ML17299B155

Rev 1 to Deficiency Rept Re Weld Failure on Unit 1 Pipe Support Structure Identified During 860313 Surveillance Insp.Caused by Weld Fracture Due to Overloading.Documents to Implement Repair Issued for Six Supports

ML17299B155
Person / Time
Site:	Palo Verde
Issue date:	04/09/1986
From:	Bingham W, Hodge M, Van Brunt E ARIZONA PUBLIC SERVICE CO. (FORMERLY ARIZONA NUCLEAR, BECHTEL GROUP, INC.
To:	Knighton G Office of Nuclear Reactor Regulation
References
ANPP-36021-EEVB, NUDOCS 8604110106
Download: ML17299B155 (18)

Text

4' REQULAT V INFORNATION DISTR,IBUTI0 3YSTEll (R IDS)

ACCESSION NBR: 8604110106 DOC. DATE: 86/04/09 NOTARIZED: NO DOCKET 4 FACIL; STN-50-529 Palo Verde Nuclear Stations Unit 2i Arizona Pub AUTH. MANE AUTHOR AFFILIATION li 05000529 VAN BRUNTP E. E. Ari zona Nuclear Power ProJect ( formerly Ari zona Public Serv BINQHAN> M. Q. Bechtel Corp.

HODGEi N. F. Bechtel Corp.

REC IP. NAl'1E RECIPIENT AFFILIATION KNIQHTON> G. W. PMR Prospect Directorate 7

SUBJECT:

Rev 1 to deficiency rept re weld failure on Unit 1 pipe support structure identified during 860313 surveillance insp. Caused bg weld fracture due to overloading. Documents to implement repair issued for six supports.

DISTRIBUTION CODE: A001D COPIES RECEIVED: LTR ENCL SIZE:

TITLE: OR Submittal: General Distribution NOTES:.Standard.i zed p lant. 05000529 RECIPIENT COPIES RECIPIENT COPIES ID CODE/NANE LTTR ENCL ID CODE/MANE LTTR ENCL PNR-8 *DTS 0 PWR-8 EB 1 i.

PMR-8 PEICSB 2 2 PMR-8 FOB 1 PNR-8 PD7 PD 01 5 5 LICITRA> E 1 PNR-8 PEICSB 1 1 PMR-8 RSB 1 1 INTERNAL: ACRS 09 6 6 *Dl'1/LFNB 0 ELD/HDS3 1 0 /TSCB 1 NRR/ORAS 1 0 REG F L 04 RQN5 1.

EXTERNAL: 2@X EGR<G BRUSQUE S LPDR 03 1 1 NRC PDR 02 NSIC 05 1 1 TOTAL NUMBER OF COPIES REQUIRED: LTTR 30 ENCL 26

I

, 4 ff ff >>

4$

lt, I

>*I>> fl >>f 4

NNI

~ N>>NNk Nfl C" I 4 I'>>'N N Il f) C(

-, '" fl 4>>" If $ IIN

'.IN 4

IIN>>4' I'"

4 I II'>>,."

, if>>,

Arizona Nuclear Power Project P.O. BOX 52034 ~ PHOENIX, ARIZONA 85072-2034 April 9, 1986 ANPP-36021-EEVB/RAB/98.05 Director of Nuclear Reactor Regulation Attention: Mr. George W. Knighton, Prospect Director PWR Project Directorate k/7 Division of Pressurized Water Reactor Licensing - B U. S. Nuclear Regulatory Commission Washington, D.C. 20555

Subject:

Palo Verde Nuclear Generating Station Unit 2 Docket No. STN 50-529 (License NPF<6)

Report on Pipe Support Failure Pile: 86-073-419 P.3.04.1

Dear Mr. Knighton:

Attached, for your information, is a report concerning the single pipe support failure in the Unit 1 containment which was found on March 13, 1986.

This report describes the condition of the failure, the evaluation conducted to determine the root cause and the extent of the problem, and the corrective actions.

The corrective actions have been completed.

Very truly y urs, E. E. Van Brunt, Jr.

Executive Vice President Project Director EEVB/RAB/dk Attachment cc: E. A. Licitra (all w/a)

R. P. Zimmerman A. C. Gehr J. B. Martin Bb0411010b Bb0409 PDR 8.

ADOCK 05000529 0OOI PDR I)II I

0

~ ~

~ >" l e ~

~ ~ P A

P 1

1 a . ll I'

Rev. 1 WELD FAILURE ON PVNGS UNIT 1 PIPE SUPPORT STRUCTURE I. CONDITION DESCRIPTION A surveillance inspection of Unit 1 on March 13, 1986, identified a broken pipe support (Support No. 1-SG-005-H008). The broken support was located inside the containment building and was a strut type support for the 24 inch main feedwater line to Steam Generator No. 2. The break occurred in the 'weld between the flange connection of the upper and, lower support beams. The lower portion of the support separated completely from the upper support beam (see Attachment A). The axes of the two beams were oriented 90',

and it.,was discovered that the two 4 inch long, 1/4 inch leg fillet welds between member A (upper beam) and E (lower beam) fractured.

Attachment A shows'he location of the failure. A distortion of

,t the beam flanges at the weld location occurred.

The time of failure of the support is not known. A visual inspection of the hanger was performed by ANPP in March 1985, prior to Post-Core Hot Function Testing, at which time the support was intact.

II. EVALUATION A. Su~azuZ The evaluation of this condition included both an evaluation of the specific failure of support 1-SG-005-H008 as well as an evaluation of the potential for similar failures of other large and small bore Q-class and Seismic Category IX pipe supports, the overall pipe support design, and the additional support designs for raceways, HVAC, instrumentation, fire protection lines and major Q-class equipment.

B. Metallur ical As ects at Weld Failure Location on Unit 1 The broken support was visually inspected. The upper beam's bottom flange edge showed substantial bending. Similarly, the lower beam's upper flange showed some bending. Part of the fractured weld remained on the upper beam with the remainder on the lower beam.

The weld surfaces on the lower beam were also examined under a Stereo Microscope. Part of the fracture was in the base metal near the fusion zone and part was in weld metal. The weld quality was satisfactory. The fractured surface of the beam had a woody appearance, but the fractured surface in the weld metal appeared smooth.

The fractured surfaces were additionally examined under a Scanning Electron Microscope (SEM). The examination showed elongated dimple structures typical of a ductile overload under shear stresses. The base metal had some equiaxed dimples and

Rev. 1 elongated fibers indicating ductile overload fracture. No evidence of any striation was seen under the SEM, proving that fatigue was not a factor in this fracture.

Chemical analysis of the base material and weld material showed that the materials conform with ASTM A-36 and AWS SFA 5.1 Type E7018 requirements. Hardness tests showed the base material had an approximate tensile strength of 72,000 psi. The filler material tensile strength was approximately 90,000 psi.

Typically this material has elongation, at fracture, of approximately 30X.

Based on the above observations, itDeformation was concluded that the weld of the flanges fractured due to overloading.

indicate that the weld transferred the loads until it could no longer accommodate strains imposed by deformation of the flange, causing'the .weld to fracture.

C. Unit 1 Pi e Su ort Desi n and Loadin Attachment B summarizes the results of loading combinations and design capability of the failed Unit 1 support (1-SG-005-H008).

The originally installed pipe support had a Faulted Maximum Permitted Load of approximately 15 Kips. The actual normal load (dead weight plus thermal) applied to the support was 17.7 Kips, thereby resulting in overloading of the support. The support has been modified to provide for normal and faulted conditions with adequate structural margin as shown in Attachment B.

The pipe support design loads were reviewed and determined to be proper and appropriate loads. Therefore, the design loads used for evaluating the pre-modified and modified design are the same for the three PVNGS Units.

D. Unit 2 Pi e Su ort Desi n and Loadin Investigation of the same support in Unit 2 (Support 2-SG-005-H008), which had experienced hydrotest and precore Hot Functional Testing (HFT) thermal loads as well as dead weight loads, determined no flange deformation or weld damage.

Further examination revealed that the lower beam on that support was not a W6 x 12 but a W6 x 15.5 member. Since the flange is 2 inches wider than the member used in Unit 1, it provides four additional inches of weld (two inches on each side of the beam). The larger beam size substitution was permitted per installation Specification 13-PM-204, since it was an upgrade in beam size.

t

'I Pl t

I

Rev. 1 Results of the calculation of this support (with the larger beam member) is also shown in Attachment B. The increased capability (approximately 21 Kips) of the Unit 2 support to carry the thermal and dead weight load (approximately 17.7 Kips) of the associated piping without any flange deformation is apparent. This support has been modified to provide the structural margin as shown in Attachment B.

E. Review of Other Su orts The failure mechanism of the Unit 1 support was a result of localized flange bending not being adequately considered in the design.

all Therefore, it was decided to review the drawings for Class supports for large bore pipe ( ~ 2-1/2 inches) to Q

determine if other instances existed where local flange bending may have been overlooked.

A review was conducted on 100X of the large bore Q-class pipe supports which involved 3,678 design drawings and "as-built" documentation. Five supports in the Main Steam and Safety Injection systems were determined to be inadequately designed for flange bending and required modification as shown in Section V.A. Attachment C lists the six pre~odified and modified support calculations and indicates the location of the support configurations within these calculations for Unit 2.

A review of design documents was expanded beyond the specific area of the support failure to include:

a. Review of the overall pipe support design and a review of the design of pipe support attachment to plant structures.

The result of this review confirmed the adequacy of the designs.

b. An investigation which sampled small bore Q-Class and seismic category IX pipe supports. The results of this selective investigation indicated the supports had adequate designs.

c. An investigation of the design of other equipment (structural steel, raceway supports, HVAC, instrumentation, fire protection, major Q-Class equipment supports) utilizing configurations that could be subject to localized flange bending failures. The results of this selective investigation indicated all had adequate designs.

These reviews were conducted to ensure that the loss of design control resulting in the failed support did not extend to other areas.

d I

'4

Rev. 1

.III. ROOT CAUSE The root cause of the support weld failure on support 1-SG-005-H008 is improper consideration of localized flange bending loads during the original design of the support. The flange deformation without stiffeners, caused additional bending stresses on the fillet welds and beam flange leading to weld failure.

IV. CONCLUSION the evaluation the limited of Based problems on identified (six), itconducted and is concluded that there number was a loss of design control during the original design of the support but this loss was an isolated case resulting from an oversight on the part of the designer and checker, and not a generic breakdown of the design process.

V. CORRECTIVE ACTIONS A. For the six supports that require rework, the following work documents have been issued:

Unit 1 6 2 Engi- Unit 3 Document neering Document to Implement

~Setem ~Su utt to Im lement Re air Feedwater 13-SG-005-H008 EER 86-SG-056 SFR-3SG-008 Mainsteam 13-SG-036-H011 EER 85-SG-094 Unit 1 SFR-3SG-008 EER 86-SG-056 Unit 2 Mainsteam 13-SG-042-H011 EER 86-SG-056 SFR-3SG-008 Mainsteam 13-SG-042-H017 EER 86-SG-056 SFR-3SG-008 Safety Infection 13-SI-220-H020 EER 86-SG-056 SFR-3RC-089 Safety Injection 13-SI-220-H024 EER 86-SG56 SFR-3RC-089 B. Procedural Modifications The Bechtel Group Supervisor of the Plant Design Stress and Support Group issued a memo on April 1, 1986, to all personnel providing definitive instructions regarding the requirements for stiffeners to be placed in the center of the load path.

W. Bingha Dat'e Project Engineer ng Manager Bechtel Power Corporation M. F Hodge Date Supervisor, Mechanical/ Chemical Engineering

P

~4

PJPE SUPPCRT 1-SG-005-H008 FAILURE LOCAT ION CH%%T FAlLURE OCCLRAED AT THIS STRUCTIJRAL

%B.D ATTAQ4ENT

~EN CCHKCT IOH FLJPCK là LPPER AJC LCNER 54PPCSP Mhk KXlKMIW ~

FLlSCK OKFCAhQ,TIOÃ AT J5KA OF GBW IKLOED CCXhKCTlGS APO LOKI SLPPCRT SEAII ATTAQCdBll'DO CIA%9)

PLATE OTHER SZIK LONER ~RT FLANGE DEFOQJKD SMUT LOAD END VIEN KO REQ ' OESCR I PT IM mS X ZS X 3'-8" Crs X 25 X 3'-6" ST1FF Pl . t/O' 2$ X 0'-5 1/2" 1 I%6 X 12 X 1'-4"

E

~ C

ATTACHMENT B LOAD

SUMMARY

TYPE IOAD (KIP)

I WEIGHT -16.642 I

I THRM &.873/-1.042 I

ISSE %1.839 IE SSE (SAM)l M+.736 I

ILOCA I + M+6.245 I J.I.

I IW.H. +34.950/-6.437 W.H. includes DW + THRM IOAD COMBINATION NORMAL LOAD ~ DW + THRM ~ 0/-17.684k FAULTED LOAD DW + TH + SSE + SAM + WH ~ &9.525k/-21.012k OR FAULTED LOAD ~ DWt.TH+[(SSE+SAM) N(LOCA+JI)2]1/2 +26.698k/-39.509k DESIGN LOAD y 49.5k/ 39.5k RESULTS TABLE PRE-MODI FIED MODIFIED NORMAL i, FAULTED NORMAL FAULTED MAX. ACTUAL MAX. ACTUAL MAX. ACTUAL MAX. ACTUAL I IPERMITTEDI LOAD IPERMZTTEDI LOAD IPERMITTEDI LOAD IPERMITTEDI LOAD UNIT ILOAD(KIP) I (KIP) (LOAD(KIP) t (KIP) (LOAD(KIP) I (KIP) ILOAD(KIP)I (KIP) 7~5 -17. 684 I +53. 4 +49. 5 -41.4 -17. 684 +53.4 +49. 5

-15 -39.5 -43.86 -39.5

-10. 35 -17.684I +53.4 +49.5 -41.4 -17. 684 +53.4 +49.5 I -20.69 -39.5 -43.86 -39. 5

S 8

'E F

H 0

ATTACHMENT C PIPE SUPPORT OLD CONFIGURATION NEW CONFIGURATION (UNIT 2) 13-SI-220-H-024 Problem 8SI517B 3/13/82 (Old Gale) Problem SI517B 3/23/86 (New Gale)

Last Page Last Page (Titled 13-SI-220-H-024 Rev. 2) (Titled Engr. Eval. Request 86-SG-056) 13-SG-042-H-017 Problem SG-501D dated 4/8/82 (Old Calc) Problem SG-501D dated 3/23/86 (New Gale)

Last 2 Pages Last 2 Pages (litled Engr.

(Titled Sheet 2 13-SG-042-H-017 Sheet 1 of 2) of 2 and Pages Eval. Request 06-6G-056, 12 and 13) ~

13-SG-036-H-011 Problem SG-501C dated 11/12/81 (Old Gale) Problem SG-501C dated 3/22/86 (New Gale)

Last Page Last Page (Titled 13-SG-036-H-Oll Rev. 2) (Titled Engr. Eval. Request 86-SG-056, Page 8) 13-SG-042-H-Oll Problem SG-501D dated ll/12/81 (Old Calc) Problem SG-501D dated 3/23/86 (New Calc)

Last Page Last Page (Titled 13-SG-042-H-011 Rev. 4) (Titled Engr. Eval. Request 86-SG-056, Page 14) 13-SI-220-H-020 Problem SI-517B dated 8/20/80 (Old Calc) Problem SI-517B dated 3/23/86 (New Gale)

Last Page FCR 59, 451 P (attached)

(Titled 13-SI-220-H-020 Rev. 1) (Dwg. included in new cele is for Unit 1 only) 13-SG-005-H-008 Problem FW-501A dated 5/21/85 (Old Calc) Problem FW-501A dated 3/23/86 (New Gale)

Last Page Last 2 Pages (Titled 13-SG-005-H-008 Rev. 8) (Titled EER 86-SG-056, Pages 5 and 6)

Iv t

tl I

Attachment C

~ ~ 1 ss or.. I or.. rq, g5(-f IELP CHANG CUE'S'f AIO D*r 'I Jg CCII HO, g DuAI.IT< .y, ~

gag i<O. l O>O7 Cl AIS DATC + 5 g'3 >/i j.

her D'II'G OIT SS'CC

-< i- 4'xo- Qd<Q SSLV TITLC

<Jape@. As~ -SlVcw gC. C. C I OC'Sl Qk ORIGIN; Q gNOIQC EQINP Q tVSISII.ICh (IOCNTIFY bY t4AIILK) T UNIT I uNIT t, VNIT j CO%MON

'A Ic 'H 6~L. 0 ~~@

4scck.=~l-&g

'L&g~ I

. 1 KgI~L

~g~kJO

~. CHANCE REOVEST/SKETC< %f 8 EVI~'R

't ~ '2 '4'i f'- + @

~~V Agi OQ I5 Tl K~ISx~g

~,

~c~c-'o4 lD VI CWCQ lIYI

. f'~

olscltI IHc rICI 0 cttcI (u

ccN

~47.g I I.

F'STSFASSrO IX A F'f'F SVI l4 FI Cl 5

. 7 4Q 5FOSITION gC OISCIiI.WC r ICI.D CSIaIIICCa DATC toostcT FI cLo ctlc)II DA C

+gcgygg rNONEEIIINO g+(~~~All Qpi I O a r OVC 4/4 0 r C ICOSI DATC HVCLC A+ CIIOVS'VrcSI VICOSI )IS'I COD I II col DATC ACCT C

+/~gzDATc FDAC IO-I.ICTCO raI neo cIIIoI.C I.IIIc owotl IOQALITT cLJLts 0 AsID sl sTcc jl SI g LI A lT K S

~ <</aiba tSIIlrT COOSIOIIIATOSII C0S'Ils T0 CI.ICIIT. iVITVrr.DIjCII'I.ISIS, ASID hljIOC~QIQIPC 0 DIST SI IQLST IOkI OSIIOISIAI AC QI T I Ok AL DISTR IbVTIPNI 0 2%0)CCT FSIOCVIICIACIIT MANAOCSI Q COST 'TSI SIIO CIICINCCa P~ lt 2 I II0 loll s+ 2

~ it the A ~ go~ ~ oy

% e ~ n ~ w s p oo ~ ~ ~ q~ ~,, ~

na rn nn no wu s