ML20114E702
Text
g. pl a
- v., i w-5+"'""*
MEMO ROUTE SLIP b)
coaca"* w Id, swrm uc-om... w?i.. a<n n ee.. e 4.., v. a.
r -.ic uo..
r e m-ne.
s :2-To (n.m..no unn, mmus m-sAt the June c'ecting we requested the ACRS to give.
_.,. } l mc C. K.
Beck, DR CPJ g -*
M*
M. M. Mann, DR
=
any advice they could on whether or not we should
) (
O R. Doan, DR o,rr hk
-[
L.
G.
Case, DRS release the two AEC stub tubes in the Ovster Creek
'I opJ t
b i rg 4
to (n.m. and s,mt; emnes Pruanits P. A. Morris, DRL vessel for repair. The Committee decided not to
't g
WPJ F. Schroeder, DRL f
O
- ive us advice in this area, however, they have Sci l
oAre UWu Y
OtDO furnished us the attached statement which represents (3
WH p
<t yo (N ged yMF)
INITIALS PfMARR$
g cc: /11. L. Price, DR the infornial viewpoints of various mernbers of the Mg R. Boyd, DRL 7
j R. Tedesco, DRL our Committee.
H. L. Price plans to have a discussion fp R. DeYoung, URL f
f f
L. Forse, DRL sometime Tuesday afternoon, June 11, 1968, atatime/
- -j h
FROM (N.ene and umt)
PtMAAR$
/
sN I
k Saul Levine, AsstFJDiv which he will establish. f'
~
J r
i for Reactor Tech.
'7[
' " >[
_/)
-/
J Div. of Reactor 1
_[(~__]
//
/SN
/.
I &
Licensing I
/
i v' V
/
\\ ik
-m e.it
- x..
7471 6/10/68 e
USE 01+sER SM FOR ADDtTK**L REMARKS c e s e _,,,es.,
spo, see,
.,n.,sp
_,, ?g 5,.J'
- hi M. *
U I'i'* *-Y
- IWOM a**
-l9, 1
s, *f
.. :?.
' n-f a
_%.i
'h5 e
[?. ~ :*
N
'u'" '.
y
. _..= / l e
. S-%
'g,*,
e e**e
+
t>
. - to
')
n PROPOSED STATEME!ir TO AEC REGULATORY STAFF The Committee believes that s,:tematic, non-random, concurrent failure of redundant elements should be considered in the design and review of reactor instrumentation systems.
Systematic failure of a protection synt em and, part icularly, common f ailure modes of control and protection systetrs are of concern. The Committee suggests that the staff consider sur '.ementation of the IEEE criteria for use in review of instrumentation, and propose criteria which take the possi-bility of systematic failure into account.
4 9
)>
'iec'd 0)fy Dir, of R88'
/f Dato-s
- ;U-~ --
- ~ ~ ~ ~
Time-5--
3eth.
~~~~
v
~
s.-w g
PROPOSTD STATDEKr TO Arc F,EGUlXf0P.Y STAFF The Committee believes that systematic, non-random, concurrent failure of redundant elements should be cont.idered in the desicu and review of r(actor instronentation systems.
Systematic failure of a protection r ynt e m and, particularly, conunon failure modes of cont rol and protection systems are of concern. The Cocenittee suggests that the staf f consider supplementation of the IEEE criteria for use in review of instrumentation, and propose criteria which take the possi-bility of systematic failure into account.
1
.tec'd Offy,ir, ei R2D D
Dato JC-- y'7~~~
i e
Ilme. - -
~~~~~~
3 eth.
v l
~
k 4
Oyster C. cek St ub-t ube Repiir The Applicant contends that no safety problem is involved in the repair.
could arise from con-iderations such as the following:
AEC interest 1.
Possibility of in-service deterioration of stub tubes, CRD housings, or field welds might be construed as causing a credible hazard.
For exampl e, sinultaneous ejection of a suf ficient number of tods to cause f ailure of the catcher might be considered credible -- thi s highly improbable coniit ton co;1d be postulated on the basis of undetected general deterioration followc3 by a severe pressure transient or carthquake.
be consideted desirable to minimize any rish tho' the 2.
It might catcher may tc " tested" by a single. control rod ejection.
3.
lailures in service that caure leakage may place on the-AEC the burden of a dif ficult decision concerning the safety of continued operatien.
4 It might be considered that quality assurance of the reactor vessel is important even where safety is not an issue.
The following discussion is based on the supposition that the AEC is interested in a critical review of the repair procedure.
Ficid-weld Contribution to Stub Tube Cracking The reactor supplier has concluded that the observed intergranular cracking of the stub tubes is caused by senuitization of the material, a corrosive agent, and stress _ caused by the manuf actur,)pp, process.
The underscored assut.ption is questionable.
The supplier cites calculations of as-manufactured stress by three independent organizations, but is silent on the much greater stress imposed by the field welds.
1.
Har the reactor supplier calculated the stresses caused by the field Does he recognize that the calculated tensile stress from the field ycJd?
weld, at the location of cracking, may be an order of magnitude greater than that from shop fabrication? Did he measure the relief of strain on cutting the field welds before he established his repair procedure?*
- So far as we know, the only measurements to itivestigate the ef f ects of the field welds were made at the request of AEC.
The Applicant has been requested to measure the relief-of strain when the remaining field welds is possible that the measurements will be reassuring. (continued) are cut -- it
'I 4 1-
_. _ _ ~ _. ____ _ ____
i 6
i j
l
.p.
4 The linitntiene cf elactic annlysis end the uncertaintice enused by highly locali cd defor:t tien are obvicun, of potentially extrce.cly hi:h field uc1d :trecccc7but doen this jut t!fy co:nplete diere:ard 2.
Inver tir tion-hivo failed to explain tho ceverity er tbc inter-trnnuler attne':.
- Mvc the uotn11ur-ical invceti ct6ts been alerted to the potni!ility thnt s treca cnd s train at the lecctlen of failure ray have been extraercinnrily hich?
3 The 4.nter;r nular erncha chew n vertien1 br:nch parallel to the incenel ucid interftee.
Mi:h tencilo stress in this rct on is indicated i
by eniculation.
The strene vill be inerensed by rctovel of part of this weld.
'lculd r.ot thir. Le a reacon to pay ecmo ct',ention to the stress at the bacc of the c tub tulo cauced by the field ucidt Sirnificanco of Fabriention and 'feldin~ Strernen, A eteci ve::cl uith no tuntenitic intertn is essentirlly tresn-free as delivtred.
Oeleule tions of operr.tienal Otre:oco nre therefore normally bn cd on :
- tre,:-free vertel.
!!oucver, the Code doec net prohibit use of austenitic : art: in t ferritie vec c1, and come level er residuri strena is therefore ire.licitl condened in nn a:-dolivered ve:sc1 ende of t sterinla hsving Alfferen' coefficiente 50 expnnnion.
1.
Does the cu??lict residuni e trc 's it? lie:
take tha pesition that Oode silence on no,concitization) are unimportant, re ardice:that etreeces produced by field eeidin; practice doca not require eny effort to dininich theatof na nitude, and thet ;ood 2.
icnore the re:ultin:If a chrinh ring were applied to a tube, uould the cupplier rin:, end (c' in the tulo auny from the rin;7(n) in the rin:, (b) in +he tube unde treece:
3.
If the cupplier doco not tahe the positien that ucidin; and 1
shrinh-rin; otre:se: are totally unimportant, what are his criterin for ceceptabic stro:: Icyct:
i.e. does hi:
cnd hou han he ap?1ied thou to his repair procedure, mininire "oldin;tr. field weldin; procedure reficct a con:idered atte: ?t to
- e
- end if so he'.rt 4.
I'n: the sup? lier conducted a inboratory develoynent te:t,' with the ac','ni conficuration and cateriale, mininire r:'ini cenirnction of the hourin; and stub tute?to develop a procedure that uill Use of :hndrol in Field '.!eldine i
The f altici field tc1dn have cauned radin1 shrinha e of the CRD housin;, nr.d for'.her chrinha;c durin: 'tcidin; is anticipated by the su ppli er.
He t?ecific that a candrcl is to be ured uhere the over-all l
l 1
., -. ~. _
.m,
v
-5
'l hero the chrinb :e vculd interfere uith inrcrtion L."
thornal cloevec.the r,trere c:ndition in tne CRD houc ld uill to ir.creaccd.
tendrel in uned, will to alleviated, but the hoop strene in the wc
hnt ic the juctificat!cn for truto force cenetraint durin:
"ny>cheuld a precedure that so greatly effects the rtrece 1.
l Weldin;?
diatrilution te op'iorn17 tuo j
I!ou doen the proecre differ in principle fron ucidin:
h Enre?
tnre crd-to-end end trcventin; chrin':n c by rigidly cln.npin-t e I
2.
'loald the nu:olier cencider thic Troceduro ecceptable?
4 t
Control-rod-drive !'ounin-c defor:ntion rhou (cinstic theory) i.
C lcairtions frc 1 nen::ur:rente of 'he houcin:en extremoir hi-h the field veld,
,l in n rc: ion that is aire hcat-nfrected.
i c*ticuinted the tennile eti c;c incide 1:nc the renc'.or carplict 1.
1 the houcin.'
l ar.u exanincd it for censitiention 1:ac he rectioned a C?D houcin 2.
and cacceptibility to inter:ronninr correcient 4
Did hic procedurce nderatc1" protect n in:t poccible overheatin:
J f
durin; Thuc ctrni:htcrin:, and hne he innpceted houcin:c for eny evid 3
overhentinp' 1
m It ic stated that the hou::in:c are hept belou red hent by vicuni l'ouever, the uninhibited tendency of the chop nn face 1
l observation, vi th n in r ;c cro~' of etici-htenin:
In torch hentin:, it ic poccible to heat to ucil above incipient color until the finto is i
heat.
redne:c (even to d 211 red hon.t) without cccine; turned eff.
1:cu ic the o;cration tonitored?
t Conclunion It ic probatic thtt the repnited stub tubec uill be in cod the role pinyed by netn11urcien11-iiovever, the Inc': of curiosity concernin:
tresses vill field strece in the fnilure is dis turbin:, becauce thece ceceWe reactc,r sup of strccc.
be icposed en the !CIL overIny.stroco is uninportant, reyrdlecc of capitudo, J
the position that ucidin:
This does not ceca to no effort to control cuch ctrcotes.
and is nn::in a
e 4
m c
n c4,,
9 i
4 g
j
. 4.
o 4
s c.
o.
i 4
1 be consistent with a good quality assurance program, or to reflect the consi cration that thorough in-service inspection vill be virtually J
impossibic and that the problem will apply in some degree to the entire i
GE product line, The possibility of troubic in the CRD housing should also not be l
ov e r l oo;'.ed.
4 4
4 i
6-3-68 i
d l
4 i
t i
a a
4 w., - ~ -
y y
-,.,y--m.,
,-w-,-,-
- ~ --
e--v.-
w -,4qw.
p-