. , +. r....,... ,. m, .,.,o.,. n..s :.f,*L ' m...:. y,;.s ,".,c f..,.~... .,.,...c..,...,.,. < .o. b,,, w..r... p. ,c. a. w P.- .r,,. ,,....y s .. m. a *.,,. .. ~ s. . M J.,U. s .g -1...'. -
- s :L.,, z a.~q\\.,W, pa.3.s >o.
.J.,,.~.~..,- . i p. .a. .. -: i..... s ' * { L.L ;.y ~..L.. . e..,., -,. r. -..... 6 ;. a WY;~g'N*.9 ;. e., i *;.3. -i.f.: *;y p.' :-' '( tg..... w,- ..a ^*.(.. ,e. .f G. c of.., < J 4 (;r,.. d. ,.~
- .~ q. <
.-* s. .;,. x.. *<. : s. c<p 's .. c...., s,% ' L,y,.:g,y-(4. o,.;, Ls. e. ;%.. ,e... d.y., ;;-3... i., .. y Q :!.%m..: e, . A. y -~ n,
- t
- .u.. r, n.;--
+. a g, wt /% .s a.,, :
- 5.,,
4 ,). ' '.s a,. * -.. ,.,u.- ,o ..... _ r s. -. i q ' i ' ' j,y!'Y. y.,, t -* ^ " ' '. ~. .Y=* 's' 1* v .s d, .m-s rz.e ..'.-.s L. .L , l * " ;,y s. c . y. s, $ te-3./ *'~.. ' ' * * ' i. w.' -.l'% ~- .y ,., s. :s es
- +:. - x.n;s
.,".~.i. v r. : w~.g. ?..- J %. ;". ;,c %.y..,,Q%.~ s .p v .. :,. ~. ' .\\ sJ t ~.. ,~ ..., x.. m. e... .~-. ..N.'~q..q, :<p... r p i a-Q..g,., 3 f '.4...' q,, ., ' ',,..y-\\,. ! x.. :.h.,. i. '.?..,.,...,D.. 4,.+At t achn.ie'n.t' . ^.., . t,. v \\ s.
- ?,. 2
... -. na j..;~y-4 .~ ..a .j, s/. s. .c q. -).... .W 3-....,...,. .., r. M., .e .,A . ;j,,-;, pa. s. c s ,E* 2... % r .g.r . {c' .,a.. . ~ '., s -J '..n 's,4,,- ., v 1 ~ .e >~ l. [5., s. " m' ,.A {..-l.,,'.;.',',*.,'- f .,s g
- s s, *,. i.. ' * * ' 4
. d.. ]
- s e'>
=.. ,.s* at c.,
- 'g l * *. o
.le g. q e, { f '*,, t. .' *,s,.
- ".f f
'?.* * * - /* * % 8., 6 ^ i e s J-y .s,....,.. '- 2 r .~o... , - :' z,.,.. w s . e, '9 6, f '.*. i d*l . '. t ,, y 'e- ..s 3,_, 4 ,~ -r,,..
- *., ' + -
"r
- n.
- fo.
g --3. f) g.#[ .hs'.8'[,'[k-[ -.b '"E '} ; .t ,.,,,. [a%.' I
- ~
9 e. . < >. \\.
- s.k.*'.
D' g a s s 1 n, s,r s.s.s.'.,, s'1,.
- s. - L '. s..,
. a 's ".,..,- T *-(:)a. s ,y, ..%s. w r s ;,J2 ~.-.-. ; @.s,.q,t
- > S V.. %. 9% Tf..
. ~.. .- /
- . 's.b. '
~ .<e 3 se. ~~. . p~... W., .,,b y Q ; y.a n.. e ' '. h,. ',,. 1 ,'.- -> W u.:. 4.; i-3 3 y. m,+. 4.,; *.,..v s 3 e.a s... <4 y p # p., 2' y, .s. .m ;.?' .,. w. 0: r s<... ..a ,..a ..s s i... u % r..t, c,,,. q% r..,,c'.,. - r ^ .h2.. s. .*+Y- .. ) ~r.E,. t ".. L L_'M %sptsh,p'W. n=.,.g.Lz.i;.k..' g.? @ ..9 .',g - J m = ^ ' 8,e + w, /. s s.*s :.'. M~q.,. w (... p. .O -w s,., f - 3..,..s K *w,.e 2..K) s.s. ~ W... ..;w y* s m ~., ~
- Pt"Tr&'m.i', s y s, w-
... - ~ s.,,n, y'.....y c%.yt.Q -s
- o. :
a.o u WL'4*i&W,1hy "T'i.(,.. n. '%. . Mk= g ** ' '..9: -4 u. -*,'~;*--^v-r.wi'~'n C~"l6,. -p% %: <<4
- *;.p: - - 5 *^ ~
~ ',. ~ ' - - s~ s Y - 4' e.H A. J. - --~ . Rr
- A
'" O f.Y@...p,.y ..&'s;..* :.wQ..;y,..... Q Q<W; w;M" p;.M QWh. {'4* r.,_.r;M'%* 2 @H.&;L)!@ g F
- -~
s; .. u. : J. ,.W&C' -s
- J
- 'T
.t Qy:j.+ l J ; %. f.. t s-p = W:y y a 7 .e 6f n *- v*-,.- .s, :?.. '2;}. e L l, %. 7 g.] _e' 4.,,;. > s s' .~..*!. . ; *'. 1 - j ',., N:., 4.g..'.;*. *,.g....'f ; ; y . "e . (',() - ).pg,y.-Cy. A*. r y. 3.p' ;'4 9 ,s ..c.-.....; 5 ',.. ?'.'.t. ~ {- h?'g.f'. {..,et.v.. },,: '~ - k. f E. '?! '} ~ h' h ./. -' f .s r ;. :- jt,. J , q / 1.'.;;. 9.. < ~ R. e . 4p..~ ' y -(. 4?..Q', ' ',3'-l. - d' r s
- 7
. Q'":" :.' a 4 /*
- ~
h*. N..y ?f.,g.*,'.J,.'d.. '. M. 7y,g%.' % p %.. v, ? . ~..' ?,c. r.7.,. V,~.,. f's 4 c - % J.9.g, :. R:. 3 y, c. r; w.,',,,.,,. y,.. ~ ' ~ .,..,.3.,.y3. ?. 2,,3 rP 't. w. . ~.. .y -:: r e..,,. t.: s. ?...., % s.- g c q..., w.. m....Fs, ? .e,v ~.....,. / s
- g. e(&,. ) ' '
.+ .e .. 4 3.y 2 ). C' ? g.,Q e *' *, Q< '2 i .' l. e; 3 h **ff 4 4,. e. y, +. ssu..', k.' c A-f. '_e r ,wm T. f g.f m :.T -e f* $..-* '.. i . o.,. L Weys c-
- . y
. a,, y.h, . Q. 3 q-w.Qm *j. d,:, 9.s..'.%,%'.,,Y's Q f }Jy ; r4,,. --.. y '.. . s:rp _.- W h. M* & y-**h. Y.IUM[is.. .[ . e. :,.
- s -...
~Qe s-
- f*
~ l\\ k. .~. $.A":Y'%, 4. l.h"b. ? t,.' i I . f.'f Y M .M,',' ,.'~.~ l.f'f e .?M.N. finn 3'3.Wik'.b.TUM.t f'%..,. . A,.,s h.$'s ~ ->* ~ t i 3 , m... Y
- .. ~.-\\,",>.
'p..%...
- 4*
,.e. to ,...J.+ , s. 2. i.%.. y ' %. 4 p. .sc,.,~ 3 ,g ' - *, 5.*:c* *.< L,*'i,f*t' yv}. g.t."?.. ! u,;,.*A,*. x, <. [~./'
- g..
. ' e, *4,.s ': d ..L. ..*r~.'. ' l g'.!.,..t fr.- s n "* 1,. ,o n-r! 4, ~ ~ w...h...g :"'... ..w - .s 0 -
- ** L'. s s
nt w ;,, c..f.;* W.g G L '. < v d.. - q. *; 'M,,.'. r 41
- a4 L
(- .,g 1, c:.,J. p?'- '. 'h; : . j,.: Dw. N?'.1 y.4 w: - ' '.~* C. c p;.,:. 4.h.d'.f@.,hp. W. k,.'w%'. ?.3 5 s,,d.f,p.. QP W.'J;* t *W.. i ~ w** ~
- u
,9-
- s. L C*
% %.:%'3: k* 's. . :.f.' ? C' .. g%) M.f 4.. %.>m. vg,...,-H y;4.....% p * ;'+ W:4..y..- *.. aM d,. 4 * ~... v,. : r.. . S h.y S Un,. <- Q.,.A Y i*. , > y v..,. v,~, ,=:. - r w a f~f.'.. m,. >/r!. *.:,e...., a .a 4. ,,p ; cc $- ~ ; ". s3 . l+W. n.,,,% W} V.?.. y T *.$j.b.., N q. c-. .. m% y%w.s,)...[.M;4 ; *L,.,g J- ' %r.;n : f :.
- r_,:'..3
.J. s ,X s -~ .p c n.- .',%..;<9.; :y w. g- ..m..,.s.,L. ., m, ~.. m. M.. *..* n.
- .. w i
.r.- s . '...i.a
- : ; :.
- p..Ce.;.#
i*.r .. g... .s 5 : . <;s.yfl *. @ ;:. y.*,vj'.x y? . g. r.w4 *l.f L* y
- =
- .,p% h h ; Q / ]./* (tN y p-- 4-,'y..
j '[. Nj'y.. .4 n.- y.. @f '?.h &uq,~ >,i.MlgG%gp. A:pk w[?.& q.lll1 c:.. * ;'.Y [< h + .4. *..a *;" N. x......, ). w* y,p %,NW5" Q . 'h*, k A. l: '. - :. .t' .s.*. Y b,p& ,,.s.:.<: ' ? ;-:: k W.;p4.. j*~w.'*'.M; :* w-W' O a; h
- e
%g h e t x i =. \\< ,. % n Q W m y .'t ~,:
- f.,,.
_, s 4. ; ? %
- '..t
~ C ~ M.f.%q,:w,:s &.-K.% Q)V *- w.. ', ;m.?..D}y.,%.W'2..e 4 ?%,,..,_ 41 e ',~,p r- . ~ - y?>_f. :.r y @.. -
- p:
.n s-
- 1.,:. \\,., %., <
f.,,.,,". %_ ', &,- ~..G..r.gf;dQf fier.J., f.q'*?... (* ' " fy...r.,...G., a.W.. :,e~ n.' . ;.,1... ~ JT. -
- l".. r <j.. >....... s.
' we _ > *W, s? ~g,. 3 ; 2 j' ' s,.~e- f. 3.
- * *l, e'
e,p. ,n Zo , k. 'r -f.. *3 /,, O,..ty %. ;.-. * :.'4, ? 4' ^ *,. .x ..,t.. ,,;*l ? 'c t* .,..+ , %,. 3 g ,, - e / 5 N UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISS po0M,. // ust*~ ,1 ATOMIC SAFETY AND LICENSING APPEAL BOARD .2 2gggg y ~f pEB 1 Alan S. Rosenthal, Chairman gttSgff'l Dr. John H. Buck Offjg g Michael C. Farrar O 4 -a ) In the Matter of ) ) VIRGINIA ELECTRIC AND POWER ) Docket Nos. 50-338 OL COMPANY ) ~ 50-339 OL ) (North Anna Nuclear Power ) Station, Units 1 and 2) ) ) Messrs. Michael W. Maupin, James N. Christman and James M. Rinaca, Richmond, Virginia, for the applicant, Virginia Electric and Power Company. Attorney General Marshall Coleman and Assistant Attorney General Anthony Gambardella, Richmond, Virginia, for the Commonwealth of Virginia. Mr. Richard M. Foster,i. Charlottesville 3 Virginia,..for_ m,., _ the intervenor, Geraldine 'Afnold.~ ~ ~ Messrs. Hshry J."McG6rfe'n, Daniel T. Swanson and Stuart A. Treby.for the Nuclear Regulatory Commission starf. DECISION February 11,1980 ( ALAB-57 8 ) In December 1977, the Licensing Board authorized the issuance of operating licenses for Units 1 and 2 of the North Anna. facility. LBP-77-68, 6 NRC 1127; see also LBP-78-10, 7 NRC 295 (1978). No exceptions were filed by any of the parties to the proceeding; accordingly, we undertook to i review the decisions below and the underlying record on our own initiative. On that review, we concluded that additional information was needed on two safety issues: (1) the signifi-cance of the past, and potential future, settlement of the ground beneath the' service water pumphouse; and (2) the like-lihood that turbine missiles might strike and damage vital facility structures or components. In all other respects save 1/
- one,
-we affirmed. ALAB-491, 8 NRC 245 (1978). ~~ Upon receipt of the requested information, we found ourselves unable to resolve either of the open safety issues without the benefit of an evidentiary hearing. ALAB-529, 9 NRC 153 (1979). Accordingly, a three-day hearing was held in June 1979. The applicant and staff produced testimony on both issues. Without adducing affirmative evidence of their 4 ~ m mus.ern,,... 2,, o r ; .m ..a -.3 own, the intervenors conctonwealth of Virginia and Mrs. Geralalne,e Arnold participated through the conduct of cross-examination in that part of the hearing concerned with pumphouse settle-ment. Subsequent to the hearing, all the parties submitted proposed findings of fact. 1/ We deferred consideration of the generic issue relating to ~~ the environmental effect of the radon releases associated ~ with the mining and milling of uranium. That issue is receiving our active consideration in a number of other proceedings and will be the subject of an evidentiary hear-ing later this month. See Philadelphia Electric Co. (Peach Bottom Units 2 and 3), ALAB-562,10 NRC (September 10, 1979), ALAB-566, 10 NRC __(October 11, 1F79). 1 i . It was our original intention to address and resolve both issues in a single decision. But new information of potential importance to the turbine missile issue has recently been brought to our attention, requiring us to withhold our disposition of that issue to await further developments. Our decision today thus denis only with the pumphouse settlement 2/ issue.-- For the reasons set forth below, we conclude that sett3 cment of the service water pumphouse does not pose a tnreat to the public health and safety. I INTRODUCTION The source of our concern here is the unexpected magnitude of' the settlement..that has Jaeen, experienced over a period of years by various parts.of the-North Anna facility's service water system (SWS) in the vicinity of the service water pump-- house. The SWS is one of North Anna's numerous cooling systems. Each system performs its own distinct functions; most of them (including the SWS) have their own independent components (e.g., piping, valves, pumps) and source of water. In evaluating --2/ We shall deal separately with the turbine missile question, including the matter of whether plant operation can be allowed to continue pending the ultimate resolution of that question. the safety significance of a potential failure in the SWS, one must understand exactly what role the system does and does not fulfill during normal plant operation and under accident conditions. 1. The service water system does not provide water directly to the reactor core -- this function (in pressurized water reactors such as those at North Anna) is performed by the primary cooling system. Nor is the SWS related to the secondary cooling system -- the system in a PWR that removes heat frc= the primary coolant to produce steam for driving ~ the turbine generators. And, as will be seen, a failure in the SWS would not pose the same immediate threat to the public health and safety as would a failure in either the primary or secondary cooling systems.; This is not to say that-the-service water system 'is unimportant to the safe operation of the facility. During routine plant operation, the system provides cooling for 3/ (1) the component cooling system heat exchangers; (2) the main control room air conditioning condensors; (3) the lubri-cating oil and seal coolers for the primary cooling system 3/ The component cooling system provides cooling for, among other things, the residual heat removal system (RHR) and ~- the reactor coolant pump motors. The RHR is designed to control the reactor coolant temperature during normal (i.e., non-emergency) reactor cooldown. For a description of the component cooling system, see the Final Safety Analysis Report (FSAR), Section 9.2.2. .- 4/ charging pumps;-- (4) the service and instrument air compressors; and (5) the primary containment pipe penetration cooling coi1s. The SWS also serves as a backup supply for the steam generator feedwater system, the fuel pit coolers and the recirculation air cooling coils. In the case of an accident, the SWS is 5/ principally used to cool the recirculating spray water and to provide an ultimate heat sink. --6/ A secondary (albeit important) function under accident conditions is to supply cooling to the charging pumps, thus ensuring their continued availability. The basic source of service water for the North Anna plant is a 9.5 acre man-made reservoir, built on naturally sloping ground about 600 feet south of the reactor building. Roughly 8 to 10 feet deep, the reservoir holds approximately u.~4..,<. 2.n. - v. .. = :.. 1 = hr : ~~ ~~4/ The charging pumps (which are part of the chemical and volume control system) provide a means for injecting coolant into the reactor primary cooling system. They play an important role in, inter alia, maintaining the proper reactor coolant inventory during all phases of operation. They also inject high-pressure water into the primary system during a loss-of-coolant accident. --5/ In accident situations, the recirculating spray water system is called upon to maintain the pressure within the containment building below atmospheric pressure, thereby preventing leakage of radioactive gases from the building. --6/ According to Regulatory Guide 1.27, "the ultimate heat sink should be capable of providing sufficient cooling for at'least 30 days (a) to permit simultaneous safe shutdown and cooldown of all nuclear reactor units that it serves and to maintain them in a safe shutdown con-dition, and (b) in the event of an accident in one unit, to limit the effects of that accident safely, to permit simultaneous and safe shutdown of the remaining units, and to maintain them in a safe shutdown condition." l . 7/ 22,500,000 gallons of water. At its eastern end, it -~ has a "U" shaped dike. The dike has an earth core which is surrounded by rock fill. Two " filter zones" located be-tween the core and the rock fill are designed to preclude The bottom of the seepage-induced erosion of the core. reservoir is lined with two feet of compacted clay which extends up the inside slope of the dike. The purpose of this liner is to minimize seepage. App. Test., p. 4; see 8/ -~ also VEPCO Figure 3. The service water pumphouse is a large concrete struc-ture (61 feet by 64 feet) embedded in the crest of the reservoir dike. It contains four pumps, each of which has a normal capacity of 11,500 gallons per minute (FSAR, 9.2.1-8). The service water.tisc. pumped from.the rreservoir, first through one 9/ (or both) of two redundant, supply headers ,,,and_ ;then,thr,oug,h,, 7/ This is sufficient water for 30 days of SWS operation for four units without resort'to rekeup water for losses ~~ due to evaporation (FSAR, pp. 9. 2.1-12 and 9. 2.1-13). 8/ The prepared testimony introduced into the record at our hearing is referred to in this opinion as follows: Applicant's Tables and Figures....."VEPCO Figure (Table) Apolicant's Testimony.............." App. Test. Applicant's Supplemental Testimony.." App. Supp. Test. Staff's Testimony.................." Staff Test.__" The transcript is referred to as "Tr. _/ See fn. 54, infra. 9 O .,. one of two (buried) 36-inch diameter supply lines, to those components which are cooled by the SWS. Thereafter, the now-heated water is returned through one of two (buried) 36-inch diameter return pipes to the pumphouse, from whers 10 / it is discharged via a spray system to the reservoir!- Backup service water can be supplied by using pumps which draw water from the circulatinc water intake system located LL/ on Lake Anna -- a much larger man-made body of water located on the opposite side of the facility from the reservoir. This backup supply is designed to satisfy the normal and emergency service water requirements for Units 1 and 2 (FSAR, 9.2.1-10). In addition to providing an independent supply of service water, Lake Anna can also serve as an ultimate heat sink for the plant (FSAR, 9.2.1-1). . :, = = a G-i ~= = = = = = = = - - = - m Aswill'beie'en'iPomourdfscusAihn"of"the5 hide 5hel !; ~ 2. later in this opinion, certain components of the service water system have settled in varying amounts during the eight-year ~~10/ The spray system is designed to facilitate the removal of heat from the service water as it is returned to the reservoir. 11/ The circulating water system provides water to the con-densor where it is used to cool the steam leaving the- ~~ turbine. generator. This system uses its own independent pumps. Should Lake Anna be called upon to supply service water, 'different pumps located at the circulating water intake structure would be used and the heated service water would be returned to Lake Anna via the circulating water discharge tunnel and the cooling canals. The two pumps at the intake structure are identical to those at the service water pumphouse. . period since the construction of the pumphouse began. For its part, the pumphouse has settled more than half a foot. Surveillance requirements and allowable settlement limits for all " Seismic Category I" systems and structures 12/ at North Anna are to be found in the f acility's technical -~ specifications.~-13/ It is there stipulated that if 75% of the allowable settlement limit for a particular structure or system is reached, the utility must initiate an engineering review of the problem. If that limit is exceeded, the plant must be shut down. See Technical Specification 3/4.7.12. --14/ The North Anna technical specifications, issued in November 1977, decreed that the settlement of the pumphouse . n. c.= _ _ = cm.....,, _ 12/ According to' Regulatory-Guide L1329,nseismicrDesign --m Classification, a Seismic Category I designation is ~~ applied to "[t] hose structures, systems, and components that should be designed to remain functional if the SSE [ Safe Shutdown Earthquake) occurs...." The service water system is so classified. 13/ Technical specifications, which are imposed upon a licensee by the NRC, establish various requirements designed to ~~ ensure the safe operation of a nuclear plant. l 14/ Other Seismic Category I systems and structures at North Anna monitored for settlement include, inter alia, the ~~ containment (reactor) building, the service b ITding and the auxiliary building. Moreover, although it is not a Category I structure, the turbine building is also moni-See Technica tored for this purpose. Table 3.7-5. Unlike the pumphouse, these structures do not rest upon a deep bed of saprolite. See discussion, pp. 11-12 infra. Although some of them have settled to some extent, (FOOTNOTE CONTINUED ON NEXT PAGE) 4 15 / after December 1975 could not exceed 0.15 feet (1.8 inches). By March 197 8, the settlement had exceeded 75% of that " allow-able limit." At that point, the applicant performed the required engineering evaluation and thereaf ter requested that the allow-able limit be increased. The safety implications of the proposed increase were the primary focus of our hearing. In assessing whether the limit could be increased safely, we concentrated on the following four issues: (1) The cause of past settlement and the potential for future settlement. (2) The level of settlement that might threaten the integrity of the service water system. (3) The effects of a failure in the service water system ',upon th'e'7s a?f e'tV ~o.~f '.t.h.e~pla nt. '.' ,,.' ' "m.,ir a,", I=., ~ ~ i z g (4) The technical specification conditions necessary to ensure that operation of the facility will not endanger the public health and safety. 19 (FOOTNOTE CONTINUED FROM PREVIC'JS PAGE) none has experienced sufficient settlement to warrant ~~ further inquiry by this Board. Sea Tr. 131-51. In this connection, the highest percentage of the technical speci-fications limits reached by the settlement of any of these buildings is 47% (at the south wall of the service building); the containment building (founded on rock) and the structure housing the backup service water pumps at Lake Anna (founded on a thin layer of saprolite) have not settled at all (Tr. 14 5-4 7). 15/ This figure corresponded to the prediction of future settlement contained in the FSAR (at p. 3-7 of Appendix E ~~ to Amendment 44, dated December 29, 1975). The "FSAR prediction" was specifically referred to in the technical specifications. See Technical Specification 3/4.7.12, pp. B 3/4.7-7 and 7-8. ' i II THE EVIDENCE OF RECORD In our order directing the evidentiary hearing, we asked the parties to include in their prepared testimony: a self-contained synopsis of settlement history; an exploration of the relationship of the settlement phenomenon to public health and safety; a discussion of the relevant soil mechanics; the basis for the staff's dewatering requirements; details of the settlement monitoring program; and additional details of the stress analyses performed in evaluating the effects of settlement. ALAB-529, supra, 9 NRC at 155-57. These subjects were all addressed at the hearing. Rather than treat each of them independently, we have arranged our
- c. ~
- w...
^* rs. :. =..- discussion of the evidence to correspond roughly with the four areas of primary concern outlined at the conclusion of part I, supra. A. History of Settlement As mentioned previously, our concern regarding the integrity of the service water system stemmed from the un-t anticipated and unexplained magnitude of settlement of various parts of the system. We therefore find it appropriate e . to set out the settlement history in some detail. We shall focus upon the settlement of the pumphouse in this discussion because most of the settlement af fecting the service water system piping has taken place in the area closest to that structure and because the pumphouse has a more detailed monitoring history than any other part of the system. We have attempted to correlate incidences of settlement with construction activity so as to gain an understanding of the causes of settlement at North Anna. The difficulties we have experienced in this endeavor parallel those which the applicant and staf f encountered in their efforts to comprehend the problem. 1. The pumphouse and the dike upon which it rests are ~ ~" situated atop a layer of com'pressible'saprolite, sixty-Yive' feet in depth. " Saprolit'e iso '"sofE,- earthy, clay-rich, thor-i~ ~ oughly decomposed rock formed in place by chemical weathering of igneous and metamorphic rocks." American Geological insti-tute, Glossary of Geolocy, 630 (1972). The saprolite at North Anna is composed of grains of hard angular quart:; grains of f eldspar partially altered into clay minerals; and bands of mica particles (App. Test., p. 38). Although the saprolite contains a large percentage of clay minerals, applicant's h ~ . witness Bruce N. MacIver testified that those minerals are 2 cemented together-into larger particles which behave more 16/ like silt than clay (Tr. 63). Saprolite, produced by the weathering of granite gneiss 17/ rock, differs from transported soil.~- Transported soils exhibit a uniformity of individual particles, allowing the soils to be classified according to average properties and thereby enabling engineers to predict their behavior with some degree of accuracy. Saprolite lacks the consistency in character and arrangement of particles exhibited by transported soils. With its more complicated arrangement of different minerals, saprolite's response to changes in stress cannot=bk ptedicted" accurately byumechanica.l_models., ppI 3N3I.^ ' ' 'I '"- ~; ~Y~! ; P ' ' +;iC Yk:1 W. App. Test., In its supplemental prefiled testimony, the appli-2. cant touched upon some aspects of settlement theory (App. Supp. Test., pp. 4-13). While there was disagreement among 16/ According to Mr. MacIver, this makes saprolite more resistant to settlement, because silt is a stronger ~- material than clay as far as loadings are concerned ^ (Tr. 63). 17 Transported soils are developed from decomposed rock and then modified during the various phases of the trans-porting process (e.g., erosion, sorting, sedimentation, and consolidationT(App. Test., p. 37). _ 13 _ the parties at the hearing respecting the reasons for the unexpected magnitude of settlement at North Anna, this treatment of the general causes of settlement was not questioned. According to the applicant, whenever a new structure is placed upon a compressible material, the application of the contact stress of the structure (i.e., its weight distributed over its foundation area) upon the underlying material should cause the structure to settle. The rate of settlement is directly related to (1) the average com-pressibility of the underlying material; (2) the thickness of the material; and (3) the increase in the " effective stress" upon the" material. Effective stress is, in turn,
- c~~
..v>-~ +nn .~m n ~ ~.. + - a function of (1)' 'th,e,c'ontact" stress r (2)' thr total weight' ' "" ~ (soil particles plus water) of the soil above the groundwater level; and (3) the buoyant weight (total weight minus the weight of a corresponding volume of water) of the soil be-low the groundwater level. App. Supp. Test., pp. 4-5. If the effective stress remains constant, a new structure should settle rapialy for a relatively brief period of time (this period of settlement is called primary consolidation) and thereafter settle at a slow, diminishing rate (secondary consolidation) (pi., p.
- 5).
, Only two factors should cause an increase in the rate an increase in effective stress or an in-of settlement: crease in the compressibility of the underlying material. If the contact stress remains constant (as it should once construction is completed), effective stress can be in-creased only by increasing the water content of the material above groundwater level or by lowering the groundwater level itself. Id., pp. 5-7. An increase in the compressibility of the material can sometimes be effected by an increase in1M the degree of water saturation of the material (id., p. 11). Because of the nature of the soil underneath the 3. pumphouse, it was expected that some construction-related settlement would occur. Before any g t getion began, the consulting firm of Dames % Moore. con' duct.ed a number:of; soil; -c 18/ The record does not make clear the circumstances in which incressed water saturation leads to increased compressi-bility. Mr. MacIver testified that the saprolite at North Anna does not exhibit greater compressibility upon the introduction of large amounts of water (Tr.195-96). He thus rejected the suggestion that heavy rains in 1975 might have increased the compressibility of the saprolite and been responsible for a large amount of settlement. The staff was of the different belief that increased water concentration could weaken the saprolite and re-sult in greater amounts of settlement (Staff Test., pp. 26-27). As the exact causes of the settlement at North Anna remain a mystery, neither party's view can be taken as conclusive. r- . borings and laboratory consolidation tests. Based on these tests, applicant's engineering firm (Stone & Webster) esti ' mated that the pumph6use might settle as much as 0.12 feet (or nearly 1-1/2 inches) during the life of the plant. App. Test., pp. 7-8; FSAR, p. 3.8-133. Construction excavation for the pumphouse was com-menced in January 1972. Two months later, Stone-& Webster began pouring concrete for the bottom mat and walls. Concrete for the operating floor slab was poured across the top of the walls on August 25, 1972; this date marks the start of appli-19/' cant's settlement monitoring record.~~ During the next two months, the pumphouse equipment was installed and a three-foot layer of : clay._ fill was..compaeted agains.t ;.the pumphouse u - = + : :.
- =
8c==lr== " :" :- ' walls. By October,18,.1972,i nostafrthwcontactEstress 41adw== : been applied at the site. App. Test., pp. B-9. It soon became apparent that the amount of settlement would be much greater than Stone & Webster had predicted. 19/ Pumphouse settlement is determined by measuring the ele-vation of the corners of the operating floor slab and comparing the measurements with the elevation of the slab on August 25, 1972. Five points on the slab were originally selected for surveying -- two are situated at the northeast corner of the pumphouse; the other three are located at the northwest, southwest, and south-east corners. As set.tlement increased, additional points were selected for monitoring. For the exact location of the original and additional monitoring points, see VEPCO Figure 6. indicated The.first measurements, taken on December 4, 1972, that,in little more than three months, the pumphouse had already experienced its anticipated plant-life settlement 20/ After reviewing the situation, Stone & of 0.12 feet. Webster nonetheless determined that construction should continue. Id., pp. 9-10. By April 11, 1973, Stone & Webster had completed pouring the concrete for the upper walls and roof of the pumphouse. i Settlement continued at a fairly even pace through the end of the year, reaching a total of 0.195. feet (2-1/3 inches)- Settlement then appears to have levelled by January 1974. 20/ Settlement..of the_,pumphousg. bas been expressed as an_ average 7 figure. The settlement at each of the four corners the. northwest.. corner. _n has proceeded: at4differingaratest has settled"5ubstantially f~asteb~ than he otheEs",'WIl'e"' the southeast corner has settled at a gubstantially lower The December 4-measurement, for example, showed an rate. average settlement of 0.12 feet, with a settlement of 0.223 and only 0.019 feet at the feet at the northwest corner These differing rates of settlement southeast corner. have produced a noticeable tilt in the pumphouse towards Tables showing the settlement figures for the northwest. the individual monitoring points are found. in VEPCO Table - 1; a graph describing the average settlement is shown in Mr. MacIver testified that VEPCO Figures 7A through 7G. the measurements are probably accurate within five thousandths (0.005) of a foot (1/16 of an inch) (Tr. 85). The foregoing discussion should not be taken as implying our necessary agreement with the use of average settle-ment of the four corners of'the structure, rather than a more appropriate measure of the severity of the problem. J As will be seen, average settlement measurements are no i longer employed. See p. 50, infra. . off until fill material was placed for the reservoir dike-between March 6 and May 10. When the dike was brought to its crest on May 10, the final structural load had been added to the foundation. A reading taken on June 27, 1974, indicated that the pumphouse had then settled an average of 0.265 feet (slightly over three inches). After June 27, the rate of settlement slowed down some-what; the incremental settlement over the next five months totaled only 0.03 feet (3/8 of an inch). In December and January, however, the rate increased precipitously. At the beginning of December, the average settlement measured 0.295 feet; by early February 1975, it was recorded at 0.38 feet (an increase of one inch). This increase occurred without any
- o
- :.w
.f 2 A..'-y = w w :;.::.a m 5 _ v. v e :, m,.2ni: additional load being plac_e_d upon_ the saprolite.. Although the applicant increased its monitoring frequency in 1975, no additional settlement was registered until the end of that year. As in 1974, December saw a noticeable rise in settlement without any additions to the structural load. That is, the average settlement remained at 0.38 feet until December 2 L/ 20, but was recorded to have increased 3/16 of an inch to 0.395 feet by January 2, 1976. I 21/ This later became the base point with respect to the future settlement measurements required by the technical ~~ specifications. See pp. 8-9, supra. l . No additional settlement was registered until August 1976, when water was first placed in the service water reservoir.. The water in the reservoir was brought to its highest level in early October 1976; the first dewatering drain was in-22/ stalled at the same time. The survey records for the end of 1976 show a slight increase in settlement in late August (0.005 feet), no increase in September, a larger increase in October (0.015 feet), no increase in November, and a sharp increase during the first two weeks of December (0.025 feet). The applicant attributed this ' settlement to the filling of the reservoir (App. Test., p. 17). (such an explanation may seem adequate for the August and' October settlements, but it hardly explains the sharp increase in 23 7 December preceded by; the, ab_sence. of s.et.tleinent.,Jn. Novgber;,.)-~~.:.- nc. .:, m.s. m.. : " t m ? =: = *. =....- -.. z. 7. " ~:. : -n.7..- . L. .a ..=- m :=: :._ w 1. 22/ Af ter the rapid settlement that took place during the winter of 1974, the staff instructed the applicant to install six horizontal drains in the area near the pump-house. The installation of these drains was designed to reduce the (then perceived) effect upon settlement of heavy rains and to avoid drastic fluctuations in ground-water level. See pp. 21-23, infra. 23/ The water reached an elevation of 313 feet in the reservoir in late August (elevation of the bottom of the reservoir is 305 feet); the reservoir was drained to almost empty through-out September; upon refilling, water elevation exceeded 315 feet throughout most of October; and the elevation hovered at slightly less than 314 feet throughout November and the first half of December. The settlement history is con-sistent with the status of the reservoir until December. (FOOTNOTE CONTINUED ON NEXT PAGE). . No new settlement was observed until the installation i of the remaining horizontal drains during the summer of 1977. Drains 4 and 6 were installed in early July. De- ~ tween July 11 and August 3, 0.048 feet of settlement was recorded. The installation of the last three drains (Drains 24/ 2, 3, and 5) was accompanied by no further settlement. The next 11 months saw no further measured settlement. In late June 1978, a gradual rate of settlement resumed. From June 30, 1978, until June 20, 1979, approximately 0.025 feet of settlement was recorded. Settlement then appeared to pick up in the next two weeks, totaling approximately 0.007 i 23/ (FOOTNOTE CONTINUED FROM PREVIOUS PAGE). t h a t v :- c- . _ -. =...,..... That month, instead of the ne'gligible' amount'of settlement which would have been suggested by the previously exhibited water level / settlement relationship, survey records indi-cate that the pumphouse experienced its sharpest drop of this whole five-month period. Subsecuent changes in reservoir water level had no effect on settlement. VEPCO Figure 7E provides a comparison of the reservoir water level and settlement. 24/ VEPCO Fi'gure 9 depicts the location of all six drains. If, as applicant has claimed (see p. 23, infra), the drains encouraged settlement by 1cwering the groundwater level, Drain 4 is likely to have had the greatest effect on settlement because of its location directly beneath the pumphouse. I 20 - feet. The most recent measurements submitted to us, taken on July 18, 1979, indicated an average settlement of 0.523 feet (a little over six inches) since the pumphouse floor 25/ was poured in August 1972, or 0.143 feet since December 19757~ As is thus seen, the applicant's predictions of the course of pumphouse settlement were far off the mark. Once again, before construction had started in 1972, the applicant 26 / had predicted a total settlement of 0.12 feet. In late 1975, it forecast that future settlement over the entire lifetime of the facility would not exceed 0.15 feet (and that forecast was carried over into the technical specifications 27 / as the allowable limit).-- As it turned out, however, by the time of the hearing in mid-1979, pumphouse settlement had _,, ;, m. g _...__2.., already exceeded' 95i~of tliaiti value and was continuing. m, In ..;._..,,z.,._... r n.=. t n.s 1. _- _. these circumstances, it is hardly surprising thst, at^the" ^ ~ hearing, neither the applicant nor the staff was prepared to 25 / The last measurements we received for individual points were taken on June 4, 1979. At that time, the nortnwest corner ~~ had settled 0.719 feet, the southeast corner 0.293 feet, and the northeast and southwest corners 0.517 and 0.505 feet respectively (an average settlement of 0.509 feet). The July 18 figure for average settlement given in the text, was furnished by the applicant in the letter of July 23,,1979,i accompanying its Memorandum of Proposed Findings. j! V See p. 15, supra. j j!j'/ See fn.15, supra, and accompanying text. . continue to rely upon predictions of future settlement. 28/ See Tr. 291-92, 341-457-4. The mechanisms of the discerned settlement are also not completely understood. To be sure, some of the settle-ment can be directly correlated with construction activities at the site. But its course over the years has not adhered to the theoretical pattern of rapid initial settlement followed by steady settlement at an ever-decreasing rate.
- Instead, the pumphouse has experienced a step-like pattern with long periods of relative stability followed by short periods of rapid settlement.
In this regard, many of the latter periods do not appear to have been coincident with additions to the structural load. _ _ _u _ _ _. I From time tortimef various"expl& nations for7this-step 3"~ like settlement activity have been suggested. One theory had it that heavy rains might be weakening the intermolecular structure of the saprolite. See fn.18, supra; see also App. Test., p. 16. For its part, the applicant now believes that 28/ Settlement measurements of the exposed ends of the service water pipes north of the expansion joints, taken since July,1977, indicate that the pipes have settled at a greater rate than the pumphouse (App. Test., p. 21). No explanation has been given for this phenomenon. We discuss it further at fn. 41, infra. , the rapid settlement experienced during the winter of 1974-75 was caused by a delayed reaction to construction activity and not by rainfall (Tr. 47-49). On the other hand, the staff is still persuaded that heavy rains might have caused'the rapid increase in settlement (Staff Test., pp. 23-31). A correlation of heavy rainfalls with settlement figures gives some plausibility to each hypothesis. Heavy rains in December 1974 immediately preceded a large incremental settlement, but equally heavy rains in September of the same year failed to alter the rate of settlement. Additionally, several rains in 1975 apparently had no effect on settlement whatsoever. See VEPCO Figures 7C and 7D. Because of its belief that heavy rains might have been . W: 1. . &91 a95ec56 rf rd...'i. M m 'Mr sif is5 i responsible for the rapid settlement which took place at 'the~ ~ end of 1974, the staff instructed the applicant to install six horizontal drains in the saprolite near the pumphouse.--29/ 1 The purpose of these drains was to forestall increases in groundwater level by keeping the level below the elevation of the drains. This in turn would prevent saturation of the I soil above the drains by removing excess water. But lowering 29/ As already noted, pp. 18-19, supra, one of the drains was installed in October of 1976, the remainder in-cthe summer ~~ of 1977. See VEPCO Figure 9 for the exact locations of the drains. groundwater also tends to promote settlement by removing the buoyancy effect of the water (App. Supp. Test., p. 6). Mr.. MacIver testified that, rather than serving to retard settle-ment, Drain 4 caused the settlement observed during the summer of 1977 by lowering the groundwater level underneath the pumphouse (Tr. 167-69). Staff witness Lyman W. Heller agreed that Drain 4 was installed below groundwater level and probably caused the incremental settlement (Tr. 362). The staff, however, still believes that the installation of the drains may limit settlement in the future by con-trolling groundwater (Staff Test., pp. 30-31). i Although the effectiveness of this dewatering attempt has been much debated by the applicant and staff, it is -.== -... -., n n.-- relevant in this prodeedirrr. ...:.'6nly'tcr the-^ extent- 'it adds' to our I comprehension of the mechanics of settlement. The applicant has made certain assertions that might be construed as imply-ing that it is somehow entitled to an increase in allowable settlement because staff-ordered dewatering may have con-tributed to the problem. We categorically reject any such suggestion. Our concern here_is with the safe operation of the plant and not with the allocation of blame for past settlement. 4 , B. Pipe Stresses and Pump Tilting Induced by Settlement Although our focus to this point has been on the settlement of the pumphouse itself, that settlement is a matter of con-cern only insofar as it, in turn, has'affected the integrity of the service water system pipes which run between the pumphouse and the f acility components which are cooled by that system.--39' For this reason, much of the evidentiary hearing addressed the problem of determining the level of settlement which would place unacceptable stresses on the SWS piping -- which, being affixed to the pumphouse, necessarily is also experiencing settlement. The staff and applicant used different methods to determine stress limits. Although those parties came to similar conclusions, we will analyze ~:cacE2netho=d-individua11y.- H ~ ., i u.. nu. a...na, o.. m _m.t m m.. - i.- Z== 1. Before turning to the pipe stress analyses, some r mention should be made of the configuration of the service water pipe system. The service water pipes penetrate the north wall of the pumphouse (in which they are embedded) and continue'for several feet beyond that wall before turning down through a 30-degree elbow at the top of the dike to enter 3V some questions were also raised relating, inter alia, to the effect of tilting of the pumphouse on operation ~~ of the pumps housed within that building. They are considered in more detail at pp. 35-38, infra. w- .. the ground. The pipes extend downward for about 65 feet,at which point 47-degree elbows (at the bottom of the dike) ) angle them to the northwest. Proceeding approximately 63 feet in this direction,the pipes encounter 7-degree elbows which bring them into a 260-foot level run toward the main pisnt. App. Test., p. 57; see also VEPCO Figure 26. The first pipe stress analysis performed by the applicant in 1975 indicated that the level of pumphouse settlement then being experienced was inducing excessive stresses in the buried portion of the pipes (App. Test., pp. 12-15). The applicant's located at: computations indicated three areas of high stress, (1) the entrance to the pumphouse wall, (2) the elbow at the top of the dike and (3) the elbow at the bottom of the dike y ;;, p;;; r;"S3 v7 2'.,3)f, In trder itoWer.ieve tfie"E'o". " ' u (FSAR, Amendment 4 9,"' 1 unacceptabis ~ stressing, tho' a'pp'11 dant" cut the' pipes"at' t'he ' "' ' * ' ' e elbow at the top of the dike and installed flexible expansion 31 / joints (App. Test., p. 15). In addition to eliminating ~~ the stresses at the point of installation, this relieved the stresses at the other two points by allowing relatively free motion of the pipes on both sides of the expansion joint (FSAR, Amendment 49, p. S3.72-3). 31/ There was some confusion as to the exact time of installa-tion of these expansion joints, which was finally identi- ~~ fied at the hearing as being between August and October of 1976 (Tr. 89). i 26 - The four expansion joints (one for each of the two ' supply and two return lines) are located in a concrete covered enclosure See VEPCO Figures 6, 8. Each just north of the pumphouse. joint consists of three corrugated bellows with connecting pipes. See VEPCO Figure 15. The bellows allow the joints to accommodate dif ferent directions of pipe motion, including com-pression, extension, lateral offset and angular offset (Tr. 93). Both because they accommodate displacements on each of their ends (thereby minimizing the stresses which would otherwise be transmitted across the joint), and because-they were installed at a point where settlement of the pump-house exerted maximum stress on the pipes, the joints are the as settlement is con-limiting components.of the SHS' insofar._$pA 229 & M O % - n.r miuu - U nfli the joints theoretically would be the'first pa'rt ra a u cerned (i.e., of the system to experience settlement-induced f ailure) (App. Test., p. 24; Tr. 175-76). The joints have not eliminated all i 32/ of.the settlement-induced stresses in the pipes,~~ although 4 the applicant's calculations indicate that "the worst-stressed limit of area in this piping is not anywhere near the [ stress] 32/ Applicant's witnesses indicated that portions of the ~ 175-76). pipes are still settling and bending (Tr. ~~ i i l the piping itself, but that in fact the most limiting component is the expansion joint itself and not the pipe" (Tr. 175). 2. We turn now to the stress analyses performed by the staff and the applicant. The purpose of the staff evaluation was to obtain assurance that the stresses in the-SWS piping do not exceed the allowable values set by Section III of the ASME Boiler and Pressure Vessel Code and that expansion joint movements do not exceed their design valu'es- (Staff Test., pp. 35-36). The staff used its analyses in formulating the revised technical specifications which were issued on June 28, 1979.. See pp. 49-50, infra. x 337 l With regard to the _capabjil,1.t.ies pf the,gxgapsion,,;iointsZ,.,- ~,, _ n e :..: 1. ._ u.. e.w- -. . c. ,,.m. _ .. = ma_w v.a the staff relied upon information' from the manuf acturer which indicated that the joints are designed to accommodate 0.25 feet of lateral movement (i.e., vertical or horizontal) of one end with respect to the other (Staff Test., p. 37)..Such dif-f erential movement is measured by comparing (1) markers on the exposed ends of the pipes that are buried in the dike fill north of the expansion joint enclosure and (2) markers - In its prefiled testimony, the staff referred to expansion 33/ joints as " flexible couplings." ~~ Y ~ . on t e northeast and northwest corners of the pumphouse' 34/ (ibid. ; see also VEPCO Figure 6). The markers on the ~- exposed ends of the pipe were not in place before July 1977. To determine the settlement across the joints prior to that time, the staff assumed that the exposed ends of the pipes settled the same amount as the top of the dike. The staff further assumed that the joints were installed in December 1975. Staff Test., pp. 36-37. Between December 1975 and July 1977, the top of the dike settled 0.079 feet and the northwest and northeast corners of the pumphouse settled 0.069 and 0.046 feet re-35 / spectively. Thus, the differential movement (or settlement)~~ across the joints for this period was between 0.010 feet (the
- J:L_w
- a =; aum : - w u v.
- rii = = = = = :.
a= u= =...un absolute of 0.#79-0_.,0 8 9 )nand i,0.,0 3 3 f e et ; j 0. 07.9- 0. 0 4 6). _ yThe -,,-.. staff adopted a figure of 0.03' feet to represent the estimated 34/ The markers used to measure pipe settlement on the south side of the joints are located at the corners of the pump- ~- Because the corners of the pumphouse are settling house. at different rates, and the pipes exit the pumphouse be-tween the northeast and northwest corners (albeit closer to the northeast corner -- see VEPCO Figure 6)', the staff's estimate of pipe settlement on the south side of the joints is based on an interpolation of settlement at the two, corners. 39 The only movements with which we are concerned at this juncture stem from settlement. ~ ~' 4 . 36 / differential settlement before July 1977,~~ and established an allowable limit of further differential settlement after July,1977 of 0.22 feet (0.25 feet design limit of the joints minus 0.03 feet experienced differential settlement yields 0.22 feet allowable future dif ferential). Ibid. The staff also evaluated the stresses in.the buried pipes. Using a simplified analytical model, it concluded that the pipes could withstand 1.00 feet of absolute settle-ment (id., pp. 38-39). An existing settlement of 0.77 feet as of August 3, 1978, was derived from various monitoring records (id., p. 4 0). This estimate, which assumed that the pipes became embedded in the dike in August 1972 and settled there-af ter at the same. rate.as. the dike, was;the. basis _ for the 'r. staf f 's imposidon: 5f-an-allevable limitiofa0.22* Teet tof '. crz:.';. 37/ ~~ absolute settlement after August 3, 1978 (id., p. 41). In reality, the pipes were not embedded in the dike until 3F This figure is a conservative one. As mentioned pre-viously, the. expansion joints were actually installed ~~ sometime after August.1976. Had the staff used settle-ment figures based on August 1976 instead of December 1975, it would have arrived at a differential settlement figure across the joints of approximately 0.017 feet., see Staff Test., pp. 36-37, Table A. 37/ The stress limits would seem to have allowed an addi-tional settlement of 0.23 rather than 0.22 feet. Pre-I sumably, the staff chose the lower figure to ensure a degree of conservatism. e . August of 1973. The staff believes that, in light of its misapprehension, the allowable settlement could be increased 38/ to 0.37 feet without exceeding code limits for pipe -~ Nonetheless, the staff did not increase the stresses. allowable limit in the technical specifications to 0.37 feet; it merely pointed out in its testimony that the 0.22 feet limit contained an added degree of conservatism. Ibid. For their part, the applicant's stress analyses utilized a computer code called NUPIPE (Tr. 100-01; App. Test., p. 55). That computer code is used to model the portion of the SWS piping that extends from the pumph6use to a point 63 feet into the 260-foot level run towards the plant. This portion, approximately c200_:fecti ini length',7 in'cludes all' the piping - that is af f ected4y"pdmphoude fand (dlke rsettleinpnt"(?NUPlPE?-f, Q-. ~~ ~ ~^ takes into account pipe flexibility, forc'es resulEing from pipe motions resisted by soil friction, and (in the applicant's recent analyses) the presence of the expansion joints. It also utilizes a relative settlement distribution profile 38/ The dike settled at least 0.15 feet between August 1972 and August 1973 (pd.,p. 41). ~~ 1 . constructed by the applicant to account for settlement of 39/ the buried pipes. App. Test., pp. 55-59.-- NUPIPE was employed to determine the necessity of the installation of the expansion joints. It was also used to determine the technical specification settlement limits for piping connected to many Class I structures. The initial technical specification for pumphouse. settle-ment, however, was based on the expected amount of future settlement and not on an analysis of pipe stress limits. 40/ App. Test., pp. 13-16.~- l .s --39/ Additionalidetails -on -thb1NUpIPE -rode' and ll,ts appiN l ~ cation in this case can be found in documents relied on and referred to by the staff in its safety evalua-tion of the applicant's request to revise the settle-ment technical specifications. This evaluation was attached to a letter from staff counsel to this Board dated January 9,1979. Other documents per-tinent to the applicant's analyses include: the proposed technical specification revision (contained in a letter from Mr. Stallings (VEPCO) to Mr. Case (NRC) dated June 13, 1978) and the applicant's response to the staff's request for more information on its stress analysis (contained in a letter from Mr. Brown (VEPCO) to Mr. Denton (NRC) dated August 2, 1978). 40/ The technical specifications are discussed in greater ~~ detail at pp. 4 6-53, infra. ] 0 A later NUPIPE stress analysis was performed in 1978 after the average settlement of the pumphouse had exceeded 75% of its allowable limit. The results of this analysis were reported to the staff on May 31, 1978. According to the analysis, the expansion joints could safely accommodate an average pumphouse settlement of 0.33 feet from December 1975 (as opposed to the technical specification limit of 0.15 feet from that date). Id., pp. 18-19; see also Tr.96-103. The computations also indicated that pumphouse settlement of 0.33 feet since December 1975 would not result in pipe stresses i exceeding the ASME code allowable values. See applicant's "Special Report - Settlement of Service Water Pump House North 41/ Anna Power Station-Unit No. 1," dated May 31, 1978, p. 10. 41/ For the purposes of this analysis,, the exposed-ends.-of. r the pipes were assumed to se'ttle at the same rate as the ~~ ~ pumphouse. As mentioned earlier (see fn.28, supra), the exposed ends, in fact, have experienced a greater settle-ment during the past two years. Peak stresses in the pipes, however, are not strongly se.nsitive to the actual. amount of settlement (strcsses in the pipes are due to a variety of sources 1. App. Test., pp. 59-60. To assure that such stresses will remain below code allowable values, the staf f has placed a limit on the amount of pipe settle-ment that will be allowed. That limit (0.22 feet from August 1978) is based on the staff's own independent stress analysis and is much greater than the value of, roughly, 0.04 feet (over the same period) that had been recorded just before the hearing. See p. 29, supra; see also VEPCO Figures 25A, 25B. The applicant's analysis also failed to take into account the effect of more rapid pipe settlement on displacement across the joints. The joints have been designed to take up to three inches of lateral displacement (Staff Test., p. 9; Tr. 320-21, 330; see also p. 27, supra). A witness for the applicant told us that, at an average pumphouse settlement of 0.33 feet (measured from December 1975), the displacement across the joints would only be about 0.5 inches (Tr. 330-31). Al-though the displacement across the joints may be larger (FOOTNOTE CONTINUED ON NEXT PAGE) . The applicant,'therefore, requested that the technical specification be revised to increase the allowable settle-ment limit for the pumphouse to 0.33 feet from December 1975. 3. Because the expansion joints are the limiting com-ponents of the SWS (see p.26, supra), the applicant presented detailed testimony concerning the operational capabilities of the joints. We were told that, under settlement-induced compression to 100 percent of the proposed technical speci-fication limit of 0.33 feet, an expansion joint could undergo 39,000 cycles before failure might be expected to occur (App. 42/ Test., p. 26). Further, according to the applicant, the ~- bellows of that joint would not be compressed solid.(i.e., there would still be room lef t for further compression) until ..._ hen _the,,,,j_/ settlement gr. eater _than.0. 66 feet had occurred W bellows are compressed solid, the joints can still absorb more than 2,500 cycles (a number f ar greater than the 1,000 cycles ~~41/ (FOOTNOTE CONTINUED FROM PRSVIOUS PAGE) than anticipated by the applicant should the pipes continue to settle faster than the pumphouse, that displacement is restricted to an acceptable value by the differential settlement limit imposed by the staff. See pp. 27-29, supra. 42/ Cyclic loads, as well as settlement-induced loads,. af fect the integrity of the bellows. The nanufacturer of th.e ~~ joints defines a cycle as one expansion and contraction of the bellows assembly (Tr. 239). The major component of such cyclic loads at.. North Anna,is the thermal change induced from starting up or shutting down the plant (Tr. 176-77, 217-18, 239). The system is expected to experience 1,000 cycles during the lifetime of the. plant (App. Test., pp. 26, 29; Tr. 218). . e"cected during the plant lifetime). Id., pp. 28-29; Tr. 221-22. To induce a failure directly, (i.e., without regard to cycling), a settlement of about six times that observed between December 1975 and April 1979 (approximately 0.75 feet) would be required (App. Test., p. 30). Tests performed by the manufacturer indicated that any expansion joint failures resulting from settlement would most likely be in the form of pinhole leaks (id., pp. 29-30; 43/ Tr. 113-14, 177-81, 221-2277 Leaks of that character can eventually develop into complete breaks, but water would continue to flow through the pipes (albeit at a lesser rate) while the break propagated. This is in contrast to a guillotine type of failure, which is usually assumed for convenience in analyses ~ 44 / v ~= involving the'eIfecE'bI b EafFiiH 1uiu1systenis. .' :such a= f ailure entails a complit'e"se~verahce of TIsijiii"t'6gether:With" - ~ a displacement of the severed ends,thus allowing an unimpeded flow of liquid from both ends of the break. Guillotine failures are more likely to occur in high pressure systems such as the primary cooling system (which operates at a pressure of 2235 pounds per square inch, FSAR, p. 5.1-13) than-in about lower pressure systems such as the SWS (which operates at only The applicant. produced a witness who specifically addressed 43/ the subject of expansion joint testing and the nature of ~~ expansion joint failures (Tr. 234-49, 281-83). licant assumed guillotine failures in the analysis 44/ The ar it ma regarding the effect of postulated failures in ~~ the ShS. See App. Test., pp. 31-35. , 250 pounds per square inch, FSAR, p. 9.2.1-27). It is thus very unlikely that a settlement-induced failure in the service 45/ water system would result in a guillotine break.-- 4. In its testimony, the staff expressed some concern that pumphouse settlement might bring about a misalignment of the pumps from tilting of the floor of the structure (Staff Test., pp. 43-44).~-46/ As we noted earlier, the pumphouse has settled with a noticeable tilt towards the northwest. See fn. 20, supra. Being affixed to the floor of the pumphouse, and having their rotating shafts oriented vertically, the pumps shift out of vertical alignment as the floor tilts. According to the staf f, the applicant is required to shim the pumps as needed to keep them within the limits on vertical misalignment recommendedbythemanufacturerkSta'ffS'ttlementSER,p.6)T. ~ ~ e
- ; ;;;. ;,
- ..:, - " T ~~" :-.,~;- :"; '" ~;L' 3 ~_,~, _T, 2 ~
Measurements are taken of pump performance' parameters (e.g., differential pressure, flow rate and vibration amplitude); should those measurements exceed the prescribed tolerance limits, 45/ Factors other than pressure also militate against a guillotine break. In particular, the pipes are buried over much of -~ their length and expansion joint displacement will be minimized (in the event of a break) by tie rods. See fn. 60, infra. 46/ See also the staff's " Safety Evaluation of Virginia Electric and Power Company's (VEPCO's) Request to Revise Technical ~- Specifications of Section 3/4.7.12, ' Settlement of Class I Structures' of Operating License NPF-4-North Anna Power Station-Unit 1," pp. 6-7, attached to a letter from staff counsel to this Board dated January 9, 1979. This document is hereinafter referred to as the " Staff Settlement SER." . ~ 36 - corrective action is mandated (id.,p. 7). The timing of, and procedure for,the measurement taking, together with the values for the tolerance limits, are specified in Article IWP 3000 of Section XI of the ASME Code (ibid.). The applicant is required by Commission regulations to adhere to these code requirements. See 10 CFR 50.55a(g). In the course of this monitoring, one SWS pump for Unit 2 47/ and both of the screen wash pumps were found to be outside ~~ the manufacturer's allowable "out of alignment" limits (neither of the two SWS pumps for Unit 1 had likewise tilted beyond its allowable limit). The manufacturer and the applicant both be-lieve that the pumps can remain operational until they exceed 26-27,189f their alignment limits by a substantial margin (Tr. Nonetheless, shimming of-the-pumps is requirednif theyrtilt:
- v. 7.)
t m - : w. me ro a :.- r.- u 2 u : m, nc nore than 0.011 inches per foot, and means of resolving the 49/ problem are currently being investigated. At the time of ~~ the hearing, the two service water pumps for Unit 1 were re-ported to have tilted 0.008 and 0.005 inches per foot, still within the 0.011 inches per foot limit (Tr. 190). 47/ The screen wash pumps are used to cleanse the screen which. filters water drawn from the reservoir (FSAR, pp. 9.2.1-14,16) 48/ Although the limit is currently set at 0.29 inches of total allowable displacement (corresponding to 0.011 inches per ~~ foot for a 26-foot long pump), the manufacturer indicated that the pumps can still function with a total displacement of 0.5 inches (Staff Test., p. 43.) 49/ North Anna Station Manager W.R. Cartwright stated that the applicant had encountered some difficulties in shimming the ~~ pumps, but still considered shimming possible (Tr. 26-27). i 1 5. The staff identified three other potential problems associated with' settlement of the pumphouse: (1) reservoir leakage, (2) cracking of the pumphouse floor, and (3) the effect on service water spray piping connections (Staff Settlement SER, pp. 8-11).,Because none of them relates directly to a potential immediaue threat to the functioning of the SWS, they were not extensively discussed at the hearing. We take note of them here,"however, for the sake of completeness. The staff was concerned (presumably) that leakage from the reservoir would contribute to pumphouse settlement by weakening the underlying soil. The staff resolved this concern by relying on the fact that groundwater flow is monitored. Thus, any leak-age through the reservoir liner would be detected, because it would contribute to groundwater flow. See Technical Specification 50/ 3/4.7.13. A measuremen't ~of]threc~ times thc_3yerage. an.nual 7 groundwater Efow'rektilres "f'YtNef'evaliiatilon' and dipdr~ ting.
- 92.;
u ' " ' ~ 51/ Respecting pumphouse floor cracking, it is now required that the out-of-plane distortion of the pumphouse foundation (determined by measurements of settlement markers SM-7, 8, 9, and 10) not exceed 0.06 feet. Finally, the staff's concern that settlement of the pumphouse might result in excess stresses in the spray piping was addressed by limiting the differential settlement 50/ The applicant argued that the design of the reservoir, together with precautionary measures it had already taken, ~- would prevent leakage (App. Supp. Test., pp. 13-16). SL/ The record does not reveal the reasons for the staff's concerns in this regard. l . between SM-8 (the southeast corner of the pumphouse) and markers E-569 and H-584 (pipe support hangers) to 0.175 feet (measured from June 1975 when the hangers were installed). C. Relationship of Pumphouse Settlement to the Public Health and Saf ety The staff and applicant based their conclusions that settlement of the pumphouse would not adversely affect the public health and safety on the following premises: (1) a settlement-induced break in the service water system is high_y unlikely; (2) any break that might occur'would develop slowly enough to allow time for corrective action; and (3) the system contains sufficient backup equipment to accommodate any such break. A break is believed to be unlikely because the level of w.,m. 4.y.;- ...w
- 7...,.. g,.
g g-y;,q g. settlement,is not expected to exceed the limits established 52/ by the revised technical specifications.-- Moreover, so the theory goes, even if settlement should reach these limits, stresses in the pipes and expansion joints would not yet be great enough to cause a break. See pp. 33-34, supra.- At this point, new expansion joints could be installed to relieve then-53/ ~ existing stresses. 52/ While most settlement predictions were shown at the hearing to be unreliable, settlement generally diminishes ~~ with the passage of time (see p. 13, supra). SV Just as in 1976, " physical system modifications to regain the original flexibility of the expansion joints * *
- could be accomplished by cutting and rewelding the pipe section of the expansion joint or the adjacent service water pipe" (App. Test., p. 27).
See also Tr. 178. Nonetheless, both the applicant and staff produced i evidence dealing with the effects of a failure in the service l water system. Before discussing this evidence, some additional details of the service water pipe configuration should be provided. As previously noted (pp. 6, 26, supra), two' redundant expansion joints are connected to the supply pipe lines, and two to the return lines. See VEPCO Figure 12. The service water pipes drop seventy feet in elevation once they have left the expansion joints. See VEPCO Figure 22. Thereafter, the pipes are joined near the main plant by the auxiliary service water pipes. See FSAR, Amendment 24, Figures S9.18-2 through S9.18-9. The auxiliary system thus uses none of the piping threatened by settlement of the pumphouse. 1. With these details in mind, we now consider settle-L _ a ::: = =..~ n -e L. =: ~ ~ = ~ ~ ment-induced poitulated'faildr65.[HEthels6rvice' water'd st'sm'.'" N - The applicant analyzed a failure of an expansion joint in a a return header, a failure of such a joint in a supply header, and a simultaneous failure of all four joints. Each of these failures was assumed to occur after the plant had been placed in a shutdown condition. In all of these postulated cases, 54/ A header is a section of piping that provides a common ~~ conduit (path) for flow of fluids originating from different sources or leading to different points of use. For our purposes, a header should be understood to in-clude that part of the service water piping between the pumps and the expansion joints. 55/ This assumption was based on the belief that the expansion joints would not fail until stresses in them exceeded (or at least approached) the design limits. The technical specifications require the plant to be shut down if the settlement limits are exceeded; even at the revised settle-ment limit, the expansion joints would have reached only ' 40% of their elastic limits. Tr. 115, 180; see also pp. 33-34,, suora. plant operators would have ample time to realign the system to located at the lake or to switch (in use the auxiliary pumps the case of a single failure) to the redundant header at 56/ the pumphouse.~~ In the event of an expansion joint. failure in a return header, there would be no immediate effect on the cooling capacitp of the SWS since the break would occur " downstream" of the point at which heat is removed (App. Test., p. 31). The reservoir has the capacity to supply service water for forty hours under these conditions (unless makeup water is supplied to the reservoir, in which case the system could remain in operation indefinitely). A switch can be made to the auxiliary 57/ service water system in less than 15 minutes,~~ well within the forty-hour time period before the reservoir would be t.,..,. : :.. t ~ C.7.~O. If " " h f ~' a .. Lw;;4c n, ;;;.3 3. .w.7,: depleted. Ig, pp. 32 : : Tr. ;117.-'.. Should an expansion joint in a supply header fa$1, the consequences of the resultant loss of service water coolant 58/ would be encountered relatively quickly.~~ There would then 56/ The staff reviewed and concurred in this aspect of the applicant's testimony (Staff Test., pp. 9-11). ~~ 57/ Switching to the auxiliary system requires the use of motor-controlled valves to realign the system piping and startup ~~ of the auxiliary service water pumps at the circulating water intake on Lake Anna. The switching is done from the control room and takes no more than 15 minutes. App. Test., p. 33; Tr. 24. 58/ Although service water would be lost immediately, Mr. Cartwrig:, estimated that the plant (while shut down) would not experiencj any adverse effects following a total loss of service water for at least thirty minutes * (Tr.184-86). i _ 41 _ i be two alternatives: the redundant supply header could be put into use or the failed header could be capped and 59 / the auxiliary system used. App. Test., p. 32; see also ~- VEPCO Figure 18. Flow could be reestablished in fifteen minutes using either alternative. Given a-failure of all four expansion joints, a switch would have to be made to the Lake Anna-based auxiliary service water system. As mentioned previously, it would take no more than fifteen minutes to switch to that system, which, once operating, is capable of performing all the functions per-formed by the reservoir-based service water system (see p.7, supra). 59/ The record does not cleariy..indis.at.e.the actual. procedure..- the applicWn_t' wouldifolihwf6f,'the'.Eimef neces'saryitio ~. cap"...TJi ~- a header. Nor does it. indicate how a switch would;be made. -f, s x to the redundant header. ' Examination of various piping diagrams reveals, however, that many options exist for making various valve alignments in the SWS. See FSAR, Figures 9.2.1.2-1 through 9.2.1.2-3. The simplest pro-cedure would probably be to turn off the pump on the failed header and to make certain that both pumps are operating at full capacity on the redundant header and that both valves are open. Two pumps are lined up to each header with only one pump normally operating (see VEPCO Figure 13; FSAR, Amendment 25, p. 59.17-1). In any event, there is testimony that a header could be either repaired or capped within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> (App. Test.,
- p. 32).
In addition, the auxiliary system can be used before the failed portion of the header is capped (see p.42, infra; see also VEPCO Figures 18-21). 9 -,s . s Although it might be at least 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> before a failed header could be capped (see fn. 59, supra), the auxiliary system can operate effectively before such capping is accom-plished. This is so because the expansion joints are located seventy feet above Lake Anna, while the' components cooled by the service water system are located just slightly above the level of the lake. A small portion of the auxiliary system's flow might rise seventy feet and be lost through the break,. but the great majority of the flow would still be directed to the plant. App. Test., pp. 33-35. The applicant's prepared testimony dealt only with failure of an expansion joint after the plant was placed in a shutdown condition. At the hearing, we asked if an ,,s. <..z--.., ... ~ analysis had been made of a sudden guillotine break in an z e :c_uc. x ir.+ c g e m.4..e.p. r. :.r. ::... a,a s .s.,,A,, c Mr. expansion joint while the plant was a full power. Cartwright did not believe this had been done. Tr. 284-85. e But he had earlier testified that, if the service water pumps failed, the auxiliary pumps could then provide sufficient service water to shut the plant down in con-formity with the technical specifications (Tr. 30-31). i Another witness for the applicant pointed out that the loss of one of the four service water pipe lines while the j plant is operating at full power is one of the contingencies taken into account in designing the system (Tr. 285). A staff witness later testified that the staff had reviewed, and found acceptable, the consequences of such a loss (Tr. 377, 453-54). If the service water system is designed so that the plant can safely withstand a failure equivalent to a sudden loss of one of the four pipe lines, then the failure of one of the expansion joints would likewise not produce untoward consequences. Counsel for intervenor Arnold questioned the applicant's witnesses regarding the consequence of 'a failure of. all lour.. -_ _. _ expansion joints during~ operation. While "such.a ; failure vas.v.. r...r considered by him to be incredible, Mr. Cartwright indi-cated that,even if it occurred, a switch to the auxiliary pumps would still provide a means for safely shutting down the plant. Tr. 116-17. 2. Another subject that was probed extensively con-cerned the ability of the operators-to detect a break in. the expansion joints so that appropriate corrective actions 9 . could be taken in timely fashion. A leak in the service -) water system would have no significant effect (i.e., adequate cooling would still be provided) until the leak rate approached 3,000 gallons per minute (gpm) (Tr. 226; App. Test., pp. 30-31). A leak rate of this magnitude would be detectable by the monitoring (in the control room) of changes in service water temperature and flow ( App. Test., p. 31; Tr. 183). If the leak continued long enough, it could also be detected by a decrease in the level of the reservoir, which is likewise monitored in the control room (Tr.181-82). The control room instrumentation readouts of temperature, n flow, and pump pressure would probably not indicate leaks at Rates considerably rates below 3,000 gpm (Tr. 226-30).. e n._ _. _. m __m less than that' amo'untN6 Eld;Meyer) ibe'IVisible '-on the :-.- ~ ~
- n ground outside the pumphouse (Tr. 229, 249).
Mr. Cartwright noted that a plant procedure requires inspection of the pump-house twice a shift (every 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />) and, to reach the pump-house, the operator must pass near the small enclosure housing the expansion joints (Tr. 248-49). He stated that a leak of about 1,000 gpm would cause the enclosure to overflow and t n n w-m ,-y.--- r~ m 4 would be visually detectable, even during a heavy rain 60/ (Tr. 249) ~~
- 3. Because operator action is required to mitigate the consequences of a failure in the service water _ system, plant procedures prescribe the steps to be taken in the event of such a failure.
The operators at North Anna are all trained in those procedures. Tr. 186-87, 285-86. D. Technical Specifications Technical specifications are made a part of the operating i license for every nuclear facility. Commission regulations divide the items to be covered by technical specifications j ~w a : 2..
- .w.
.,,a. x.m es.-m.u ni,.:. 2 L,
- q' = ;,I into five cat'sgories.
10 CFR 50.36(c). Cate~g'ory~ 1"~r~e fh' T r + l .%... q.3, g/ to, inter alia, safety limits
- when a safety limit for a l
nuclear reactor is exceeded,- the reactor must be shut down. j Category 2 refers to limiting conditions for operation. These are defined as "the lowest functional capability or performance 1 60/ Even a guillotine-type failure might result in a leak f rate of less than 3,000 gpm because the expansion joints are attached to tie rods that would prevent their complete separation (Tr. 298). Pinhole leaks necessarily would be accompanied by smaller leak rates. 6F Safety limits for nuclear reactors are those which are found necessary to protect the integrity of certain of the physical barriers which guard against the uncontrolled release of radioactivity. 10 CFR 50.36 (c) (1) (1) (A). ) 4 . levels of equipment required for safe operation of the facility." 10 CFR 50. 36 (c) (2). When such conditions are not met, the licensee must either shut down the reactor or take certain remedial actions prescribed in the technical specification. The settlement limits imposed on the North Anna facility fall within this category. Categories 3, 4, and 5 refer, respectively, to surveil-lance requirements, design features, and administrative con-trols. Section 50.36 is silent on the subject of licensee failure to meet the conditions imposed upon items falling in these categories. But the violation of any condition imposed by a technical specification is also a violation of a license issued pursuant to Section 103 of the Atomic Energy Act cf 1 1954 (4 2 U.S.C._2133)._ Depending on the seriousness-ok the _ =. = :. =:. -=., . a :vv, m. -,:- n c:. - - violation, it'thus could subject the licensee to:the range of sanctions provided in Sections 186 and 234 of the Act (4 2 U. S.C. 2236, 2282). With this brief background, we turn now to the technical specifications regarding settlement at North Anna. 1. In October 1977,'the applicant transmitted to the staff its proposed technical specifications for " Settlement 62 / of Class I Structures" (Technical Specification 3/4.7.12). 62/ Technical specifications are joint products of the staff and applicant; the applicant initially furnishes proposed specifications which are then re iwed by the staff. The commission is not bound by the scope of the proposed speci-fications, but may include any additional conditions it deems appropriate. See 10 CFR 50.36. l l 47 - They (1) set forth allowable settlement values for certain safety-related piping and structures; (2) specified monitoring requirements; (3) called for certain action; by the applicant if settlement exceeded 75% of the allowable limits; and (4) required that the plant be shut down if ' settlement reached 100% of the allowable limits. The proposed technical specifications were adopted by the staff and issued on November 25, 1977. They placed an allowable limit on the average settlement of the pumphouse of 0.15 feet as measured from December 1975. See pp. 8-9, supra.~~6y It must be noted that the' original pumphouse settlement limit was based on a prediction of the future settlement that
- c. r.,
ra:. e v. g stir c c.y n = = : e.. r..i -. w :. s un= cJ.1 that s tructure was.egppdtM,fo,gxperience;over. its. 4 0-year._,,...,., y n. _m ,. 7 rf.r life,rather than upon an analysi.s of expected pipe or expan-64 / sion joint stress limits (App. Test., pp. 15-16).~- It was j; p/ The technical specifications also established'a differ-j ential settlement limit of 0.25 feet between settlement point number 7 (near the northwest corner of the pump-house) and the points on the north side of the expansion joints (settlement points 15, 16, 17, and 18). --64/ The applicant's prediction of future settlement upon which the limit was based was submitted to the staff in December 1975 as part of Appendix E to the FSAR for North Anna Units 1 and 2. It was admitted into evidence in this proceeding as Applicant's Exhibit AV-1. See Tr. 17-19. e . known at that time, however, that the pumphouse could settle 0.25 feet without inducing limiting stresses in these com-65 / ponents (FSAR Amendment 4 9, pp. S3. 72-1 through S3. 72-5).-- On May 31,,1978, the applicant submitted a report to the staff providing its evaluation of " field conditions and the consequences of additional settlement", as required by Technical Specification 3/4.7.12 when settlement reached 66 / 75% of the technical specification limit. -- The results of that evaluation indicated that pipe stresses would remain within their operating design limits even if further settle-ment occurred. The applicant's analysis further indicated that settlement of up to 0.33 feet (from December 1975) would still not cause the pipes or expansion joints to exceed per- .-,s
- m. r _
missible design imits QAccordirifl'y F66 %. -. e((13p197 {f_rjhe '1.2 7],.f.. applicant submitted a proposed change to thd technical speci- --65/ Although the limit of 0.15 feet was based on a prediction of future settlement, the applicant (at the time the expan-sion joints were installed) assumed a settlement of 0.25 feet for use in its NUPIPE evaluation of the stress capa-bilities of the pipes and expansio.. joints. The analysis indicated that both the pipes and tne joints could safely tolerate a settlement of 0.25 feet. FSAR Amendment 49, pp. 53.72-3 and S3.72-4. --66/ Measurements in March of 1978 had ir#.icated an average' pumphouse settlement greater than O. 125 feet (which is 75% of the allowable limit of 0.15. _et). A " licensee event report" regarding these measurements was submitted to NRC by the licensee in April, followed by the detailed special report in May referred to above. See pp. I-4 and I-5 of NRC's Inspection and Enforcement Report No. 50-338/ 78-44 attached to the staff's testimony. . 67/ fications incorporating this higher value. 2. After requesting and obtaining additional infor-mation on this subject, the staff provided its safety evalua-68/ tion of the request on January 9, 1979. It e sentially agreed with the applicant's basic conciusions; however, it altered the proposal in some respects. The staff's version, 69 7 which eventually went into effect subject to our final review 7-did not impose an absolute limit on pumphouse settlement. 70/ Instead, the staff established four distinct limits: 67/ Letter from Mr. Stallings (VEPCO) to Mr. Case (NRC), dated June 13, 1978, transmitting the applicant's re-quest for " Amendment to Operating License North Anna Power Station Unit No. 1 Proposed Technical Specifi-cation Change No. 12." 68/ Letter fzom staffacounsel:ltos_this soard :(andhalL tru u-v_. parties), dated ~. January. 9., 197 9., transmitting. the. staff ' p.,, p., ~~ 4 "Saf ety Evaluat. ion of 'Viryinia7 ElectYic~ and Power"-Companp's - ~ " ~ (VEPCO's) Request to Revise Technical Specifications of Section 3/4.7.12 ' Settlement of Class I Structures' of Operating License NPF-4-North Anna Power Station-Unit 1." 69 / On June 21, 1979 -- af ter the evidentiary hearing had been concluded -- we issued an unpublished memorandum notifying the Director of the Office of Nuclear Reactor Regulation that, subject to our further review, we would not foreclose him from increasing the allowable settlement limits if he considered the increase to be warranted. Shortly thereafter,the staff put those revisions into. effect. See letter from Mr. Parr (NRC) to Mr. Proffitt (VEPCO), dated June 28, 1979, transmitting " Issuance of Amendment No. 12 to Facility Operating License NPF-4-North Anna Power Station, Unit No. 1." 70/ For locations of the various settlement points identified ~~ in the following paragraphs, see VEPCO Figures 2 and 6. 1 I O J ' (a) For the expansion joints -- a r vised value for the differential settlement limit between either SM-7 or SM-10 and any of the four points SM-15, 16, 17, or 18: 0.22 feet from July 1977; (b) For the service water piping -- the total settlement of any of the four points SM-15, 16, 17, or 18 on the north side of the expansion joints: 0.22 feet from August 3, 1978; (c) For the pumphouse floor -- a new out-of-plane distortion limit between any of the four points SM-7, 8, 9, and 10 in the pumphouse: 0.06 feet; ..., s. . s.. u,. .. m u.- =.m
- u.o _ ; un.. - - ;
~ Fc r the, spray' piping - l. :he'w Sifferential 1imit a' (d) , ;== :- u ==a
- = =.: = = e= =. =. : :..- 1:=q: -
between SM-8 and either H-569 or H-584 on the pipe hangers in the reservoir: 0.17 feet from May 1976. 3. The applicant has now accepted the staff's revised settlement limits. See "Vepco's Memorandum of Proposed' Findings,"
- p. 9.
On the other hand, intervenor Arnold opposes them. S'ee "Intervenor Arnold's Memorandum of Proposed Findings Regar'ing d 4 . Service Water Pumphouse Settlement," pp. 2-3. Her position I can be summed up as follows: The initial technical specifi-cations were designed to protect the public health and safety. This goal was to be accomplished through the establishment of settlement limits. Settlement has now (almost) reached the allowable limits. It follows, she maintains, that either the plant should be shut down or added safety precautions should be taken to minimize the impact of settlement.
- Instead, the settlement limits are being raised, unaccompanied by any additional safety measures.
As a consequence, a greater risk to the public health and safety is being created. At first glance, the logic of this position appears unassailable. It fails, however, on its initial assumption. The original sett;1emeD.L l.im. z.s.._f.or.w..No..r,th,mA,nna wore.,no..t...d. e.t.e..r,,.,.., p,- it 3..=, =. ~ n.- mined by a careful analysis to ascertain-the. level of.. settle _. q, ment which might threaten the public health and safety. They instead, based upon predictions of future settlement.
- were, This is not to say that safety considera-See pp. 31, 47, supra.
tions were not a factor in the establishment of the original limit. At that time, the staff analyzed the applicant's pro-posed limit, determined that it would not threaten the con-tinued integrity of the pipes, and therefore approved it. e -ms i 1 52 - On the basis of the pipe stress and expansion joint -l 1 i analyses more recently performed (see pp. 27-34, supra), we - i are satisfied that the revised limits challenged by Mrs. { Arnold similarly pose no safety problem. There is thus no reason to overturn'the staff's settlement allowances. Nonetheless, we feel constrained to register our strong disapproval of the manner in which the applicant and staff went about the development of the original pumphouse settle-ment limit. As we have just pointed out, 0.15 feet was established for that limit on the basis of the applicant's i prediction that the further settlement of the pumphouse would not exceed that amount over the facility's lifetime. To be sure, the applicant's analysis, endorsed by the staff, reflected that this limit was suffihiehtTP ~c~ohnseWa~tive' (eYse/p'res'u~niab'lyi23 3_7 -s_.._ it would not h' ave been= adopted'notwithstan' ding'the predi~ti~oh)~.~ c But the analysis showed something more: that, in fact, 0.15 feet was an unnecessarily conservative estimate. More parti-cularly, it appeared that a further settlement of 0.25 feet would involve no danger of the pipes exceeding their design capabilities. See fn. 65, supra. Had the applicant's prediction proved accurate, it would have made no' difference that that prediction -- rather than an ' appraisal of what was a reasonable outer limit of permissible 53 - l 1 further settlement -- had been used in fixing the technical specification value of 0.15 feet. But the vice of that 1 ~ approach becomes readily apparent given what turned out to be the significant inaccuracy of the forecast. With the further settlement approaching the 0.15 limit within a matter of a few years, the applicant found itself required to seek a revision of that limit. Although we have found the revision devised by the staff is adequate to prevent a threat to the public health and safety, the fact remains-that, by setting an unduly conservative limit in t'e first instance, the staff h opened itself to the understandable (if unjustified here) i charge that technical specification restrictions will be altered anytime the licensee is unable to satisfy them. - 3. % e, = .u _,. -i E u cr_r : =.=.
- ..n.
c. ~ 4. Apart. from the;matyer._of'.-allowable sett1~ement. limits,. ..c4-i n +. - the evidence addressed'both4 (1) thenfrequencynpf f settlement. "rd4 monitoring and (2) how rapidly the monitoring results must 71/ ~~ be reported back to the applicant. According to the original technical specifications for North Anna, the settlement of all Category I structures was to be monitored every six months. As part of the revised 71/ Settlement monitoring is presently being donc by Moore, Hardee & Carrouth Associates. They, in turn, submit the ~~~ results to the applicant. Tr. 80-81, 122-25, i a specifications, the staff increased the monitoring frequency for settlement markers SM-7, 8, 9, 10, 15, 16, 17, 18, H-569, and H-584 to once every thirty.one days. See Technical Speci-fication 4.7.12.1. The staff believes.that this monitoring frequency, which is to be observed for the first five years of plant operation, will provide adequate warning if settle-72 / ment limits are approached (Staff Test., pp. 42-43). At the end of five years, the applicant is to make an engineering study in order to determine whether there is a need to continue monitoring on a monthly basis (ibid.). Although (at the time of the evidentiary ncaring) the applicant was monitoring settlement monthly, Mr. Cartwright stated this was being done because pumphouse settlement had reached ninety -percentcof its al o'dabiielIihik aiid'shd ~ahplirt'Idf I ~ x : :_ m, ;. .n -9 c 2-cant wished to make certain that lisit was not exc'eedbd (Tr. "" 204-05). Mr. Cartwright also expressed the opinion that: From a good common sense engineering point of view, a frequency of once per month is excessive, once we can establish that the rate of settlement is sufficiently slow so that there would not be an opportunity for a significant amount of further settlement to occur between readings. Upon our ex-planation or indication of cessation of jg/ Groundwater levels and drain flow rates are also monitored monthly. Another reason the staff is interested in monthly settlement monitoring is that it could provide the oppor-tunity to establish a correlation (if one in fact exists) Tr. 339. between groundwater conditions and settlement. l . the recent further settlement of the four service water lines, an adequate monitoring program would see the reduction of that frequence perhaps to a quarterly basis.and eventually back to the original semi-annual basis, and this would be adequate from the standpoint of verifying compliance with the technical specifications. Tr. 206. Thus, while Mr. Cartwright believes monitoring once a month might be excessive, he conceded that the applicant does not understand the recent rapid settlement of the four service water lines.--73/In any event, the applicant must abide by the present monitoring requirements unless changes in those require-ments are approved. Although, as they currently stand, the technical speci-fications require.that.sett1.epeng surveys,be,.taKen y nthlyd. u w_. 3 they do not stipulateJ.how,sooni the%p}icant jusstcl.acqu, ire;m m,._ the survey results. Nor do they establish a time period with-in which the applicant must plot and interpret the results. These matters are covered instead in the applicant's written internal procedures, which provide the operators of the facility with day-to-day guidance on normal operations, maintenance and --73/ Theso lines have recently settled somewhat faster than the pumphouse itself. See fn. 28, supra. . surveillance. Specifically, these procedures require that (1) monitoring results be supplied by the surveyor to the 413-14) applicant within seven days of the survey (Tr. ~122-25, and (2) the entire process be completed (i.e., the results be computed and interpt %d by the applicant) within thirty-one days of the actual survey (Tr. 412-13). Intervenor Arnold requested that the seven-day require-ment be made a part of the technical specifications. See "Intervenor Arnold's Memorandum of Proposed Findings Regarding Service Water Pumphouse Settlement," p. 17. Both the applicant 74/ and staff opposed this request. In support of her request, Mrs. Arnold called our atten-tion to the applicant's past record insofar as reporting was . ;.,y.,. u. m,,,y.,o c.,y.,.... .ra.. 1-z. m.m :. - e~. concerned. In'particular, she" alleged: '(1)' Stone and~ Webster' ..3_ (and perhaps the applicant) should have been aware that pumphouse settlement might have exceeded 75% of its allowable limit even before the original technical specification went into effect (Arnold Proposed Findings, pp. 7-8)25l (2) in 1977, the 74/ See "VEPCO's Reply to the Proposed Findings of Intervenor Arnold and the Commonwealth of Virginia," pp. 14-15; "NRC Staff Reply to Intervenor Arnold and Commonwealth of Virginia Memoranda of Proposed Findings Regarding Service Water _ Pumphous.e Settlement," p. 7. 75/ Although Stone and Webster was no longer performing official surveys for the applicant in 1977 (this function was then in the hands of Moore, Hardee & Carrouth), it had taken a survey in 1977 which indicated the 75% mark had been exceeded. These results were not reported to appropriate personnel until more than five months after the survey was taken. Tr. 159-62, 165. m . applicant was aware that settlement was occurring more rapidly than expected and yet allowed the monitoring fre-76/ quency to be lessened (pl., pp. 8-9) ; (3) the applicant had been slow in the past in receiving results from its sur-veyors (id., p. 10); and (4) the applicant imprudently failed to place survey markers on the pipes north of the expansion joints until two years after the installation of the joints (id., p. 11). This " lack of diligence," it was said, justifies the establishment of technical specifications to replace the reliance on applicant's internal procedures to assure timely reporting of survey results (id., p. 12). The applicant conceded at the hearing that, prior to the institution of its seven-day reporting procedure, its track record in comp 1"itiYiig silrYefphYOi!6fbeNEYo"oddhne'~.* Y ** # h .u.r,~,_,_...- -.. m er ,-+n.s- -m s r ur z r : sometimes took'-Es<1ong as fo~ur month,s Tor monitoring results ~ to be transmitted from the surveyor to the applicant (Tr. 123-24, 128-30). We were told, however, that those results have been reported back in as little as two or three days since the 77/ seven-day procedure was adopted in early 1979 (Tr. 12 5) T-While the applicant intends to see to it that its procedures are followed, it perceives no reason to incorporate them --76/ Although monitoring at this time was required only every six months, Moore, Hardee & Carrouth was monitoring "on demand" and had actually been taking surveys more fre-quently than twice a year. The frequency of these surveys decreased after the summer of 1977. Tr. 15 6-7 0 *. 7jV The record does not indicate when the thirty-cne day pro-cedure was established. N, 58 - x int, a technical specification. See "VEPCO Memorandum of t Proposed Findings," pp. 24-27. We encountered two difficulties in evaluating the inter-venor's position. To begin with, she mentioned only the seven-day reporting requirement. Although the record is sparse on the point, we infer from the thirty-one day procedure that the applicant must make some computations after it receives results from the surveyor. If this is indeed the case, it would seem i that this final compilation deadline should be a greater object' of concern than the seven-day reporting deadline. We have thus treated Hrn. Arnold's request as directed to both procedures. i Beyond that, we were confronted with the apparent inability of.any witnesstat the. bear {pgtto. delineate the practical dis-3 tinetions between. violations:o,f, intern A reporting procedures, and of technical specifications. See Tr. 414-16. As we see it, for present purposes little (if any) distinction exists. Surveillance requirements constitute Category 3 of the See.10 CFR items to be included in technical specifications. 50.36(c). Thus, if the internal reporting procedures were adopted i as formal technical specifications, they woul'd fall into this But even remaining as internal procedures they will category. have virtually the same impact. This is so because Technical e 4 .,m.._-., Specification 6.8.1 for North Anna dictates that: Written procedures shall be established, im-plemented and maintained covering the activi-ties referenced below: c. Surveillance and test activities of safety related equipment. It therefore follows that the failure to " maintain" (i.e., observe) internal surveillance-related procedures is, of itself, a violation of technical specification requirements, in this case, of a Category 5 requirement. As it turns out, as a practical matter the difference tweeen Category 3 and Category 5 technical specifications is not substantial for our purposes. A violation of a Category 5 technical spegification subjects the-licensee to the same sua _c r s = "* z 3 = _LE.;wa unc.m mm, 3,. ~ range of poteritial.;pensities -as does :a Wio1'ation%f a Ca'tecjory 1 3 specification. And in neither instance is a licensee re-quired to shut down its reactor automatically upon realizing that it has failed to observe the technical specification in question; only violations of Category 1 and 2 specifications require this result. See pp. 45-46, supra. Nor is there 78/ Category 5 of the technical specification items includes " provisions relating to organization and management, pro-cedures, recordkeeping, review and audit, and reporting necessary to assure operation of the f acility in a safe manner." 10 CFR 50.36 (c) (5). 4 , reason to elevate the procedures in question to the status of Category 1 or 2. For, in our opinion a failure to observe these procedures (whether or not they are directly expressed as technical specifications) does not perforce warrant reactor shutdown. The failure to complete a survey procedure within thirty-one days (or to receive survey results within seven days), unlike the exceeding of a settlement limit (which is a Category 2 technical specificatien) does not indicate per se a potential safety problem. Such situations are routinely 79 / (and properly so) handled by the staff on a case-by-case basisT-For our purposes, there would appear to be only one possibly significant distinction between internal procedures and technical specifications. According to Technical Speci-fication 6.8.2, procedures established under Technical Speci-T ~ ' M!.3I 'r' .,;u[il 6.UN_b'!.N jI'E W d:r *'"?' 2E N " fication 6.8.1 may be changsa~bp~EhE facility's station manager after the change has been reviewed by the Station Nuclear Safety and Opezating Committee (a committee composed of engineering and operating personnel at North Anna -- see Technical Speci-fication 6.5.1). This is consistent with 10 CFR 50.59 (a) (1), which states that a licensee may make changes in procedures described in the safety analysis report (which include sur-veillance procedures of safety-related equipment) without i --79/ We need not, and do not, consider here the question of the Commission's authority to impose sanctions against a licensee in the event of willful or repeated violations of technical specifications or other regulatory require-ments. / i Commission approval unless the proposed changes involve " change [s] in the technical specifications incorporated in the license or an unreviewed safety question." The licensee must, however, notify the Commission of any such changes. See 10 CFR 50.59(b). Technical specifications, on the other hand, may not be altered without prior Commis-sion approval. See 10 CFR 50.59, 50.90. But that consideration does not aid intervenor'.s position here. We have held that technical specifications "are to be reserved for those matters as to which the im-position of rigid conditions or limitations upon reactor operation is deemed necessary to obviate the possibility of an abnormal situation or event giving rise to an imme-diate threat to.t.he public health and. safety."-. Portland o w.. k..., i,.,, ,..,._,.; w. _., - General Electric ~ ' Company' ~(Trojan Nuc1' ear -Plant)i= ALAB-531',: , ' ~,, _ 2 9 NRC 263, 273 (19/9). Applying that standard, we there refused to convert certain commitments which the applicant had made into technical specifications. In this instance, no different result is warranted. To repeat, unless and until they are amended, the applicant is obliged to comply with its established internal procedures to the same extent' as if those procedures were incorporated in the technical ~ specifications. Needless to say, the staff should be duty bound to monitor that compliance and we have been given no cause to believe it will not do so. In that regard, there have been three reports issued by the Commission's Office of Inspection and Enforcement within the last two years ' dealing with exten-sive inspections covering settlement of the pumphouse and service water pipes. See Staff Test., Appendix A. Given this concrete indication that the staff inspectors are alert to the settlement phenomenon at North Anna, we fully expect that they will fulfill their responsibilities in this area 80/ in the future. Further, should the applicant later decide to amend the internal surveillance reporting procedures here involved, that fact will come to the staff's attention early enough to permit an evaluation of the justification for the amendment before it might have safety implicatio:n :-- -. - ~..tr-:.-~ ~. m i.y -, ------.._; 7 m m n ~ ~ ~ ~ s. To ensure r.; - t c.1 e y n u u>.
- . = :a y :..
w.c. w.i.v. that this is so, we are ordering the applicant to notify ' staff of any changes to these procedures, made pursuant to Technical Specification 6. 8.2, within 10 days of the Station Manager's final approval of such changes. 5. Finally, two related points warrant our attention. In their discussion of the technical specifications at the hearing, all the parties assumed that the motions of the - buried pipes and expansion joints would continue to be de-termined from calculations based on settlement surveys. The 80/ See also Tr. 272-74. Board asked witnesses for both the applicant (Tr. 231-32, 286-88) and the staff (Tr. 444-49) whether it would be possible, and if so, whether it would not be more desirable to determine the motion of the expansion joints through direct 81/ measurement of the~ joints themselves. Although witnesses for the applicant (Tr. 232) and the staff (Tr.' 445) seemed to indicate that direct measurement would be possible, they maintained that reliance on surveys would be adequate. Unfortunately, both parties misconstrued the thrust of our questions. We were attempting to ascertain whether :(on the assumption of its feasibility) direct measurement might be preferable in that it would offer greater accuracy while eliminating the wait for survey results. Because the expan-sion joints aro-the11iriitin.g3qmpoJ;cnts _o;fAhegeg;14Ae water-~ g system (see^ pr26i supra) ;-timely and -accurate. monitoring.of, the joints would seem to be highly desirable; direct measure-ments could conceivably provide more precise information as to the amounts the joints have moved in each of the directions 82/ in which they are capable of moving.-- The effect of cyclic events upon the joints could perhaps be documented.
- Finally, 81/
Although the record is silent on this point, it would seem that there is no direct way to measure the buried pipes. 82/ The joints are capable of' accommodating many different directions of motion. See p. 26, supra. ~~ if a problem in the joints developed, in many cases direct
- { l measurements would provide immediate notification; there would be no time lag awaiting a surveyor's report.
While for these reasons we believe that direct monitoring of the expansion joints has advantages over surveying, the evidence of record establishes that surveying provides an ade-quate means of assuring that these joints remain within their design capabilities. This being the case, we wouid not be justified in ordering direct monitoring, regardless of its feasibility. Nonetheless, if a more accurate monitoring method 83/ could be employed,--'we strongly urge the staff and applicant to consider its adoption. In that same vein, we believe the applicant would be well-advised to adopt.otherameasures.sto; enhance early-detec ~~ ~' ~ tion of any expansion-joint Jenks. -. whether rsettlem&nt-induced-or otherwise -- that might eventuate. To be sure, we have found that the regime being employed is adequate to provide a reasonable assurance of safety against settlement-induced problems. But.in relying upon analytical predictions of the effect of settlement upon the expansion joints, it largely 83/ There is nothing in the record which indicated how (or even, for certair whether) direct monitoring could be accomplished. S _r. 232, 445. w-. a b. overlooks the practical expedient of checking for small leaks in these sensitive components of the system. In this connection, the testimony made it clear that any damage to those joints would likely show up first in the form of small cracks (" pinhole leaks") in the bellows, which cracks would eventually propagate circumferential1y. Yet it appears that no effort will be undertaken to monitor the joints for these small leaks. True enough, small leaks initially have no safety impact. For until any leaks become quite large (at which point their effects are detectable by control room instruments), the service water system's ability to deliver sufficient cooling water would not be impeded. But there are significant advantages -- both in terms of upgrading safety 3and 14 retaininggopeggting, flex,i,.,,,,,,,,, ~ 3 v.- ;. --.- - -. w.. z.. A _p5: - -.,.; acn 14y=.,,+ 75n, bility -- inherent in detecting a small leak before it becomes a major one. In theory, direct visual observation of the expansion joints would be the surest way to detect small leaks. But this may not be feasible because the joints have been con-structed with a protective covering (Tr;: 232). In any event, it would be unwise to require the applicant's employees to make repeat' e 5 i 84/ trips down the manhole into the expansion joint enclosure-~ when there is an easier way to accomplish the same objective, on the latter score it should take very little effort and involve negligible expense for the applicant to install a device in the enclosure to detect rising water. If such a device were made to activate a signal outside the enclosure, the presence of a leak would be readily apprehended by those who -- under present procedures -- are already required to 85/ pass directly by there on inspection tours every four hours. In light of our finding that the other measures being undertaken are adequate to provide a reasonable assurance of safety, we are not conditioning the plant's operating license to require that this. additional precaution:be observedd But the extra measure of-safety'IFFouWa'ch'iefe^,~$EMdG1Co'E,~ ~ " leads us to record here our expectation that the applicant will soon install and put such a device to use unless, within thirty days, it furnishes us a statement of reasons why it believes it ought not do so. 84/ Apart from the possible diversion of the employees from other and more important duties, there would be ~- an increased risk of inadvertent damage to the joints.- 85/ See p. 44, supra. d . l III 4 ULTIMATE FINDINGS On the basis of the disclosures of record summarized l above, we make the following findings: A. Settlement With respect to the settlement of the various components of the service water system at North Anna: ~ (1) There has been much greater pumphouse-related settlement than had been anticipated. See pp. 15-20, supra. The reasons for this are unclear. (2) The record does not permit any conclusions re-garding the ef ficacy of the staff-ordered de-s T, me texpansien cints are c =- Wi e M
- P. e ~^ '
watering system.!'~X1though the installation 'of ... n a. ; r.,.. r,.. - --:...-.. -e L ca s-
- r. T x=='u x=
a u::u Drain 4 apparentl.:.reme::y was responsib~1e for one incre-ment of settlement, it is unknown whether instal-lation of the drains will prevent future settle-ment through control of the groundwater level. Additionally, there is insufficient evidence to allow any conclusion as to whether heavy rains may have caused settlement in the past or might. e e 1 w cause settlement in the future. See pp. 21-23, supra. (3) Settlement has followed a step-like pattern (only in part correlated with construction activity), rather than a pattern of slow, steadily diminishing movement. See pp. 15-P 23, 17-19, 21, supra. (4) While its rate has generally slowed down in the past two years, settlement has not stopped. It is of some concern that more settlement was recorded between August 1978 and July 1979 than was recorded between August ' 19 7 7 - and ~~Augus t"19 7 8'," and"that' the- - - - - - - - - , ~ - - increhen't oT Tsttfe'm'ehY Wdo"r3e'd*Ti 3ahsTB'1991 N'P W ~ although small, was the greatest of any month since July 1977. See pp. 19-20, supra. Based upon the above findings, we cannot conclude that settlement will cease in the near future, nor is it certain that settlement will never reach a level requiring further modification of the service water system. 6 1 o 6 69 - B. Pipe Stresses and Pump Tilting With respect to settlement-induced pipe stresses and pump tilting: (1) Stresses in the service water pipes will not exceed ASME code allowable values and expan-sion joint movements will not exceed their design values at the settlement limits set forth in the revised technical specifications. See pp. 27-34, supra. (2) The expansion joints are the limiting component of the service water system insof ar as settle-ment is concerned. See pp. 26-27, supra. (3) The expansion joints are capable of absorbing, .- - th e n t -, 4%'m r.-, 2 n r m i.m +,g.,,, s, e much greater se'ttllidsnr thin that*allcwedf,by the revised technical specifications. See pp. 33-34, supra. (4) Expansion joint failures would likely begin as pinhole leaks rather than as instantaneous guil-lotine breaks. See pp. 34-35, supra. (5) The concerns regarding pumphouse tilting, reservoir leakage, cracking of the pumphouse floor and stresses in spray piping have been adequately resolved. See pp. 35-38, supra. l 4 4 1 4 .. s C. Public Health and Safety Although not a matter of certainty, it seems likely that settlement 'at North Anna will not reach the level necessary to threaten the integrity of the service water piping. In any event, even were that contingency to materialize, the public health and safety would not be endangered for the reasons that: (1) A leak of less than 3,000 gallons per minute (gpm) would have little effect on the service water system's functional performance. Leaks greater than 1,000 gpm should be detectable visually. In any event, leaks greater than 3,000 gpm would touch off. alarms in the control u" r z,; 7.1.: _ j 8 6/f"" " 7....,,,.,C < T ;. -' ~ ~ ~ ~ " - ~ ~ - room. See ppt 44-457 supra.- -1. (2) once a leak is detected, the plant operators have ample time to switch to the fully redundant auxiliary system or to the other operational header (if available). See pp. 39-43, supra. (3) Although no settlement-related f ailures are likely to occur before the plant is brought to a shutdown condition (see fn. 55, supra), the facility has 8v In this connection, see the discussion regarding a possible method of detecting smaller leaks at pp. 64-66, ' supra. 4 e 7 .'g' -D _ _ 71._ been designed toi withstand the. loss of one . service water line~during. operation. See pp. 42-43, supra. (4) Operators are trained in plant: procedures which prescribe actions to take in the event service water is lost. See p. 45, supra. D. Technical ~ Specifications The revised technical specifications. provide an adequate assurance that future settlement of the service water pump-house for North Anna Units l'and 2 will not endanger the-public health.and safety. This is because: .(1) The revised technical specification settlement limits ~should ensure that appropriate actions _ will-be taEsnTefbr'e Tset't'lement-induced; falluris ~;ini thegyfm, 4 ~ ^ service water"systhm 'occu'r'. 'A see'pp;"48-53';;' supra. ~ ~ - (2) The monitoring program required by the revised technical specifications, together with the ap-plicant's internal procedures regarding surveillance reporting requirements, will enable the applicant and staff to become aware of any settlement-related problem.well before the technical. specification j limits are reached. See pp. 53-62, supra. ) b 4 4 e ,e
- Based upon the foregoing, we conclude that there is reasonable assurance that the public health and safety will not be endangered by a settlement-induced failure in the service water system of the North Anna facility.
In light of the importance which attaches to the continuing monitoring of the settlement of that system, the applicant is to provide written notice to the Director of the Office of Nuclear Reactor Regulation within ten days of the Station Manager's final approval of any changes in its internal procedures (made pursuant to Technical Specification 6.8.2) which relate to the reporting of the results of such monitoring. On the basis of this opinion, our jurisdiction over the 87/ pumphouse settlement issue.is hereby terminated.-- For the reasons set forth at p. 3, supra,, turisdiction ogr the turbine missile issue is retained pending further order of the Board. It is so ORDERED. FOR THE APPEAL BOARD 1 W C. ean Bisnop Sec tary to the Appeal Board 9 ) 87/ This .ermination does not affect the expectation referred at p. 66, supra. e