ML19209B554
| ML19209B554 | |
| Person / Time | |
|---|---|
| Site: | Nine Mile Point, Brunswick  |
| Issue date: | 06/07/1979 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML18136A035 | List: |
| References | |
| NUDOCS 7910100110 | |
| Download: ML19209B554 (3) | |
Text
.
s SRUNSWICK STEAM ELECTRIC PLANT, UNITS 1 AhD 2 REVIEW OF PIPING REANALYSIS PER I&E SULLETIN 7F.07 SAFETY EVALUATION REPORT INTRODUCTION / BACKGROUND In their April 24, 1979 response to I&E Sulletin 79-07 Carolina Power anc Light Company (CPht) stated that the recirculation and main stear piping nac ceen analyzec by GE using a computer coce that comoinea cirectional seismic responses oy algecraic summation.
All other safety relatec piping was analyzec by UE&C using a computer coce that combinec cirecticnal seismic responses by algeoraic summation. CP&L has supplied st..:plemental information on this suoject at meetings with the NRC staff anc in tetters cated May II, 21, 22, 29, anc June.4,1979.
DISCUSSION CP&L has statec that a reanalysis of all affected piping in :ne "as-Duilt" concition will ce completec witn the results showing all piping stresses remaining below.tneir allowa::le values, as specifiec in :ne 3SEP FSAR oy July 21,1979.
Accitionally, all loads on attached equipment (no::le loads) will De acceptaole.
Upon completion of mocifications to certain pipe supports whicn were cetermined to De originally uncercesigned, all pipe supports attached to safety relatec piping or equipment in tne plant will De "operacle" and within FSAR criteria. These mocifications will De complitec prior to return to pcwer operation.
The recirculation ano mainsteam lines were reanalyzec cy GE using PISYS.
The responses from two directions, the most disacvantageous comoination of one
.urirontal with the vertical, were comoined oy the aosolute sum anc the results were witnin FSAR allowaole. Refer to Evaluation Section for reanalyses done by UE&C.
The licensee's response to IE bulletin 79-04 states that no VELAN swing chect valves are in any of the affected piping.
Further, IE Bulletin 79-02 was accressed wnen. piping support mocifications were founc to De necessary.
The licensee has stated that the reanalysis has no effect on pipe creak criteria since tne postulatec creak was analyzed to occur at any point on tne pipe, insice or cutsice containment.
EiALUATICN
-ne reanalysis ecnni0ue em-loyec,as a lumosc mass res:c".se Ipectra mocal Inis Cy".ami inaly sis pr0cecure is an ac~e":ti le ~str.00.
Ihe i~alysis.
i s 0"..te 00 moi ~.a'.".Or ~f res:Cnses '-- two ci recti 0ns is ilio i~;e";aole to
. e 3 ~.1#f.
300R ORG NAL m, m 7910310[(Q-
s Cnce the support mocifications are complete, the affec.tec piping stresses, a. acned ecuipment loads, and support cesigns will all be in accorcance witn FSAR criteria and accep' aole to the staff.
UE&C nas reanalyzec category 1 (pressure councary) anc category 2 (otner safety relatec) lines using Square Root of Sum of Squares (SRSS) load combination instead of algecraic summation.
This analysis employeo a conservative factor cf 2 to convert from CBE to DBE. When the use of SRSS methocology for a 23 analysis was questionec cecause it dio not conform to FSAR commitment, :ne licensee appliec a conservative factor of 1.38 to the SRSS results to convert to aosolute sums, whicn is an acceptable loac combination methoc wi a a 2D analysis.
In such cases, crecit for.conservetism in the OBE/DBE relaticnsnip was taken (a more realistic f actor of 1.2 was usec instead of 2).
Lnen this exercise was completec, one of the first 39 reanalyzec lines was fcunc to exceed total allowaole stress oy 2%, out was still less : nan 0.9 Sy as permittec Dy tne FSAR and was founc acceptaole.
For the remaining 111 unreanaly:ec lines, SRSS stresses were estimated from tne algecraic summation stresses oy applying a fr, tor of 1.5.
The SRSS results were :nen converted to absolute sums for use wi:n tne 2D analysis cy applying a factor of 1.36.
Crecit was again taken for une conservatism in
- ne CBE/DEE relationship. Wnen this exercise was completec, 39 accitional lines were founc suspect. SRSS stresses were computec for tnese lines wnicn eliminatec tne f actor of 1.5.
However, several of these lings still cave stresses in excess of coce allowable. To resolve this proolem, tne licensee recomputec the total stress using coincicent point values insteac of maximum values.
Inis method is more realistic anc is acceptacle -* :ne staff.
The new total stresses were all within code allowable.
Tne above procecure whicn the licensee took in completinc :ne piping seismic stress reanalysis uncer IE Bulletin 79-07 is acceptacle to tne staff.
We finc the licensee's responses concerning IaE Sulletin 79-02 ana 79-04 acceptacle. The reanalysis nas no effect on BSEP pipe creak criteria committed to in the FSAR.
The staff stil1 has some concern as to whether the reanalyses effort reflects the true as-built conditions in the plant.
However, CP&L nas completed a walk-cown of the Diping ano supports insice ne cry well to verify.that the as-cuilt ccncition nas inceec Deen utilirec. Acci:icnally, ne licensee nas commi ttec to a 'eic verification of all lines en oo:n units by June 15, i979.
as finc :nis commitment acceptacie.
=-n
,e ' ice see.as '.cica:ec.na: :ne follcaini computer :-: grams were Jsec in
~. i -ear.a'f i 5 O f
.n i s '. i n*. :
" 5!
- P00R OR8 NAL
s P!SYS - General Electric Company ACLD1?E United Engineers and Constructors GE nas statec that the coce o!SYS combines the responses cue to seismic multi-axial excitation oy absolute summation; the mocal response due to eacn ' excitation are comoined by methods as specifiec in Regulatory Guide 1.92.
A Fortran listing of the cynamic response calculations section of PISYS nas seen suomittec Dy GE anc these statements nave seen certifiec and confirmeo.
GE is also presently solving a set of NRC generatec bencnmark proolems as part of the coce verification effort.
DEC nas incicatec tnat the coce ADLPIP-11 comoines the esponses due to seismic excitation Dy.the SRSS method when used with the response spectra technicue.
Inis nas oeen confirmec cy examining tne coce listing and by verifying :ne code my solving a set of cenchmark problems.
In accition to satisfying the coce verification requirements, :ne licensee nas also agreed to provice tne NRC two problems for confirmatory analysis.
These confirmatory problems will ce solved incepencent!j sy consultants to the NRC at Srocknaven National Laboratory. The mocels suomitted for these piping proolems will De confirmec oy the licensee ;s corresponcing to the "as-cuilt" concition. We find these commitments acceptaole.
CONCLeSION dasec on the discussion and evaluation presented above, we concluce that the requirements set forth in laE Bulletin 79-07 are acequately satisfied to allow resumption of operation upon completion of the mocifications icentifiec in oce meeting of June 4,1979.
P00R ORSM k
1i3l 362
.