ML19269F521
| ML19269F521 | |
| Person / Time | |
|---|---|
| Site: | Oconee, Arkansas Nuclear, Rancho Seco, Crane  |
| Issue date: | 05/19/1976 |
| From: | Grier B NRC OFFICE OF INSPECTION & ENFORCEMENT (IE) |
| To: | Goller K Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML19269F522 | List: |
| References | |
| IE-RID-76-13, NUDOCS 7912300013 | |
| Download: ML19269F521 (2) | |
Text
CI:T.'I-Mc % M ~
U.
- :r
- e. ->.
"'.c:
- ic.;
ScT 14 MS
.$ T-
,-3_..
I
...e..
a.i:1 R--
C'. t.: si l l e i-PO!O. AMLU:' F0i:: Chair'an H.,.'drie My -.v. '. ',,,
F F
Comiss'oner Gilinsky
.N. P " "- '
a co=issioner Kennady Cor:nissioner Bradford Comissioner Ahcarne Fr.0'* :
Harold P.. Denton, Director Office of fluclear Reactor ite;ulation
}
I THnU:
Lee V. Gossic'i, Executivc Director for Op: ration SUCJECT:
1 LAF.-~ PIPE LEAR AT TVO I, CLKILCOT;;,
I!:L.':D.ZO l.
C"
.1.5 Cl,USED OY FAULTY FACXICATIO: R.'OC55 I'. F~ACCE LARCE PIPE LEAX AT TV0 I, OLKILUOTO, FINLRiD TVOI is a 662 Oc BLT. supplied by the Stfedish Conpany Asea-Aten.
It first went critical in July, 1970, and operated at full powcr since January, 1979. /it 9:35 p.a. August 29, 1979, a larte leak occurred in c G-inch ciameter stainless steel pipe in the rea:: tor water cleanup system. Tnis syster, unlike cocparable systcos in U.S. C'.JP.'s, operates at the full primary systcr crcssure of 1232 PSI.
The f ailed section of ;:ipe was upstream of the filters, se tw.;atcr that leake::
out uas the same as in the reacter vessel.
The plEnt protection systems were actuated by high roc-ter.;erature, an:, by water level switches on the ficor. The protection systars actc: to cause reactor trip, closing the contcinnent isolation valves of the auxiliary syst5ts, and sritching the reactor building ventilation from nornal to filtered r:c::e.
The failed pipe section was isolated by closing manually operated isolation valves at 11:00 p.m., when it was determined that the radiaticn level in the area was not too high. Reactor pressure reduction was started at 12:45 a.m.,
the residual heat removal system was started at 8:00 c.c.
The plant was in cold shut down at 10:20 a.m.
The total amount of water leaked was esticated to be about 1300 gallens, with a total radioactivity of 10Ct:Ci. The dose rate in the room was 10 m rem /hr.
lL The iciled pipe section was at and downstrea: of a T connection where a sy-pass ar0ur.d a heat exchanper connected to the cipe dischcr'it.g free the heat ex-ci,&.. _ _ r.
As a result of an operatienti errcr, the ry c:s vaiv:.:as :.ositicr.cd
- srli. tb
-Tio..; hcif.mnt throck the heat :xc':ar r, ;h:,thcr half..:s Ne-c.:::. This c.ute:. urccold wat:r at 2C:: tc a in;:::5.' i o c cipe c:rryir.
c.:i'.....tcr :: 15 0L '.. Cit. 1 illustrc;;; t % syst: cc' i urcticn an;. i.;c:.:ie, L.
1 Cr;ck.
I
- ~- 'Oll crc
--i :.i0n Of t's. f;ilC2 tipe Lisclosc: :l-t tS: ~ i'. '"i--
i c. U.r-
- In i.
i c r 1
. i -- : l i.,,
.s i crcs ic
,... v i.
-o=
2166 054 791230o O'
)$
M,"Sl 'ka
accition, there.:as a smaller through-wall crack cricr.In circurfcrcr.ticM..
sfter s2cticnin/, it could be se:n that ther2 usrs r:r.y cr:cas en th. ir:i :
surface in c " crc:c-cracking" pattern, typiccl cf ther.1 faticue fcilur:s.
i: ore detailed r.ctallogrcohic and fractccrc: hic cnalysis confir. cd thct tr.
j cracks were ccusec by fatigue. Fig; 2 shows the craze cr:cking on thc insitie r
surface of the pipe, as delineated by dyc penecrant. Frcctographic evicence of fatigue on the fracture surface is shown in Fig. 3.
The parallel lines running horizontally in the photo are typical fatigue " striations," showing the pro-grcssion of the crack each stress cycle.
It was concluded that the failure was causec by the rasid ther'.al fluctuations created by the mixing of hot and cold water. The piping was r:placed, and no further probicas are anticipated with the by-pass valve pr:perly closec.
Sicnificance To U.S. Plants i
A very similar f ailure occurred in 1962 in the Vallecites E;7:. Cold feedwater entering reactor rceirculating water through c T :onnectica ccused similar cra:c cracking and a leak. SWRs in this country have had cracking on the inside rcdius of feedwater noz:les caused by leakage of cold feedwater around the thermal sleeve. Thc recently discovered cracking in the P'.;R feecuater piping near the steam generator noz:les is now believed te be primarily caused by cold feedwater nixing with hot water free the insico of the steam generatcr.
Staff Actions i
I 00R is preparing an information letter to DSS cescribing this occurconce, and recoc:nending that they be on the alert for similar configurations in new plants, f
A letter will be sent to licensees of all operating plants asking them to check g
their syste-s for locations where hot and cold water are nixed, either uncer nor=al conditions or could nix if valves are set incorrectly.
CRAC1'S CAUSED BY FAULTY FAEP.ICATION PROCESS IN FUf!CE i-i We have had several reports of cracking in staa : cenerator tube sheets and reactor vessel nozzle forgings in components manuf actured in France. Although specific technical details arc not yet available to us, it appears that the problem is caused by a carginal w21d cladding process used to clad the compon-ents with corrosion resistant weld depotit.
In the case of the stean generator F
tube sheets, Inconel 600 is weld deposited on the prinary sidc, as is cone on t.
plants in this country. Cracks in soce of the French components have been E
I found beneath the cladding, in the base cetcl fer-ing.
Sese of these crccks cr r:;,crte:' to be up te 1/2 inct uenp (12 c '..
In the case of the stainicss steci cla: renter vessel no::les, the cracht ar: osti; cb:ut 1/lc in
_ee:,
but 501. are u; to 1/a inen Ocep. All cracks :r: r:lctiv ly shcrt, caut 1/4
~.
l/2 inch.
Th
-;;,;cct Ccus Cf the CraChinf is still so L ha*. OcsCure.
lJccut IC f ars cgo r ;. - cli. crf.0'r.i"" U a3 alSO discoVCrs!., Er.7. 'C; E~tTi'";;; t: 0 0 rrir..itic Of n
' e : t i M...r i n. "O WId cl % in~ c C r ct i : ~-, c~..'sc N t2l IM iU~s Of
?00R OR 8BL a66 055
susceptible (high alloy) composition. The probler was first discoverc:' in Gemany, and it appeared that forgings made in Europe i.cre screwhat c:.re susceptible th:n those of United Statas canufact:.re. This prefer was re-solved by codifying the weld deposition procedure, and in sene cases by
['
modifying the chemical composition of the ferging.
The French habe stated that tte current cracking problem has a different Although it is still related to fabrication processes, and also cause.
appears to be related to forging btse metal com,,esition, the cracks appear
' @=.
a to be much deeper than those of 10 years ago. Although we have not yet been given enough technical detail, the French have said that the process sequence causing the cracking consistr. of a first weld deposit pass with the usual pre-heat c post-heat, but the succeeding pass or passes are acnc without pre-heat or post-heat. Further, they feel that the cracking is " cold cracking," and
[
related to hydrogen either in the forging or added in the weld cladding oper-p ation.
It could also be that not all the cracking found is due to the same F~-
cause.
Sicnificance To U.S. Plants At present there appears to be no reason for corcern that similar cracks are present in cocponents of U.S. plants, because the weld deposition processes have been carefully checked out as a restilt of the problem 10 years ago. The only U.S. plants with reacter vessels manuf actured in France (Creusot-1.oire) are Prairie Island Units 1 and 2.
According to Uestinghousr* these bere clad I
using Westinghouse approved procedures that required pre-heat and post-heat H
for all weld passes.
F Staff Actions y.-
The French have promised that full technical details will be made available F
to us early in f:ovember, when several Eepresentatives will be visiting !GC for f=
the Research Infomation Meetings. Appropriate mestbers c'.the staff will met EM with them to discuss the problem and the technical details.
{
n fiorthern States Power has been informed of the French problem, and that it is f
possible, although unlikely, that their reactor vessels could be affected.
If J
after full details are available it appears prudent to inspect the Prairie ff accessible nozzles.
If further information leads to the conclusion that other'.
@f Island nozzles, the staff will require inspection of at least the two most'-
?
plants could be affected, the staff will take th.atever action is appropriate.
E..
cc: OGC SECY
, /!
C4. :,..i,.;
u x-r
- e....!
__.a 7
Harold R. Conton, Director
~"
- Office of ;uclear r,cacter Regulation E
y 7v AD:EP: DOR )
Actina Acting DND0'O t:RR IGR ED0' ES: DOR
'":conan LCShao
-9Ekb ECase HDenton L30 sick
/lC]
10/
10 10/ I /10 10/ /10 10/ 10 10/ 10 10/
gi U
d l
Faulty fenetion:
263 C A1: kg/s c
c IgC C
210 c 233 g er
- a c'x 10 kg/s h
Planned functien:
n flow 253 C
cocier 20 kg/s l
J 2166 057
, i i
f fi f. [i. W W>
R
-] %
.S$[0
&,?
?'*}h.~4-$; ; &y?
~
, k)' %$ b, G
)v '
,(5.
~
,i g
.tt x.'
. l c+ ; % '.,..R. ' 8v.4 w ' T. '~@..
g 6.v.
s is 4
uq
\\ ; 4 'g.',j j%.d.
- lpr,
'9
<.. :. y
- f.4 epi,3 3.v$$@~T.
i 5
- V gjlj.irjy f
~t h.
/gh :[ % f'. '
u'y ~u'y,'$
,,y 'y 9.Q
ij.'
x
!,{1 m& _\\ :.
1
. T y.. i
.. ;. s. y
\\F 1, ".
.c.. ; *ll1.,t n}, :.is. [y..,%
- y):'n w
1, g, -e t
g.
~,
a.< k,,5 :..v,q r.,
./
or
- 3 n-y
- x p.r.:w4.. A
?\\
..1.Jv.
- K..I.Lf.. :' w'%cy%eq g&.' j y'
r.,-
',. ' t u-c s
. q..w. c.-
r
- t.,..,.
p e
': b, P..
.t. g/.
2
.. =
. g d,, y i.
. -. ' - ~
g h..
+
4 o
4
.-..m-Q..
., p..
y b.
s.
, (.
r.
_m
-e ia""-
.x>.:----~~-
b
- ^--
ro
~ ~
2166 058
?00RORBNJL
e
.e e
8 mm'.*
o-M Ti
~ ki
%e i
4
.L a
t e
{.
6M
. m,.
,h -
m u
rIT.'
3
\\j,.h J._.-
j L
~~
g f.
- 4*j
~
w
/
t-h.
s s
e g
t.
e r..
s.
1 4-,
/
s l
m....+i
,y f'. 2.
o 21660$9 POORORGN1
on L~-1 3 I
f FrauToe timnAD c=:s P;t=:.su FRAPKTOE hrrwIw L2ERC2.AD OtACKI3G MAS Ltit:1.1w IN lit cur CF 15 REACTOR N EZZLES.
t WIM.M IEFN:
7M
~
a.
r:=u tscm:
Dm FRAPRTOE ALSO 3EPGtTED L3EiE2.AD CRACKDE IN ST~r.M GEERATM luE SHEETS.
MIM.M EFDC 13 m f
mIMLMLysm:
Em
~
cAUSC OF CRACKIE:
RtIRCEEN EMNtITTUMBfr
- 2166 060
'00RORIGNAl.
SUBJECT 3@ CATAL
- "~
NO*
M A D E l:
m'
/
l
.r e
it xt Kr\\
eB o n.Rcb r
SeMPni l
A"N P e h P
K l
L A"
X C
0 RR2NA s 8 D 0. M A
i 1
CA.A0
^
X I
v s
i 1K
~
R (v C
f A-
\\
\\'\\
b D 2
~
E A
~
O. L i*
9 C
R EP MIR B
~
U T S
S
~
, )
/
X e
f
(
S 3D:
3
" CC2 cu C bW 3
C
~ @ o@~
j 13 A
D n
1
( C W c_ i O -
I2 ON5a I
}1, 1
l I
I FRffAT&E 11DRIAD GWXItG IICIDEhT KTim PLM FOR ESO.1 Rim 0F FRRD1
- SUREY C USOT LOIE #0 OTER EETm WSSEL WTOTS AS TO CU@lfG PROES (EED m ESTIffriCUSE EACTOR MS
- If6ECT QJD THEE ( 3 ) rn771FS Fim ms / 00S EACTm WESF1 S
- RFUN FPACRE EONICS #RYSIS T0 hSSESS ItfIEGRITY T ' PRAIRIE IRAf0 EETOR WESFIS
- IfEECT ( UT ) THIE ( 3 ) TUE SEETS AT TA'PA FDR llCERCIAD CPAXItG
- ONUIER #RYSIS T iM)ROEN DERIlTLHENT PEEL.
SUBJECT 3 M CATALOC
INSPECT (UT) il;?IE N0ZZLES FROM OPS / OQS PIACTOR YESSELS TASK COMPLETED P.EPAIRABLE INDICATIONS IN ALL THREE N0ZZLES PER FRAMATOME CRITERIA FOUR BOAT SAMPLES TO BE REMOVED FROM THE THREE N0ZZLES FIRST BOAT SAMPLE TO ARRIVE 11/18/79 FIRST B0 SA WLE EYALUATION 11/20/79 REQUESTED THAT ONE OUTLET N0ZZLE (OPS) AND ONE INLET N0ZZLE BE SENT TO WESTINGHOUSE 2166 633
?00RORGlW1
- +h**M e
=m+=--eee
-m.-
-em e
._me..
.m=
..m.,m m
I!!SEP.VICE INSFECTION DEVELOPFINT SCHEDULE FOR DEVELOPING ISI PROCEDURE 1.
FABRICATE CALIBRATION BLOCKS 12/7/79 2.
ASSESSMENT OF 45*,
60',
AND 70* SOUND BEAM USING CALIBRATION BLOCKS 12/14/79
- 3.
ASSESSPINT OF 45', 60',
AND 70* SOUND BEAM USING N0ZZLE FROM OPS / OQS 1/8/80 4.
MODIFICATION TO ISI TOOL 1/15/80 PROCEDURE CAN BE DEVELOPED WITH0 STEP 3 2166 064 h
/
. ~. _. -. _ _. _ _ _. _
9~i'A 'EC 3 n'iTri t.----in (. y. t;.,;]$
SLEA 19 - OTH5Tlm E61ME
/EP ALA' APR ARZ A52 01 CSL CTL W
IIP d.
=
H UK U.P LM' iM lui EU Q1 P!U KJ FEli PEE RA FR)
TF TF WIS WS SlFB TFR - PAKDry A'O Wil QX
- 6 CE QE
, CE '
QE aM RA FR KR LEM L&l IEE EP SLFPLIER - GICRD ERIDGE & IPat W
CF.
GL OL RR suRIB - !UDEM KHEAR SP' Ta TW E'
VIR' n'
m 2
300R ORl8 NAL
"'"~"""
Ex 1-am a adro:
witax 2166 065
(
2-aADwSum SUBJECT 3 M CATAt
au - snu.
EER IfM5TIGATim
~
SLRtIFR - FP#%TifE H
PEP f&
TS GS F9f Sella - BJ ta m M n FAR IIP DCP W
WT SPPLla - tHI -
e w
a
?bOR ORGINAL aiu ou SUBJECT 3_@ CATALO
DEFINE LABORATORY PROGRAM WORK UNDERWAY HYDROGEM DIFFUSION CALCULATIONS UNDERWAY RESULh5 FURTHER WORK TO BE DEFINED BASED OM 0F BOAT SAMPLE EVALUATIONS
~
/
P00R ORGINAL 2166 067 6
e 1
-- -