ML20127N559
| ML20127N559 | |
| Person / Time | |
|---|---|
| Site: | Monticello  |
| Issue date: | 11/14/1969 |
| From: | Dromerick A US ATOMIC ENERGY COMMISSION (AEC) |
| To: | Boyd R US ATOMIC ENERGY COMMISSION (AEC) |
| References | |
| NUDOCS 9212010309 | |
| Download: ML20127N559 (12) | |
Text
a
~
4 DR NOV 141969 Original signed R. S. Dayd, Assistant Director for floactor Projects, IRL R. C. DeYoung DC TERU:
S. IAying, Assistant Director for Deactor Ted"lhE7p IRL FONTICELI4 NUCLEAR GENERATING FLANT (DOCERI NO. 50-263) - REACIOR YESSEI.
We bave corgleted our review of the reactor vessel for time Manticello Nuclear Generating Plant.
Our report is enclosed.
L
\\ l5/
A. W. Droserick, Chief Containment & Con @onent S chnology RT-825A Besach URL:C&CTB:LP Division of Desctor 1.tcensing
Enclosure:
Report on TSAR Review of Monticello Plant ccw/ encl D. R. Wiler, DRL D. Vassallo, DEL M. B. Fait; tile, DRL J. P. Knight, IRL bec:
S. Levine, DRL R. C. DeYoung, DRL A. W. Dromerick, DRL L. Forse, DRL DISTRIBlTTION:
Suppl. 4
('(4 DRL Reading C&CTB Reading AD/RT Reading DRL:C&CTB DRL:C&CTB,
DRL:D@[
DRL:AD RT otrat >
0
$URNAME>
11/M) h.
11/fh/fh..
11/J3/bh,=.
11/h/bh....._
Daft >
. _ __.___. Form AEC-518 (Rev. 9-53) AECM 0340 e.oveenstat P.ivti.e orms. me e-n.-ei, 9212010309 691114 PDR ADOCK 05000263 O
4 I
ACRS REPORT 110NTICELLO REACTch VES3EL l.0 IllTn0DUOTIOM 1.1 _0cnertd Serious problemt arise with respect to the transportation of large power plant components to remote cite locations for which access is limited to secondary roads and railroad tracks.
The Monticello site is such a location.
For this plant, it was_ decided to pre-fabricate the reactor veccel cubcomponents to the maximum degree compatible with the transportation limitations and to complete the vessel fabrication at the cite.
The ASME Code,Section III, does not requiro vessels to be fully fabricated in one place such as an established chop. However,_
because it was a major departure from conventionrd practice, our evaluation of the vocael has been conducted to a somewhat greater depth.
e
+
l.
- ~,.... _
)
3 t
1.4 Status of Reactor Vessel The reactor vest.cl fabrication, including the CRD housing field velds, has been completed by CN11.
The /GZ Code required hydrostatic test is completed ani th: Code St:vap has been affixed.
2.0 EllACTOII PRE 33URE VESSEL 2.1 Generni The Ibnticello react.or vessel is designed and fabricated in accordance with the 1965 edition of the /G2 S:ction III, Claus A, Code plus 1966 Addenda, and Code Cases 2332, 1335, 1336, and 1339 The vessel core support structures are accigned to accommodate jet pumps,-
and the internal support structure ic mtde of Inconel and stainleco-ateel vhieb has not become furnace censitized during the fabrication cycle. CRD (control rod drive) stub tubes are made of Inconel, and the distance between the shop veld and the field veld (to CRD housing) is approximately 4 inches.
It chould be recalled that this distance was a minimum of 0 5 inch for Oyster Creek and nine Mile Point stub tubes and thic, together with the stub tubes bein6 of stainless steel, resulted in high ctub tube stresses during operation. The stub tuben are attached.
to the bottom head in counterboreo with a partial penetration veld, a Section III approved method.
9
-' l
j
.k.
The reactor vencel is twie of lov alloy carbon steel type A-533 cl.1, crade-B.
The vescel interior 10 clad with '?ype 308 stainless steel applied by i
veld overlay technique.
The Monticello vessel han the came Ceometry no opecified for other mm i
type venoclo except for the recirculation inlet nozzles.
Those nozzlee i
i are located at a 9-inch higher elevation then for similar vocselo. Thic.
feature was necescary in order to accomodate the field velding of a girth scam without incurring concernable veld dictortion in the nozzle vicinity.
Thermocouples are attached to the vessel utarting at the clocure region and continuing down along the side of the vcosel to the support skirt region.
These thermocouples are connected to a panel in the nain control room where the temperatures will be observed and recorded during the initial run of
~
operational transients. The observed temperaturco vill be uced to verify nocumptions for the reactor vessel trancient stroca analysis.
2.2 Dasir;n Specification The design epocification is basically identical to cpecifications used l
~
for_ reactor vencels for plants such as Dresden 2 and 3 and Millstone.
Specific differences in the specifications are ascociated with special t
provisions related to items such an electroslag velding (not used for this vessel),andseparationofresponsibilities.
j t,
i 1
3 4
.,.a.._
.m, _ _.u_
.. ~.
l t
5 Tho FSAR specified 50,000 cycles of Rod Worth Tests for thermal loading.
{
This transient vna interpreted at a meeting with the applicant on October 5
2, 1969, to consist of: 50,000 cycles 'of 10 *F temperature changes, and j
400 cycles of 30 'F tegerature chan6co.
Decause a 10 'F change in water 1
temperature hno an insignificant effect on the vessel strecs icycl, this i
trancient is not analyzed. UE's atandtud equipment specification for future vesseln vill-lict only the 400 cycles of 30 'F.
/nendment No. 21 to the -
F0/Jt, dated 10/17/69, includes a corrected Table 4-2-1, REACTOR VESSEL ANALYSIS, with the Control Itod Vorth Testo reduced to 400 cycles.
The staff concurn that omission of the 50,000 10 'F cycles does not influence the veccel streso level and the cumulative fatigue usage factor in a signi--
ficant manner.
{
Alco included in the specification are inspection requirements over and above the Section III required.
Notably among these are 100% ultrasonic j
testing (UT) of plate material, Ur of cladding bond'on a grid pattern, Charpy V-notch toughness' testing with enough camples to obtain a full Charpy curve including upper and lover plateaus, and nil ductility transition.
temperature (ITMT) determination by Drop Wei ht testing.
6 Since CBLI did not initially have a ctress analysic Group of sufficient size and experience, this tack vac divided between General Electric (GE)'
i and CBLI.
GE vould perform the transient thermal analyses.and the:cyclici
.t l
'i
,_.,A._.....,...,,_-.,
- c.
. _..,.....,,,,,,,,,. _. _ m
/a.., _
,,--,,,..._.,,,.u.
._,..r,
...,.e
4
( i operation analyces for the components.
In tvidition, _ the applicant agreed to provide an independent review of the strean analycos.
2 3 _Strecs Analyg The results of the strecs analynco obov that the highest cyclic strucceo -
vill be experienced by nozzles for recirculation inlet, feedwater, CRD bydraulic return, and core spray and flooding.
All of thece nozzles.nre provided with *aermal cleeves to reduce the streco 1 cyclo to within code acceptable limits.
}!igh strecsco are also indicated for the vescel support chirt near the veocel bottom.
Stress levels for CRD stub tuben are lov enough to permit chakedown to elactic streca levels' after the initial operational cycles.
/dl calculated stress levela conform to the limito-of Section III.
Cumulative usage factors have been calculated to be 0 52 for the core spray.
nozzle, 0.4 for support chirt attachment, and.0 56 for the closure stud bolto. Other calculated usage factora are less than 0.1.
~
/cendment No. 21, dated 10/17/69, referenced GE Topical Report / SED-5703 on CRD Ienetration thermal and streso analyses. Vnile ve have not yet made a detailed review of this report, our preliminary review indicates that the design assumptiona agree with the CE reactor veccel specification.-
~.. -,.._
..-_..__.___._..,._._..:.a_-.-.
,_...u._n...._._..___,._._..u.._.
.,a-
1 i
i l
.l.
t
.._J The Teledyne IMterials peccarch Division, acting for the applicant, made
,/
an independent review of the strece analycio report and found it consistent
{
j i
with industry practice with respect to the analytical' methods used and the-ASME code interpretations enployed.
Baced on our review of the cummary ctress analycio provided by the applicant and the independent review by Teledyne, we have concluded that the' design of the reactor vencel vill provide adequate margina of cafety.-
2.4 rabrication Data 11e a
Fabrication of the reactor vencel proceeded without any unucual problems aricing.
During stainleos steel clndiing of base metale come arcan vere found to develop microfiscuring.
Examination of the problem diccioned a ferrite content of 1-2%, while practice has established a 9-10% range au being normal to prevent cracking.
Backtracking, it was found that a human error in calculating the ferrite percent in the qualification procedure had prevented corrective adjustnent of the flux composition. Remedial action van undertaken by grinding off the affected areno^to remove the curface crach indications, and in areas where the cladding thickneco was reduced to less than 5/16 inch additional stainicac steel was redeposited.
i Although not considered a problem the fabrication procedure for the-vcscel differed from conventional practice in two vays;
(
e 1
(-
4
o,,
8 1.
The flat base platu material vao Q & T (quenched and tempered) by tle stoe?. supplier rather thtut by the fabricator.
2.
The plates vere cold forned ani streso relieved by the fabricator 4
1r. stead of the conve!)tionta hot forming followed by a Q (4 T heat treat-nent. The cold forming van performed after the platos had received-nufficient prehent,100 'F min., to move the meterial out of the transition temperature rtu2co and into the ductile range. The totta forming strain vac appraximately 2-1/2f,.
Experimental vork hac botn performed to predict the degree of influence of cold forming or procencing on plate' materials,1/ however, t.oct attention has been given to uniform-strain nnd relatively little to-the non-uniform ctrain coniitionc which occur during bending operations.
ASE Section III considero cold forming effecto in paragraph N-515, except that it does not pince a cpecific limit on the r;nount of cold formin8, provided the tensile and impact properties of the parts are act reduced below the minimum specification values. ; Eocentially this means that the manufacturer must produce a procedure qualification test for forming operationc which are not followed by an nuotonitizing heat treatment.
CILI performed cold forming tecto on A302-B and A533-D,- class I, materiale i
Total strain selected for theco testa was M which ic in exceso of actual forming strain.
Results of thece testa show taat:
i l
y W. T. Lankford, "Effect of Cold Work on the Mechanical Properties of -
Preocure Veccel Steels." The Velding Journal, April 1956, pp.195-s to 200-S.
t 9
Cold deformation of 4% otretch and subsequent stress relier does-n.
I not significantly affect tensile properties of the plates at surface and 1/4 T levels.
Surface tensile strength was reduced
[
f approximately 2000 poi and 1/4 T tencile strength approxitr.ately i
1000 psi.
11ovcVer, the plate center tensile properties improved
\\
approximately 1500-2000 poi due to the cold vorhin6.-
b.
Impact properties were clightly improved at the surface levelo,.and they were conciderably enho.nced at the pinto center.
c.
Lit ited UW (drop veicht) test information was obtained on the A533-B material.
Imr temperatures varied from +10 'F to +50 'F for these tests.-
We have concluded that for the lionticello vessel cold forming of the-reactor vessel plates did not produce a concernable de6radation of-materials properties.
Drop veight test results to determine NUTT for -the vescel shell and dome.
material ranged from +10 'F to +50 'F, except for one piece at 0* F.
~
These tect results are from the 1/4 T (thicknesc) level at the tensile stretch side (outside) and they reprocent the more conservative location, 1/4 T versusL3/4 T,. for the samples.
Based on the results of the drop veight tests, which are better larT indicators, we intend to require that the vessel pressurization temperature:
be 110 'F instead of the 100 'F proposed by the' applicant.
I
4
'e 10 25 pimencional Variationc The fnbrication sequence did not result in any dimencional variancec which vould compel code consideration or introduce difficulties with fitting _
the ' internal structures into the vecr.el. 'llovever,ethe final-overall mencuremento resulted in a 15/16 inch longer vescel and one inch larger t
incide diameter than called for by the cpecification. The effect is an=
l incignificant increase in tha verici stena space and approximately lHi cu. ft. of additional vater crace, which is lesc than 1% incroaco of -
volume below nermal vater level.
It is noteworthy to obcerve that loeni veczel chell distortions at veld -
i i
Jointo are lens for this reactor vescel than for chop velded vencels.
This condition is attributabic to the veld grcove coometry as compared to the deep J-Groove Ceometry employed for cutot.atic chop velding.
The latter approach recults in local inward bu1Cinc at the veld location, a condition which can be of concern in voucel areas with high coolant flow velocities.-
j It in also noteworthy that the dimensional-location and verticality-variations for bottom hewl CRD penetrations and stub tubec are all-vell within the specified tolerance range and, in fact, within _ a_ band equal to 20% of that rance.
This dimencional-control van-achieved through the_une_of shop t
produced boring templates, which climinated the need for careful location-measuremento at the site once the templates vere properly aligned andl anchored.3 L'e have concluded that-the increased vessel dimencions are of insi nificant -
6 magnitude.
I e.
n-e-,,,*
,,br,,,,-w-,,w--c--ec.er-wm-m,.-
-,m-..--y e w
'+.v~
,w
~t
' vvw e
'**--*v-c---t-y
- -**"'"f f-
- r
"+P-'
- f*"
- t f
11 2.6 14aterial problemn -
l Except for the CRD ntub tubon, no concernable problems arose'about the t
materialo for the reactor voucel.
~
While unchining incido and outuide diametcrc of. the CRD vrought Inconel otub tubec, axially ali6ned inclucions vero-noticod. These inclusions were of a type not retuilly detectable by conventionni non-dectructive i
incpection methods, and vere found to exict on a limited number of stub tubes which vere then rejected.
Invectications and laboratory tents per-formed by CMI, Illinois Inctitute of Technology Reccarch IncT,itute end.
othern have dice)ored the inclusionc to be magnesium oxide. The Inconel-600 material iu a flux-cast naterial and magnesium oxide is a. constituent of the 1' lux; other potential cources of mccncolum oxide are magnesium (added as a mallcabilizer and deoxidizer) and the firebrich of the mold, I
which under adverse Jouring condition could increase the propensity toward-clac entrapment.. Although tests at CMI choved the cmall inclusionc.to be. innocuous, the une of any inclusion-bearing material was avoided. The' Monticello reactor vecsci consequently doco not contain any detectable inclusions.
27 Sensitization of Stainless Steel i
Aside from prudent material selectiono, the fabrication proceduro and sequence of operationa vere establiched to reduce to a minimum'the number of cases where'ctainless stccl component parts of the vessel vould become furnaco censitized.
As a result only the following' items =are in a fully sensitized condition:
c
'ye mm wa-m
+-+c'---tsur-9 aq'N w e Mg'--WgwM.gi,*-e+ex-ert
--F
,mism itM",
gr P
-s'N-
'ane w-y a t 71 eec err
+-
p1
'M wV
' re \\ h 17d
+r**r
=g-rv W-ver++
'pa er-e
%lu7--
s
j 1
1.
1.' eld depocited cladding overlay on vessel plates and forcing.
2.
Two recirculation outlet nozzle cafo endo. -
3.
Two jet pump Anstrumentation nozzlo safe ends.
4.
The CRD byarnalic syutua return nozzle safe end.
5 Two top hen'l inctrument nozzle. unfo ends.
6.
The top head vent safe end.
/dl of thece vessel arena not availnble for inspection from the outsido.-
3.o co!cLusion On the basis of our review, we have concluded that the Monticello field fabricated -
reactor vessel ic acceptable for cervice.
6 9
3 t
I
+
f I
j-
...