ML20198F964
| ML20198F964 | |
| Person / Time | |
|---|---|
| Site: | Washington Public Power Supply System |
| Issue date: | 09/03/1975 |
| From: | Strand N WASHINGTON PUBLIC POWER SUPPLY SYSTEM |
| To: | Boyd R Office of Nuclear Reactor Regulation |
| References | |
| CON-WNP-1100 GO1-75-181, NUDOCS 8605290159 | |
| Download: ML20198F964 (12) | |
Text
..
o .
o *-
s
?
l O.n7!m,da:D ri ,4 '
Attac1 cent I
' 1.',3
- c' Washington Public Power Supply System
( / A JOINT OPERATING AGENCY N.l tb e. c,. so. us sooo cro w.. . oros w.< m e m . o. w.. - :o ., m por. ison us.nu September 3, 1975 Docket Hos. 50-460 G01-75-181 50-513 Mr. Roger Boyd, Acting Director
. Division of Reactor Licensing Office of fluclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, D. C. 20555
Subject:
HPPSS NUCLEAR PROJECTS NOS. 1 & 4 -
REACTOR VESSEL SUPPORTS
Reference:
- 1) Letter, NO Strand, WPPSS, to A. Giambusso, NRC, same subject, dated July 18, 1975.
- 2) Letter, NO Strand, WPPSS, to Benard C. Rusche, NRC, " Supplemental Limi.ted Work Authorization,"
dated August 21, 1975.
,y -
)
Dear Mr. Boyd:
The WNP-1/4 ACRS letter mentioned a concern for the loads on the reactor vessel support structure for certain postulated loss of coolant accidents.
In Reference 1 we conmitted to submit a report which would address this concern. This report is attached.
The report ~connits to the addition of pressure retaining guard pipes to those portions of the primary system hot and cold legs that are within the reactor vessel subcompartment. In addition to reducing the uplift and asymmetric loading on the reactor vessel, the guard pipes also allow for the design of the reactor vessel subcompartment for lower differential pressures.
We request that the Staff review of the vessel subcompartment differential pressure portion of this report be performed as soon as possible. As noted in our request for a Supplemental LUA (i.e., Reference 2) this analysis could be considered as an unresolved safety issue if it is not reviewed and approved in time to facilitate the issuance of the Supplemental LWA, since the analysis supersedes that centained in the PSAR. Please note that the supplcmental LWA request included the pourir.g of the
, reactor vessel subcompartment walls. Prompt revicw and approval of this LUA with no analysis restrictionwould on theallow construction for theofissuance of the walls internal to supplemental' the .
Containment Also, the design of the subcompartment and procurcnent of taterials is
) proceeding (at our risk) for the lor:er differeitial pressures.
8605290159 750903 -
?, . t' Mr. Rogar beyd 52? erber 3, 1975
. 'Page 2 C01-75-181 To facilitate this review, we have made the report.as complete as possible and have addressed areas that have been a problem in the past. Also, we are committing to the design of the reactor vessel subcompartment for a differential pressure of 110 psi which is 120% over the calculated value for the most critical node rather than the usual 40%. It is hoped that this commitment will eliminate any difficulty should the Staff calculate a differential' pressure somewhat higher than indicated in the attached report.
We would welcome the opportunity to meet with the Staff to discuss any aspect of the report. Also, we will incorporate any of the attached material that the Staff desires into the PSAR by amendment. -
Very truly yours,
, W H. O. STRAND Assistant Director -
Generation & Technology NOS:AGH:km Attachment ,
cc: CR Bryant - BPA, w/att.
TH Cox - NRC, u/att.
JR Schmieder - UE&C, w/att.
EG Ward - B&',1, w/att.
9 9
4 I
l I
Attachnent 2 SUPPLEl4EllTAL TESTIMONY OF
!!RC STAFF O!!
REACTOR PRESSURE VESSEL SUPPORTS .
WASHIliGTON liUCLEAR PROJECT N05. 1 A!!D 4 DOCKET NOS. 50-460 AND 50-513 BY VINCEllT S. 1100 NAN e
k e
i c
L
o -
U111TE!) STATES OF N-!!:!:ICA 1:UCLF.M1 111.GULN! CRY C01:IISS10 '
~ El:FO!!E Till: ATO:!1.C SAFETY AtlD I.ICl;;;SI!!G E0A!!D In the Matter of ,
),
)
WASill!;GTO:: l'UlH.1C POWER SUPPLY SYSTUI )
)
(Washingtoh Public Power Supply System ) Docket No. 50-460, 50-513
)
Station, Units 1 and 4) )
TEST 1110:1Y OF VII;CE!!T S. !;00!!A!!
's Introduction On 11ay 7,1975, theI fluclcar Regulatory Commincion (t:RC) staff was inforced by the Virginia Electric and Power Coctpany (VEPCO) that the ef fect of one of the i loads postulated to occur in the unlikely event of a specific loss-of-coolant
(?
( acci, dent had been calculated unconservatively in the original design of the raccLor ves'scl support c.ystem for I orth Anna Units 1 & 2.
Safety analyses for nuclear plants are usually based on the ausumption that the reactor vesse) rer.:ains fixed in place or undergoes only very s:::all movenents
- j under the full spectrum of postulated design basis accidents. 'lhe reactor *
~ '
vessel supports are therefore designed to sustain, u: hin censervative stress limits, the loads calculated to result from postulated accidents, such as pipe breaks in the main coolant loops. In recognition of the fact that om or more of the vessel support loads calculated to result from a postulated LOCA was found to be greater than the loads used in the original design analyscs, the effect of the increased loads on vessel motion (and therefore on the integrity of attached piping, on the operability of control rods, and on naingenance of cb. .sle core geometry) required reevaluation.
e
.- M N p.,.. -
,. - ,,-m. , .,,,,.-n-. , --.-.r - -,
I A An noted in the- Prel .inary P.cport of the 1:dC st.af f s.ated Ju%. 11, 1975 -
c-lu tha NRC staf f ?s opinion that the questions related to the tre#tment of transient LOCA loads in the desi;;n of reactor vessel support systems that have arisen for North Anna Unit.s 1 and 2 tr.y apply to other PUP, facilities. We have, therefore, initiated a systematic review'of this matter ' td* determine hou the design basis loads ucre tchen into account on other PWR facilitics, and what, if any, ucacures nay be required for specific facilitics.
i The results of studies reported to date by the licensec for North Anna Units 1 and 2 indicate that, altliongh thic margins of safety for lateral loads niay be less than originally intended, the reactor vessel support system will retain essential structural integrity. This means that the
'ressel-remains fully supported, while undergoing some translation, and
, aat the ultinnte consequences of this postulated accident are no more severe than accommoJated by the original design. Ue are in the process of an independent evaluation of the calculated transient loads on the i
-- support systems, and the effect of such loads on essential structural ,
integrity. Uc expect this review uill be conpleted in about one year.
The purpose of this testimony is' to prpvide (1)" a more detailed discussion of the phenomena leading to the imposition of loads on the reactor vessel i
supports; (2) the NRC staff's basis f.or assessing the adequacy of reactor vessel support designs for the UNP 1, 4 facility.
y s
?
i e
- e r- \
3 Loadlug Phenon.eua In the unlikely-cvent of a rupture of Pi1R primary coolant system piping in ,.. a location uithin the reactor vessel shield cavity, transient loads ori inatind from three principal causes taay be transmitted to the reactor C
vessel support system. These are:
1.
Blowdoun jet forces (rcaction forces) resulting f ron the rapid expulsion s of fluid. .
t 2.
Transient dif ferential pressures it$ the annular region between the outsLi of the vessel and the inside of the shield uall (the vessel cavity).
- 3. . Transient dif f erential pressures across the core barrel within the
( reao, tor vessel.
Standard design procedures are available to account for the first of these forces (reaction forces) .1/ The "dif ferential pressure" forces, (produced -
I both internal and external to the reactor vessel) are, houever, three l dimensional and time depend'ent and requir,e sophisticated analytical procedurc to translate t.hcm into loads actitig on the reactor support system. Loads -
from all three sources are resisted by the inertia of the reacLor system components, by the coolant loops, and by the support members, and restraints ,
I a
of other components of the primary coolant system as well as the reactor i
pressure vessel supportn.
'd!SI Standard 11176 and Standard neview Plan 3.6.2 contan ace'eptabic ucthods for computation of the basic,bloudown loads.
"' ' ~ ~ ' "
- ~ ~ . - . ~ ~~ C :.".: = = =. = . .. ...-.... .... . _-
_~ __ _ _ _ . _ _ __ _ _._ . .-. _ _ _ _ . . _ _ _ _ _ _ .
The transiene dificrs clat prensure acting externally on th2 rca. tor
. .ad.cl is, .*; result of the flow of the bloudown effluent into :'. :- reactor e
cavity. The magnitude and the tiac dependence of the resulting forces dependa on the nature and the size of the pipe rupture (i.e. circu :ferential or slot orientation, full area or restraint-limited arca), the clearance between Ehe vcsael ani! the shield vall, and the size and location of the vent openings leading from the vescel cav'ity to the containment as a whole. <
For some time, refined analytical nethods have been availabic for calculating r
these transient dif ferential preocures (multi-node analyses) . The recults n0 such analyces indicate that the consequent loads on the vessel support systen calculated by Icss sophisticated methods may not be as conservative as originally believed for earlier designs. Accordingly, the design of the RPV supports for Ll'c t!!!P 1,4 facilitics should be based upon the use of
( 1.tinode analysis for external transient differential pressures due to blowdown.into the reactor cavity. I!awever, in typical canes, these loads n'en sign 111cantly lens than those associated with the internal differential pressures acroes the core barrel.
i The inte'rnally. generated loads are due to a momentary differential -
pressure which is calculated to exist across the reactor core barrel when the pressure in the annular region betbeen the core barrel and the inside of the reactor pressura vcsocl wall in the' vicinity of the nozzic connected to the ruptured pipe is ascuned to rapidly decrease to the saturation pressure of the primary coolant due to the outflou of fluid. Although the depressurization wave travelu rapidly around the core barrel, there is a
" cry. short period of tine (in the order of 40 to 60 millisecondsj during
..lch the pressure in the annular region opposite the breah location is
a e
calculated to renai tt, or near, the o r igi:. :1 rea. r c v c : ting pre.uure, us, transient asy:: netrical forces are encrted en the core - cci and the vcsuel vall which ult.hnat ely result in dynanic load: on the vesse] supp rt r.yo teta. It is the ef fects of these asyrer.etric dyna:t:ic loads rhich vere unconservatively accounted for by the licensee originally reporting this problen, and uhich in some other cases riay he underestimated.
Although of varying magnitude, bloudown reaction forces result fro:n postulating a rupture at any location in the reactor coolant sys t e:n. The i
peak reaction forces on the vessel supports w uld occur as a result of a postulated rupture adjacent to the supports, i.e. in the ima cdia te vicinity Vi 6- of a noz::le. The loads resulting f ro:a asyn.tetric cavity pressures vould be b present only if a rupture is postulated at or near any of the vessel n.>szles e so that a significant portion of the effluent f ro a the rupture vould be rected into the vessel cavity. Inertia forcer, resulting fron the response of the vessel internals to tran:icat differcnLial pressures across the core barrel ui]l also accrue as a result of a postulated rupture at any vessel norzle or in either a hot or cold leg pipe, but the r.agnitude and direction i
of the resulting loads varies raarhedly with the location of the rupture postulated due to friction losses, propagdtion time of the decompression uave and vessel geo:netry. Peak load. values occur uhen the postulated rupture is at the location resulting in the shortest propegation tiue of the deco:n-pression uave to travel through the reactor internals. In consideration of all these factors our interest must therefore be focused on a specific l
l r l l
l l
o n-e rupture nt the cold eg nozzle uhere t.h: prercqut,ite cond!La ;
2neunscu above extst.2/
+
\
- l ,
1 1
i 2/
- A pont ulated rupture at a hot leg nozzle raty yield pc ak cavit.y pressure t
forces, leut a very ni,nificant reductic,n in the futeranlly ger.; rated l inertia forces uare than con.pencate:s for thic increauc.
l
- l i
l
- l l
l
1
~
A -
On the iar.l, o( our revi .of thit. Innue, we conclude th thene .; recs crai be F.
- 3y taken into consideratica during the denir,n of the f acili_y , utilizin;.
inethoda cuch au those descrJbed above, and if' the applicaat proposes to use' such v.ethods, there in rennoncbic acsurance t ftat the pu'alic health and cafety will not be endangered due to decir,n of ItPV supports.
e O
h O
9 S
e b
O 1
k e
O t
e e-e
I O ATTAGDOWT 3 PROIDSEI) SUPPLDEVr 'IO OCT0flER 15, 1975 TESTIt.D.W E
V. S. IG NAN
s 6
'Ihe staff sub!aitted supplemental testimony on October 15, 1975, concerning reactor pressure vessel supports. On page 7 of this testimony, one paragraph states that the methodology necessary to model the reactor coolant syste:a is available to reactor vendors so that they can properly account for three distinct types of loads applied to the vessel support systen.
'lhe paragraph following then further discussed our evaluation and conclusion concerning cnly one of the three components of strpport system load, while referring to an applicant response. The supplemental testimony herein addresses the two remaining components of support system load, and draws an overall conclusion on applicants current status with respect to the reactor vessel support system safety issue.
Applicant has responded to NRC concerns regarding vessel support systcm loads in their letter report to the staff of September 3,1975. At this time, our evaluation of this report is not complete; but, ue can draw some conclusions on applicant's submittal.
Regarding calculation of pressure differentials in the cavity in which the vessel is installed, applicant's letter report described hot and cold leg guard pipe designs intended to reduc reactor cavity (external to the vessel) pressure differentials. Analy 1 procedures used to calculate reactor cavity pressures are gene m ,
cceptable to the staff, although detailed review and evaluation of applicant's assumptions and calculated results is not complete.
Regarding jet forces associated with reactor coolant blo.edown, applicant has stated that these forces will be examined and accounted for using staff approved techniques as presented in Standard Review Plan 3.6.2 published by the NRC.
Our overall assessment of applicant's appreach to the ITV support issue for the hKP-1,4 facility is, then, that we have reasonabic assurance that applicant can and will calculate the relevant forces and analytically apply s:uae to assess the adequacy of RPV supports, with .the provision discussed in my October 15,. 1975 testimony regarding vessel internal pressure differentials, h'e find that, pursututt to 10 CFR 50.35(a), the applicant has described the proposed design including, but not limited to the principal engineering criteria, and that the further technical and design information required to complete the safety analysis can be considered after a Cp authorization, and that there is reasonabic assurance that the infomation will be supplied by applicant and reviewed by staff prior to submittal of the final safety analyais report.
9
UNITCD STATCS.
NUCLEAR HEGtlLATOltY COM.'aff W W AStfiNG TOfJ. D. C. 20 b 57.
OCT. 2 9 E Docket Nos. 50-460 -
and 50-513 .
V.THRU:
A. Moore, Assistant Director for L.ight Water Reactors, Group 2, DRL
. I A. Schwencer, Chief, Light Water Reactors Branch 2-3, DRL
- 3 WNP-1,4 HEARING TESTIMONY ON RE5CTOR PRESSURE VESSEL SUPp0RT ISSUE I
I believe that the testimony filed in this case on October 15, 1975, regarding reactor pressure vessel supports, should be supplcmented tu include some material recognizing and addressing the specific issue Tg of containment subcompartment differential pressure analyses (SPA). O The rad safety hearing is scheduled for November 11, 1975. N
<)
Applicant, in a recent submittal to the staff (Attachment 1) con:erning O the RPV support issue, presented a revised design (they have added guard pipes on hot and cold legs) and analysis of the subccupartment @
pressure differentials.
Their new analysis uses essentially the sama
, computational techniques as the earlier design analysis which we -
reported acceptable in Supplemant No.1, June 2,1975. But for the revised design, the Appiicant claims much reduced reactor cavity @
pressures from that approved by staff in Jun21975. I feel tha Coard Q will inquire into staff review status on this issue quite apart from C the fact that the SPA pressure-time histories are only one of several A inputs to the total loads on Rpy supports. 3 The staff testimony of October 15,1975 (Attachment 2, see page 7, last paragraph) states that we are confident that if Applicant uses certain known analytical techniques like WHAM, Applicant can, in our judgement, adequately predict the RpV loads resulting from forces applied by the
" thermal and hydraulic response of the core to the postulated rupture".
The reference to the WHAM code at this point in the testimony applies only to the calculation of " internal pressure differentials" rather than to subcompartment pressure differentials or to jet impingament loads.
We must also provide testimony to our conviction that Applicant can predict pressure-time histories within containment subcompartmants, either with the engineering criteria and methods he proposes to use or those which we require him to use. We must similarly conclude that Applicant can adequately design for jet impingement loads on affected structures or components.
e s
i
/. A. 1oore A. Schwancer .
I suggest that the ASLB will not be satisfied with a verbal statement by the LPM at the he'aring that we are considering the SPA and jet impingement loads as part of the overall " generic" issue and expect a satisfactory resolution. He must provide some bases for the con-clusion, and then draw the conclusion, pursuant to 10 CFR 50.35(a),
that Applicant has described the proposed design, including but not limited to the principal engineering criteria, and that such further technical and design information as n.~ay b'e required to complete the safety analysis can be left for later consideration, and that this information will be supplied by Applicant and reviewed by staff within the time required to be included in the Final Safety Analysis Report.
I've included a suggested supplement (Attachment 3) to the October 15, 1975 testimony, and respectfully request your early consideration of this material .
At 6;/
Thomas' Cox, Proja:t !!anager Light Water Reactors Branch 2-3 Division of Reactor Licensing Attachments:
- 1. Ltr fm W?PSS to R. S. Boyd dtd 9/3/75
- 2. Supplemental Testimony '
by V. Noonan
- 3. Proposed flodification to Testimony by V. Noonan k
- F s
e e
s
s L
- r r-O.17/;W3$uCJ f.
) Attacirtent 1
- c. s
- . j)'
Washington Public Power Supply Syst2m
'/.d A JOINT OPERATING AGENCY -
. /
U.y
$.G
$ r. c. soi see sooo cio wm.~oron wo n.c.u. o. w.w. cron ., m r..o,. no2 no.o e September 3, 1975 Docket tios. 50-460 G01-75-181 50-513 Mr. Roger Boyd, Acting Director
. . Division of Reactor Licensing .
Office of flucicar Reactor Regulation U. S. Nuclear Regulatory Commission Washington, D. C. 20555 pN
Subject:
WPPSS NUCLEAR PROJECTS fiOS. 1 & 4-REACTOR VESSEL SUPPCRTS
Reference:
- 1) Letter, i:0 Strand, UPPSS, to A. Giambusso, sg)
NRC, same subject, dated July 18, 1975. c_-
- 2) Letter, NO Strand, WPPSS, to Benced C. Rusche, O NRC, "Supplemer.tal Limi.ted Morh Autharization,"
dated August 21, .1975.'
Q
Dear Mr. Boyd:
The WNP-1/4 ACRS letter mentioned a concern for the lead; on the reactor vessel support structure for certain postulated loss of coolant accidents.
In Reference I we coamitted to submit a report which would address this concern. This report is attached.
The report commits to the addition of pres *sure retaining guard pipes to those portions of the primary system hot and cold legs that are within -
the reactor vessel subccmpartment. In addition to reducing the uplift and asymmetric loading on the reactor vessel, the guard pipes also allow for the design of the reactor vessel subcompartment for lower differential pressures.
We request that the Staff review of the vessel subcompartment differential pressure portion of this report be performed as soon as possible. As noted in our request for a Supplemental LUA (i.e., Reference 2) this analysis could be considered as an unresolved. safety issue if it is not reviewed and approved in time to facilitate the issuance of the Supplcrental LWA, since the analysis supersedes that contained in the PSAR. Please note that the supplemental LWA request included the pouring nf the
, reactor vessel subcompartment walls. Prompt revicw and approval of this LUA with no
~
analysis restrictionwould on the allow for the issuance construction of walls of the supplcmenta
internal to the Containment.
I )j Also, the design of the subcompartment' and procurcrcr.t of raterials is proceeding (at our risk) for the lower differential ;.ressures.
'e . , ,
' . i
, o .
Se: tenter 3,1975 Mr. Roger Boyd .
G01-75-181 Page 2 To faci.litate' this review, we have made the report as complete as possible and have addressed areas that have bacn a problem in the past. Also, we are committing to the design of the reactor vessel subccmpartmcut for a dif,ferential pressure of 110 psi which is 120% over the calculated value for the most critical node rather than the usual 40%. It is hoped that this commitment will eliminate any difficulty should the Staff calculate '
a differential' pressure somewhat higher than indicated in the attached report.
We would welcome the opportunity to meet with the Staff to discuss any aspect of the report. Also, we will incorpordte any of the attached ,
material that the Staf f desires into the PSAR by amendment.
Very truly yours, t
' p - .
8 JJ 17L uM N. O. STRNID Assistant Director -
Generation & Technology ,
NOS: AGil: tim ,
Attachment ,
cc: CR Bryant - BPA, w/att.
. TH Cox - IIRC, w/a tt. 1 JR Schmieder - UE&C, w/att.
EG Ward - B&W, w/att.
1 4
i b
r p
\
v d
Attachn.~nt 2 - .
SUPPLEMEtiTAL TESTIMONY 0F .
NRC STAFF
.2 OH idt f4}
REACTOR FRESSURE VESSEL SUPPC?.iS Q
- IIASHII:GT0il KUCLEAR PROJECT l'05.1 At:D.4 DOCKET fiOS. 50-460 A. D 50-513 BV-VI! ICE?tT S. tl00iAN G
m 6
\
r
(,
\ .
I
e: ,
4 g
i ,
ll!41TEli STATE:: OF N41:ltiCA
. NUCLEAll RI.CllLA'!O!;Y CO:i:lISSION t
BEF01;E TI'IE AT0!!LC SAFETY A!!D LICE::SJ!!G 1:0.'tPJ)
~
In the Itatter of .
). .
) '
WAS!!L':CTON PUlil.1C ' POWER SUPPLY SYSTE!! ) i
. )
(Washi,igton Public Power Supply System ) Docket flo. 50-460, 50-513
. ) .
~
Station, Units 1 and 4) )
. TESTJf!ONY OF VI!;CE':T S. t;OO !A!!
In troduc tion On llay 7,1975, the !!uclear Regulatory Com:iission (!!RC) staff was inforced by the Virginia Electric and Power Cocpany (VEPCO) that the effect of one of the i[ loads pentulated to occur in tha unlikely event of a specifie loss-of- cooJent
- (' acc.', dent had been cale
- : lated uncor.scrvatively in the original design of the reacter ves' sci cupport system for I; orth Anna Un it s 1 L 2.
Safety analyses for nuclear plants are usually based on the acru:nption that th e reactor vessel remains fixed in place or undergoe.c enly very unall movenents l under the full spectrum of postulated design basis accidents. The reactor ' '
, vessel supports are therefore designed to sustain, uithin conservat.ive stress lir.ii ts , the loads calculated to result from postulated accidents, such as pipe breaks in th.c main coolant loops. In recognition of the fact that one or more of the vessel support loads calculated to result from a postulated LOCA was found to be greater than the loads used in the original design analyses, the 9
cffect of the increased Joada ,on vessel motion (.md therefore on the integrity of nt:tached piping, on the operability of control rods, and on ualn:pnance o f
- e. . ele core geometry) requ, ired;rcevaluation. , ,
An noted Jn the Preliminary Report of the URC staff dated h ly 11, 1975
( 2 is the N.",C staf f ? u opinion that the quer.tionn related to L'..c- trea tment-
. e of transient LOCA londri in the design of teactor venuel support r.yntems that hava orinen for North Anna Unita 1 and 2 taay apply to other PUlt facilities. We have, therefore, initiated a systematic review of this matter td'dctermine bou the design basic loads ucre taken into account on other PWR facilitics, and what, if any, nessures nay be required for specific facilitics. * ,
i The results of studies reported to date by the licensee for North Anna Units 1 and 2 indicate that, altitough thic margins of safety for 1stteral
, loads may be less than originally intended, the reactor vescel support
. systen will retain essential structural integrity. This naann that the
" ssel rcrains fully supported, t:hile undergoing coac translation, and
(' hat the ultir.nte consequences of this por.tulated accident are no uore severe than accone.odated by the original design. Uc are in the process of an independent evaluation of the calculated transient loadr. on the support systems, and the effcet of such loadr. on encential structural ,
integrity. We expect thin review uill be conpleted in about one year. -
~
The purpose of this testimony is' to pr, ovide (1) a note detailed discussion of the pheno:rena leading to the imposition of loads on the reactor vessel supports; (2) the NRC staff's basis for. assessing the adequacy of reactor vessel support designs for the UNP 1, 4 facility.
.( .
O s
- . . I t . .
_Londlun Phemcwua
- In the unlikely event of a rupture of Pt.T primary cooltint system pipiug in a locatiott within the reacter vessel chield cavit.y, transient leads originatin ; from three prjucipal causes may be trans aitted to the reactor vessel support system. These are: l l
4
.1. Blowdown jet forces (reaction forces) resulting f ron the rapid expulsion j
- of fluid. .
- 2. Transient dif ferential prcosurgs i:$ the asutular region bet.wcen the outsi.I of the vessel and the inside of the shield vall (the vecsci cavity).
- 3. . Tritusient dif ferenLial pressures across the core barrel within the
~
[ reno',Lorves$cl.
l Standard desi;;n proceduren are .;vailable to accoun'. f or t.hc first of these forces (react' ion forcer.) .1/ Th "dif f erential pressure" forces, (produced !
I hoth i$terual and external to he reactor vessel)'are, however, three i diinensional and time dependent and require nephicticated analytical procedur(
to translate t hen into Joads acting on Lbc reactor cupport sysi;cn.
Loads f rom all three sources are resisted by the inertia of the reactor syntem componentn, by the coolant loops, and by the support me:abers, and rc::traints of other co:nponents of the primary coolant nysten an well ad the reactor i
pressure vennel supports. !
t i
( UlSI Standard Nt76 and Standard Review Pl.,n 3.'6.2 contain ace'eptable sactbods for computation of the basic, blowdown loads. ,
. l
- .**+****.eeme-
- e eeeee eeee,, e e me. e , e - - _ ..,-ww... , , , , , , , _. _ , , , , , ,,
I o- .,
(. The tranulcat dif ferentini prcusure acting externally on the rc.. tor Asci is a result of the flou of t.hc blutidown affluent into t: a reactor e
cavity. The magnitude and the time dependcuce of the.reculting foreca dependo ei the nature and the size of the pipa rupture (i.e. circumferential or slot orientation, full area or restraint-limited arca), the clearance between the vessel and the shield vall, and the si::e and location of the vent openings leading from the vessel cavity to the contain nent as a whole.
por' some tir.te, refined analytical r:cthods have b6c.n availabic for calculating i
these transient dif forential pressures (culti-node analyses). The results of such analyses indicate that the consequent Jcads on the vessel support system calculated by less sophisticated methods may not be as conservative as originally believed for earlier designs. Accordingly, the design of the RPV supports for the W P 1,4 facilitics should be base.1 upon the ut.c of
{- 'tinede analycia for external transient dif ferentini pressures due to bloudou;) into the reactor cavity, li=:ever, in typir.d c . m.: , these lc~; is aYe significantly less than those associated with t.ht. int.crual dif ferential pressurcs across the core barrel. >
I e
The internally generated loads are due to a nonentary dif ferential -
pressure which is calculated to exist across the reactor core barrel when the pressure in the annular region betbeca the core barrel and the inside of the reactor pressure vessel vall l'n the vicinity 3C the nozule connected
~
to the ruptured pipe is assumed to rapidly decreene to the saturation pressure of the primary coolant due to the outflou of fluid. Althagh the depreucuri:.ation uave travels rapidly around the core barrel, thcre is a wery short period of time (in the order of 40 to 60 nil 11 seconds) during
..ich the presourc in the nunular region opposite'the break location in
-~. - , , - , _ , , - , - - - -
{
calculated to rur.nin ., or ucar, the or.lginal react oper...nu pecunare.
'( .uc, trannient. asy: xact.rical force: are encrte1 on the core .rrel and the vessel wall which ultimat cly recult in d' namic y loads on the vensci suppo* rt systeu. It in the ~cf fects of these any xmtric dyna:.;ic loads chich vore unconservatively accounted for. by the licensce originally reporting this problen, .and which in some other casos c.ay be underesthaaled.
, Although of varying magnitude, blowdown reaction forces result fro:n
, postulating a rupt~ure at any location in the reactor coolant system. The '
peak reaction forces on the vessel supports would occur as a result of a pcctulated ruptacu .'idjuecut to the supports, 1.c. in the immediate vicinity of a nozzle. The loads resulting ' Iron anyinaetric cavity pressurec could be present only if a rupture is postulated at or near any of the vessel nozzics so that a significant portion of the ef fitient- from the rupture would be 5
rected into t'he vessel cavity. . Inertia forcen resultinp, frou the response of the vessel internalc. to transient dif fer(nt.Ia1 pressures across thc. core barrel will n.).no accrue an a result of a postulated rupture at any vessel nozzle or in either a hot or cold
- leg pipe, but t.he magnitude and direction I
of'the resulting loads varies markedly with the location of the rupturc ,
, postulated due to f rict.lon losses, propagI: tion tiue of the deco.rpression wave and vessel geoactry. Peak load, values occur uhen the postulated rupture in at the location result.ing in the shortect. prcipagation Liue of the decom-pression uave to travel through the r6 actor internals. fn conn.fderatiun of all these factors our interest must therefore be focused on a specific !
i s
t
.( . .
e e
g -o-upture nt the cold led nozzle where the prercquinite con.itiens I
incussed above exi:;t.2/ -
2
\-
- s s
2/
- A postulat ed rupture at a hot. J eg no:: le r:ay yield reid: cavity ptendure forces, but a very 31.nn.lficant rc<luction in the int cranlly gener.'.ted inertia forces more than compencates for thic increauc. '
( '
.. y . .
3 .
S' thete are co,<:Jdei chle dif fe. .reu Jn the rea'ctor nuppur. : / :t:cu de:.igns
(
for varfous facilitics and in the design marginn and methoda .ct:.plcyed, the
, previoun1y d:veloped Joss conservative calculatJoan of these " differential prensure" loads may or saay. not have a significant affect ott the adequacy of the vensel support cystera for a specific facility. Additionally, t!:e actual loads imposed on the ver,nci support noc.bere are dependent on the dyna:aic renponne character--
istics of the cntire reactor coolant syctem. If the prjncipal response freq'uencies of the reactor coolant cysten are uc'll rcr..oved f rota the frequency of the forcing load, only a minor porti.on of the thcoretical load vill actually i
be trans:.:Itted to the su ) ports. If, on the othct h:;nd, the natural frequency of the support . system is close to the ' forcing frequency, a dyiucaically at;plified load must be sustained by the supports.
'11 ithodology necennary to. i.:ollel the complete rer.ctor coolgnt cyrte:a in
\
sufficient dcta.fi to deter:ainte analytically the rwginitudes and phc?.e relation-ships of the vessel support syntec: loado fro:a 1Eterre:1 prenuuro dif f ert.ntial s, pressure dif ferentialc in the cavity in ubich th- veu .el in innt.alled, and the jet forces associated uith reactor ' oolant c bleudoun are nou u Lthin the cct.::bility ,
of the nuclear necan system vendo,r. '
t Proper design for this co:1ponent of the RPV supportu requiren carefoi a :ccar. ment of the thernal and hydraulic responuc of the core to the pectolated rupture, using techniques sl:r.Jlar to thone cuployed in UlWI 3/or other analysin twtho:h.
. i I
3_/ Conputer Prop, ram Ull.*J! for Calculot.Ien of Prenmire,Veloeity, mal Force {
Trancients In hlquld Pilled Pipiny.,Uetuurks,1:niner 1:nghtrern Cupori 's .
67-49-R, Stanislav Pubic, (Novett.ber 1967) .
, t .
t e
v l 6
8 .
r
-s -
which complet ely character b:e the therm.il and hydraulle renpt ..- of spec.lfic
- f. , ores. Unjug this data, the vcucci loadiagn are detemined , stat e- of-the-art dyna:::ic renconce analytical actboda. Both of these have been dir:-
cussed with IMU and the applicant and they are in the proccus of providing a response.
Thus far, the resposice has not bcca ecmplete. Unicac methoda nuch as those described above are clearly made part of applicantn' design methods, the staff believes that they should be required au conditions of any CP issued.
Although vc believe that it is itportant to annure that proper calculational
' techniques are used, the r,cassesa:::cnts thus f ar carried out by the st af f indicate that, although the original desigiis ucre based on incomplete analytical technLquen, the RPV support designs are still adequate even when
, reassessed using the core sophisticated techniquen now uced.
I \'. . t. ecoc,. of 1: orth /.nna Units 1 and 2, the Archit.cet Fagineer, Stone & trebster Engineers, deternilnod by bsc of sit:plified, concerective sacchc.d3 that the design stress 11mitu originally specificd for : orth /.nua Unit.s 1 and 2 uould bc cxeceded uhun the "dif ferential pressure" loads ucre properly considered, and that ranterial failuro of cene elenents of the support synt.ca night occur. The original design of the reactor vessel suppor,t system for :: orth /.nna Units 1 .
and 2 had been ba, sed on purely clactic actic.n of the nupport. elc uen ts . This in an acec[itabic si: plified approach that renults in the de: inn banir. Inada for the vessel support structurco being far less than the inherent load-carrying capability of the support structure r.:aterinin. It is partially for this reason that only limited vessel imtions would be crpected when t he hit.hne deaftn loada are taken into account, and other than clastic action in considered. -
- e
(
O s
--.-__-_,3 - . - - , ~ , - - - . - -m----- ,- . - _ - - -
o -
.1
- o. . . -
In the Stone & t cbater .iJ,yne: , the peak valuer. for alt of the er; short ion Jriailn have been d ir ect.ly nn:: ..cd whereau . in rea)) t y, c:: .ut ion of the physical principles involved in the trans:::Issioti of .shoch utr.v:: through th6 reactor synten and the pat. cage of fluids from the postulated bccak ::.ahe the simultaneous occurrence of peah loads improbable. Uaing ti e co::.bined peak loads, Stonc & IMint er perforr.ed an initial analynic 'that . indicated stress Icyc1c greatur' than th,ose used in the orir,inal dcaigry in sor.c cle:.icnts of the reactor vessel supports that resist lateral.iaotipn. A conservative analysis uns then performed under the assumption that all. lat.cral restraint un's retaved. The s
results of thaso anelYFee indicate that t!.ctu would be a reactor pressure vessel lateral novement of, chout one-half lach, which nagnitude vould have no significant adverse effect on the functioning of the control redc or on the integrity of inter-connecting piping such as the etacrgency core cooling cystem OZCS) injection
, l' '-4 /
7 the total Joadt. 1.uing re:;inted by the everall reactor ceaL..t nya r.eu r.nd
}.
it a nupports. No gross- lorb, of at.ructural ru;; ort. and rc t e: Jut J e:..'
t.t; te clearly unacceptable rotlon of the vecsc1 hnu been Indiccted L.y ci 'c.
. ef these ana.l yc es .
, . I A cimilar conclusion resulted fro:n nore cophist.icated analyses conducted in generic ntudies by 1.'qstinghouso5/ In these noro .sophinticated annlyses the ti=c
[/Amendecut36toNorthAnnaUnits1&2,DocketNo.s50-338and50-339.
,5_/ !!inuten-Su:: mary of Meeting dated May 29, 1975 regarding 1Icstinghour.c 's analyu.lu on reactor internals renpenne result'ing iroa I.0CA.
09 I
- i S t I
- 6
-- - - ,,-m- -7.- , , , _ - , , - . . , . ., ,% . , _ , - y- ,,7 -.
hint ury oC* the Joading phesiamena va:. canaldered, as the flui '-ct ruct ure i ' actions reculting froo raovement of the core barrel under Lt. influence of the asynmetric pressure loads. . .
The ceverity'of the load 1.ng is reduced by thene concideratitnc. All design basic rupture 1,ocations and rupture flou arcas
~
were ac defined in We.itinghouse Topical Report 1.' CAP-80826/ "Pipo Breahc for ~
the LOCA I.nalynin of the Uentinghouse Primary Coolant Loop."
,The worst case rupture for these analyses was detergiined to be an instcntaneous suillotine rupture adjacent to'the cold Icg vessel nozzle. Structural analyses based on the loado developed by this worst cace loading denonstrated that Westhghouse reactor coolant nupport systern noa beJng designed can suctain these loads andgcraf.n within the conservative design ba. tis stres:. liuits.
?
a Since the reactor coolant support syste:as now heing decigned do not cr.: ploy cub-stantial changes in structural configuration fror. tiestinghouac support syste..s of el ar'dedign, a significant ine[caint of confJd5nce ir, tained that the rer.ults s -
of the simplified analyses discusser!'above do scope the po:,sible conocquences of m .,. che design basis postulates. ~
~
g Although tiie , von'k donc thus far indicates that the RPV supporta previously '
9'* designed are still adequate when reassessed with more sophicticated techniques, .
- ;, s q %'
the staff nevertheless is inforning all facilitics in operation and under construction of the need for careful reansessoc:it. Sinilarly, for facilities seeking construction permits, applicanta nust adequately provide for these loads in their desig;ns as described above. '
6/ WCR-8032 was reviewed and approved by the NRC staf f on /.pril 15, 1974. ,
I I
D p
.eg '
e /
p
4 On t.he 1.e:.i:2 o f ou r t als..e .of t li : .
1ssue, we coaelp*!c that timuu .: : :: ce.7 can be q
fly taken into consideration during the der: inn of the facili:. , utilirirg . . .
methods such as those described al>ove, and if the cppliccat proposes to use' such methods, there is reasonable accurance that the public health and sarcty will not be endangered due to design of RPV supports.
4
. I e
e e
e 9
- t ,
4 9
9 6
- s 9
9
's e
e d
O O
O 4
i i t
9 k
4 o
9 4
38
[
t l
- 's.
t .
\ . .
l'
[ . l .
e 0 g
e
.. 1.
, _ ,_____ -- __. . _ , . ~ . - _ . . _ _ _ _ - - -. . _ _ _ - . . _ _ _ _ . . _ . . . _ . - - _ _ _ - - - - -- -
,..s .
ATTAGMINr 3 PROPOSEI) SUI'l'LININF 'IO OCf0 Bell 15, 1975 TESTIBD.W
_BY
. V. S. NOONAN e
W
} i l :
e I
\
i
'Ihc staff submitted supplemental testimony on October 15, 1975, concerning -
. reactor pressure vesrel supports. On page 7 of this testimony, one paragraph states tirtt the methodology necessary to model the reactor coolant system is availabic to reactor vendors so that they can properly account for three distinct types of loads applied to the vesser support system.
The paragraph following then further discussed our evaluation and conclusion concerning only one of the three components of support system load, while referring to en applicant response. The supplemental testimony herein addresses the two remaining components of support system load, and draws an overall conclusion on applicants current status With respect to the reactor vessel support system safety issue.
. Applicant has' responded to NRC concerns regarding vessel support system
' loads in their letter report to the staff of September 3, :1975. At this time, our evaluation of this report is not complete; but, we can draw some conclusions on applicant's submittal.
Regarding calculation of pressure differentials in the cavity in which the vessel is installed, applicant's letter report described hot and cold leg guard pipe designs intended to reduce reactor cavity (external to the vessel) pressure difforentials. Analytical procedures used to
. calculate reactor cavity pressures are generally acceptable to the staff, although detailed revic>. and cra.luation of applicant 's asstaptions and calculated results is not complete. '
Regarding jet forces associated with reactor coslar. o '
- , valicant has stated that these forces t.111 be excained .ml xx .v.xt for using staff approved techniqua as presented in Standacd Revie< Pina 3.6.2 published by the RC.
Our overall assessment of applicant's approach to the l'.A suppart issue for the t'NP-1,4 facility is, then, that we have reasonabic assurance that applicant can and will calculate the relovant forces and analytically apply same to assess the adequacy of RPV supports, with the provision discussed in my October 15, 1975 testimony regarding vessel internal pressure differentials.
h'e find that, pursuant to 10 CFR 50.35(a), the applicant has described the proposed design including, but not limited to the principal engineerim criteria, and that the further technical and design information required to completc the safety analysis can be considered after a CP authorization, and that there is reasonabic assurance that the information will be
~
supplied by applicant and reviewed by staff prior to submittal of the final safety analysis report.
e .
e
sQ ( -
Distribution: l NOV* O 51975 Docket Fil-4 5 RKSharma LWR 2-3 File RCloni s VAMoore LWR 1 & 2 Br. Chiefs Docket Mos~50-460 ^
T and 50-513 TEssig
' PStoddart RKornasiewicz D. Skovholt, Assistant Director for Quality Assurance & Operations, DRL l J. Collins, Chief, Effluent Treatment Systems Branch, DTR i W. Gammill, Chief, Site Analysis Branch, DTR J. Kastner, Chief, Radiological Assessment Branch, DTR ,
THRU: A.Schwencer, Chief,LightWaterReactorsBranch2-3,DRLh I WNP-1,4 RADIOLOGICAL SAFETY HEARING OELD Attorney E. Ketchen has informed me that the strategy for conduct of the subject hearing will require that all staff witnesses be on hand in Richland, Washington as of Tuesday, A. M. , November 11, 1975.
As we have previously arranged, the witnesses and their branch affiliations are as follows:
T. Essig, RA3 P. Stoddart, ETSB R. Kornasiewicz, SAB
- R. K. Shama, t.r;;onne National Laboratory R. Cloni, QA&O T. Cox LWR 2-3 Ori2inal Si7ned b']
T.Ccx Thomas, Cox Project Manager Light Wator Reactors Eranch 2-3 Division of Reactor Licensing
[1 e iran n -r d 9 l =7 z;,,1 T4' I 4.-
t y - , i ID orr .. , RL: LWR 2-3 RL C/1WR 2-3
.* a m a', TCox:pga AS .encen ,
oats * .11/$ 7.5- - 11/[22.5 ft u. e. oovannmant painfimo orF8cas lef*.sas.eee Forms AIC.318 (Rev. 913) AECM 0240 J