ML20212M369
Text
.
(
(
/( J;' rTi,.'
n.
Uf410ft CARDIDE CORPORAT1011
- 1.. -
p Rt.c _;, z 3
..n :
f~2 M WJ. '.jj.
Il0CtEAR DIVISIDft
.:a.p P.O. UOX X, OAK fifDCE. TEinf ES$EE 37333
' ] ge r.
p'
' I
!!ay 31,1970 2
/ t. M tr. s.
G
- d@%,..Eri..,,,, j ).
- u...cica 5;,, ug..g.
fir..lohn C. licKinley, Chief Project Review P, ranch No.1 US Nuclear Regulatnry Cor. mission Advisory Com :ittee on Reactor Safeguards Washington, DC 20555
Dear John:
Criteria Implementation Revicu licctings for Diablo Canyon
!!uclear Plant, Units 1 and 2 During the period knuary 16 to February 2,1978,. riteria imp!cmontation review r:eetings v:ere held with PG".E and Mcstinghouse in San Francisco, CA, and l'.ontceville, PA.
The stated purpose of these meetings was to provide thr! !!RC staff an opportunity to acquire ibe detailed infoir.ation nece:;sery to understand what had been done, how it hed been donc, and why it had l'eco done. Attendecs included PG&E and !!astinghouse personnel and their conselt-ants, I!RC staff personnel, ACRS consultants, and iltervenor gr::up repre-sentatives.
I attended part! of the meetings rela:cd to piping, mechanical equipw nt, and electrical equipment as an ACriS cone.ultant.
This letter q
will suonarize rny observations and impressions.
liy general impression was that the meetings provided'an excellent oppor-tunity to determitie the depth and quality of the engineering effort on the Diablo Canycn plant.
The face-to-face discussion provided an oppor-tunity for the inmediate resolution of comunication problems that frequently complicate reviews based on written information and formal bearings.
Calculations, drawings, test results, and support documents were made available as questions arose providing an indication of the depth of enaineerir:9 cffort on these items. Although no attempt was made to thoroughly check al.1 of the material, sufficient checking uas donc by the staf f to assure titit published doctn.:entation of results acrecs with the actu.al design and $nalysis.
This revicw provided more coiifidence thaa one ficquantly has in answers stated wit.h minimal support in written cc:muni-cations or in formal hearings.
I was favorably impressed with the docu-mentation of design and analysis inforaation and the ability to quickly retrieve this infonnation considering the fact that severisi years have clapsed since some of the original engineering work was done.
i One session-of the meeting with PGT.E was devoted to the interfaces l'etween l
structural considerations and piping, mechanical, ar.d ciectrical systems s
F 9608230464 860301 PDR FOIA PCR HOUCH96-391 s.
n
?
-s, A -It+ i
~
[
(
o.
IIr. Q.::r. C. 'e t:i iley 2
liay 31, 1978 c
(
considerations.
This session, which was attended by both structural engineering personnel and n:echanical engineerino personnel, was useful in assuring that a consistent approach has been usec' in the sciswic cvaluation and that interfaces are properly covered in the analyses.
The scismic input for evaltating piping, mechanical, and electrical sys-teus was discus.cd at some lenqth.
1he ucchanical cngineering review is based nn floor response spectra dei. ermined in the structural analysis and accepted by the Structural Engineering Emnch of f;RC.
The methods used in deternining seismic input for suppor't poiht located between floors and for including torsional n:otion were reviewed, pGT.E has inclu: led the torsional r:otion based un the actual location in the build-ing and has interpolated between floor spectra io obtain specific support.
point input spectra. 'destinghouse has taken a const rvative approach in their analyses, using a maximum radius for deternin ng torsional input and using th". ncn higher ilcor spectra to deter.aint support point motion.
Ito additional inforntion a; pears necessary assu:ainn that the Moor spectn used in the analy cs are accepted by the Structural Engincering Cranch.
lhe procedures un-J in coabining modes and spatial components.in the scisni:
analyses were revictied.
The methods used for biablo Canyon credate the issuance of L':C R-gulatory Guide 1.92 uhich is nou in use.
The method used fer Diablo Can.ien ieouires that:
(a) the scismic responses due to vertical excitat. n t:e co..Sined with the responses iue to e:: citation in g
one of the tuo orthoconal horizcotal directions on on absolute sum basis for each made; (El the resul tant modal responses be combir.cd oy the square root of t.he sum of squarcs. method.
The pIocess is repeated for the other orthogonal horircntal direction and the controlling value used for desi n.
b The present regula tory Guide 1.92 presents a..uthod v;herc:
(a) for cac h componen t of care r;uake c:otion, the modal responses for closely spaced frequencies are first. combined by the absolute sua nethod and these rcsul ts are then combined t;ith the other modes by the square root of the sum of squares method; (b) the responses due to all three components of earth-qual:es (ttro horizontal and the vertical) are cer..hined by the square root of the sum of squares wthods.
Each of these methods has some conservative fea tures.
!!cwe ver, the degree of conservatirm in nach melhod varies with conditions and neither metimd is consistently uare consci vativc than the other. Sow sem i tivil.y studies have been wd" to ce cpare the two methods for Diablo Canyen piping loops.
The review of thes studies tended to give assuiance that the methods used for Diablo Cenyon vero adequate.
liowever, the regulatory guida motheci resul ted in grea ter loads in sce locations, and additional information was requested to. verity the adequacy of the design approach.
The effects 'of relative scismic anchor movements on piping systems and the combination of resulting loads with inertial and pressure leads were
~
~
' 3 r
b
{
(.
Mr. John C. McKinley 3
May 31, 1978 discussed. A seismic anchor rovement study had been reule at the request of the MP.C s ta f f.
lhis study covered ciqht piping lin: s and showed thc t -
the total piping stresses for these lines weie within the allcw.:bic limi ts.
It was reqerested that. the present study be expanded to include specific load corabinations and justification (by comparison) for piping systems not included in the present study.
The interface between different scismic design class piping systems was revie.:c t.
The cencern was tint excessive loads could !:0 transmi tted to scismic design class 1 Syster anchor by suprert failurcs in the seisuic design class 2 portions of the sy.tcid.
It u.s request.ed tha t, when process lines with design class changes are a part o' systems rcquired for shu'uee:n or branches from thru, the pipe support.s fer the clats 2 portion be cedifiti to meet the spacing criteria guidelines imp 1m:nted for the class 1 partion of the piping syt tui.
The combinat. ion of i.0CA and scismic Inads for branch lines in piping system t as reviewed.
It uns requeste.I th3t resul ts of analyses be provided fer several typical hianch lines.
It was indicated that an equivalent static LOCA branch line analysis may be ined if acceptable justification of the dynamic aglifica'. ion factors is provid:d.
Scisnic c,valificatinn tests for valves have indicated th3t a few valvas havc natural f:cquent.'*. belew 33 He.
Since the pi;>in:; systen analyses have asstracd thr.t tbc valves can be nodcled as ri i9 :i hofies, t.he potential effec:
of flexible val.es on the piping response was discussed.
It appears tha t wN ihe louer freyeries may be associa.ted iith relatively smsll masses in the valve actuator a:n the resulting avlification of piping response c:ua to resonance nay be si. 211.
Further, some of the values may be stiffened to remove the appare:e lower. frequencies.
Scismic tests on some vessels have indicated the presence of shell codes.
Additicnal infomation was requested to indicate f.hc tede shapes used in the
"~
analysis of various tanks and heat exchant:ers and 1.0 provide justification of why other modo shapes, such as shell modes, have 1.nt baan considered.
It was further requested that the possibility of slashing effects in the boric acid tanks he evaluated.
lhe damping values used in the Ilosgri cvalua Lion wcee discussed.
It was agreed that, in general, dam;iing values should agree with those given in imC Regulatory Guide 1.61 but tha t experimental test data may be used when appropriate. A value of C damping was used for the reactor coolant system (as compared with 33 in the regulaim y (;uide) based on actual plant tests ruade by l?estinghouse.
It was requested that the similarity between the liiablo Canynn coolant system and the system tested Ie demonstrated to justify use ~ f the experimentally dgtennined d roping value.
It was also o
requested that the ef fect of structural st.if fening on damping valurfs should be assessed f or equipment that, has burn modified or stif f ened and the v
results suntnarized.
B e
s s
8 y
usars.am.sa meC~%
6
(
(
r%.
o.
fir. John C. licKinley 4
liay 31, 1978 In sunnery, I third: thr' review meetings were Leneficial.
Al though additional inforrotion was requeste d for some of the items discussed, I helicve th< consensu. was that the inectings resulted in an increase in confidence in the seismic evaluation of the Diablo Canyon plant.
It appears that the additional inforcation requested can he supplied and the staf f evaluation completed without excessive trouble or delay, Sincerely, 4,5 ht d I. W. Picl:cl 111P/blta cc:
C. P. Siess Reeut o s O
=
.,,i
, s:
s p
4.: