ML17275A624

Forwards Responses to Structural Engineering Branch Round 2, Set 1 Questions Re Sacrificial Shield Wall

ML17275A624
Person / Time
Site:	Columbia
Issue date:	10/03/1980
From:	Renberger D WASHINGTON PUBLIC POWER SUPPLY SYSTEM
To:	Youngblood B Office of Nuclear Reactor Regulation
References
G-02-80-218, G-2-80-218, NUDOCS 8010090454
Download: ML17275A624 (38)

Text

REGULAT "

INFORHATION DISTRIBUTION YSTEH (RIDS)

ACCESSION NBR:8010090454 DOC,DATE: 80/10/03 NOTARIZED:

YES DOCKET FACILe50"397 NPPSS Nuc,lear Projeqti Unit 2i Washington Public Powe 05000397 AUTH ~ NAME AUTHOR AFFILIATION RENSERGER,D,L,'esbington Public Power Supply System ir,R'/g REC IP ~ NAME RECIP IENT AFFILIATION YOUNGBLOODEB~ J.

L~icensing Branch 1

SUBJECT:

Forwards responses to Structural Enginerring Branch Round 2p Set 1 questions re sacrificial shield wall, DISTRIBUTION CODE:

B001S COPIES RECEIVED:LTR g i ENCL (0g SIZE:, gg, TITLE: PSAR/FSAR AMDTS and Related Correspondence NOTES:PM:

2 copies of all material.

05000397 RECIPIENT ID CODE/NAME ACTION:

A/D LICENSNG RUSHBROOKpM ~

INTERNAL; ACCID EVALl BR26 CHEM ENG BR 08 CORE PERF BR 10 EMERG PREP 22 GEOSCIENCES 14 HYD/GEO BR 15 I 8 E.-

06 L'IC QUAL< BR MECH ENG BR 18 NRC PDR 02

.OP LIC BR PROC/TST REV 20 RA S

BR22 G FILE 01 i G BR25 COPIES LITTR ENCL 0

1 0

1 1

1 1

1 1

1 0

1 1

2 2

3 3

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 RECIPIENT~

ID CODE/NAME YOUNGBLOODiB LYNCHgD~

00 AUX SYS BR 07 CONT SYS BR 09 EFF TR SYS BR12 EQUIP QUAL'- BR13 HUM FACT ENG BR ILC SYS BR 16 LIC GUID BR MATLt ENG BR 17 MPA OELD PO'rJER SYS BR 19 QA BR 21 REAC SYS BR 23 SI T ANAL BR 24 SYS INTERAC" BR COPIES LTTR ENCL.

1 0

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 1

0 1

0 1

1 1

1 1

1 1

1 1

1 EXTERNAL.:

ACRS NSIC 27 05 16 16 1

1 LPDR 03 TOTAL NUMBER OF COPIES REQUIRED:

LTTR ~

ENCI.

~

t.

'I 1

H

%1

'i 1,

Washington Public Power Supply System A JOINT OPERATING AGENCY P, O. BOX 958 3000 GEO, WAEHINOEON WAY RtcHLAND, WAEHtNOTON 99352 PHONE (509) 375.5000 G02-80-218 October 3, 1980 E

Docket No. 50-397 U. S. Nuclear Regulatory Commission Office of Nuclear Reactor Regulation Washington, D.

C.

20555 Attention:

Mr. B. J. Youngblood, Chief Licensing Branch 1

Division of Licensing Gentlemen:

Subject:

WPPSS Nuclear Project No.

2 Responses to Round Two guestions Set One, Structural Engineering Branch (SEB)

Concernin the Sacrificial Shield Wall

Reference:

Letter, NRC,(Tedesco) to WPPSS (Ferguson),

"Structural Review of the Proposed Corrective Weld At the 541 Foot Construction Joint of the WNP-2 Sacrificial Shield Hall", dated September 19, 1980 Enclosed please find sixty (60) copies of the responses to Round

Two, Set One questions from the Structural Engineering Branch (SEB) con-cerning our p'roposed corrective weld at the 541 foot elevation on the WNP-2 Sacrificial Shield Mall.

These responses will be incorporated formally into the FSAR in the next amendment and are identical to the responses given to SEB informally in our October 1st meeting in the form of Burns and Roe Technical Memorandum No. 1204.

We understand that wi,th the formal receipt of these questions with this letter that you are now in a position to write a letter to us on our docket approving our proposed corrective weld.

We request that you expedite your efforts in this regard as we need NRR release on the yGA

I I

i;

BJ Youngblood i Page 2

2-80-218 October 3,

1980 t

proposed weld as soon as possible so as not to impact our restart schedule on the wall.

The Office of Inspection and Enforcement

'Headquarters and Region V) has stated that a copy of the docket letter to them is sufficient documentation to confirm your approval of the design and provide for the release of the technical 'hold'n the corrective weld.

Very truly yours, DLR:OKE:cph cc:

MD Lynch -

NRC (telecopy)

FP Schauer -

NRC (telecopy)

A. Toth -

NRC Regional Inspector DP Haist -

NRC Region V

B. Bedrosian - B&R JJ Verderber - BSR ND Lewis EFSEC, Olympia JR Lewis - BPA WNP-2 Files 1

D. L. Renberger Assistant Director Technology

0 0

I k

~gag

STATE OF WASHINGTON)

Responses to Round Two guestions Set

One,

)

ss Structural Engineering Branch (SEB)

COUNTY OF BENTON

)

Concernging the Sacrificial Shield Wall D. L. RENBERGER, Being first duly sworn, deposes and says:

That he is the Assistant Director, Technology, for the WASHINGTON PUBLIC POWER SUPPLY SYSTEM, the applicant herein; that he is authorized to submit the fore-going on behalf of said applicant; that he has read the foregoing and knows the contents thereof; and believes the same to be true to the best of his knowledge.

DATED

><~~ >

, agsO D. L.

RENBERGER On this day personally appeared before me D. L.

RENBERGER to me known to be the individual who executed the foregoing'instrument and acknowledged that he signed the same as his free act and deed for the uses and purposes therein mentioned.

GIVEN under my hand and seal this + ~

day of

, 1980 or

$ y

~yshi otary Pub c in and f the State of Washington Residing at

<q,a<'L' NP l

/

>,~lJ,. "i

,'//~

'dl<~~ ~, ~~gx'~

Round 2 Set 1

Responses to Structural Engineering Branch (SEB)

Questions 130.45 130.49

WNP-2 Provide typicaL welding details for the column splice platesi the skin platesr and the shim plates at the construction joint in the sacrificial shield waLL (SSW) at elevation 541'-5".

These details should include the size and Length of welds and an indication whether the welds are continuous or intermittent.

Indicate the maximum totaL Loads which are transferred at this joint.

Response

This request concerns welding detai Ls at the interface EL. 541'-5" and the magnitude of Loads transferred at the interface.

The requested information is furnished beLow:

a 0 Typical welding detaiLs for the coLumn spLice plates are shown on BRR Drawing S835 inc luded in Reference 1r Attachment 5.

WeLding detaiLs for the exterior splice plate are shown in the detail entitledi "Typical Exterior Splice Plate at Columns" and consist of the following:

(See Fig. 130.45-1 from Dwg.

S835 for convenience)

Above the interfacer EL. 541'-5" r the splice plate i s welded to the W24 column and the ring channeL by 50.5 inches of 3/4 inch fiLLet weLd.

Below the interfacei the splice plate is welded to the ring box by a complete penetration single bevel weld' inches by 14 inches.

The welding of the interior splice plate is shown in the detail entitledr "Typical Interior Splice Plate at Columns" and consists of the foLLowing:

(See Fig.

130.45-2 from Dwg.

S835 for convenience)

Above the interfacei the splice plate is welded to the W24 column and the ring channel by 32 inches of 3/4 inch fi L let weld and 12 plug welds each 1 7/16 inches in diameter.

Below the interfacer the splice plate is welded too the ring box by 46 inches of 3/4 inch filLet weld and by 4 plug welds each 1 13/16 inches in diameter.

I

~

~

~

WNP-2 b.

The connections of the skin plates to the ring members adjoining the interfacer Elevation 541'r are shown in Section 1602 on BSR Drawing S783 inc luded in Reference 1i Attachment 5.

FiLlet weldsr 3/8 to 1/2 inch in sizei are provided aLL around the periphery of each skin plate.

(See Fig. 130.45-3 from Dwg.

S783 for convenience).

c ~

The shim plates provided at the interface are shown in Leckenby Drawing F124 included in Reference 1i Attachment 4i Last page.

The shim plates were driven in place and were not intended to be welded.

In certain cases they have been welded inadvertant ly to the upper channeL by slot welds.

This has no structural significance-(See Fig. 130.45-4 Dwg.

F124 for convenience).

d.

e.

The maximum horizontaL shear Loads in one 15 degree panel of the waLL are due to a combination of Loads involving annulus pressures and pipe break reactions both due to a feedwater Line break together with seismic and dead and Live loads.

The magnitudes of the controlling panel shear Loads which are transferred across the inter face, joint are 327.0 k i ps tangential along center line of wall and 27.4 kips radiaL along Line of column web.

1 The maximum verticaL Loads transferred across the joint by one set of column spLice plates are due to a combination of loads involving reactions due to pipe break (in dryweL L proper) i seismic'ead'nd Live Loads.

The controlling vertical Loads consist of verticaL force of 307.4 kips along column axisr moment about radiaL axis at column center Line equal to 87.4 inch kipsi and moment about tangentiaL axis at center of wall equal to 8210.2 inch k1ps0

References:

1.

WPPSS reports "Engineering Evaluation of the WNP-2 Sacrificial ShieLd WaLL"i transmitted to NRC on WPPSS Letter G02-80-168'ugust 1i 1980.

P

4l

~

k

~'I k

r

I I

f

~

S o

~

~'

eP.

eCHKM~

V Ci L

5 "L

S ~

r

p vv)P

~P'-H ~IOE OF.

coL. NF5 (TgP )

7

')Q EXC,E.PT A5 r'aT-.D I~

Cl,~

4J '~

m iV

~~ f

~+TYP.

W'ii'4 l"'x El I '- 5" S PL I CE 4 7 Y P.

CEASE,H 5I 'F V'l4 COL.At":e.

~TYP.

CoPE HoLE E'TYP.)

DIAF Hv~d,V. E. (we I~M"=P. SKid TYt.

~ %

IYF.

FOR BH NINGit'NNER SKIV k. - ~/q

~

~

~g'x8"LVIDE. EEAM WING t-(TYP, THIS LEYEL)

I M'TE P, I=AC6 OF

&TKBI KI.54l'-8' 4, - A5 DEVEL.'OHTlhfUOU&

TYP.

'15'- I>'. S. iRA D.'~.

C4 Box BEAM Q9i ONLY

~ P3 SKlH g.;>

SUPPORTlhlG)-p

~'~

GER'3 7~

I'l=XI~

rr rr r

/

rr gC+TYP.I

~

YV'2+

w/4

<rYF.

'I '

N~ F R SKIH TQI

~~I= ST'F E}

E.L '534'-Q~

~ \\

F I30-'t> -3

'4

ltd Igp a

IRC ppl I

'I I R

WNP-2 Q.

130.046 The tangentiaL shear in the ring channeL above and below -the 541'onstruction joint in the SSW is taken up equaLLy by the inner and outer skin plates while your proposed corrective weld is only on the outer skin plate.

Accordinglyi describe the redistribution of Loads at the 541'onstruction joint.

Identify the structuraL elements which resist the Localized bending moment due to this "side-stepping" of the tangentia l shear introduced by your proposed corrective weld at this construction joint.

Explain in detail"how aLL the structural Loads are transmitted to the box ring below this construction joint.

Response

This request is concerned with the method of transmission of tangentia l shear from the ring channeL above the joint interface at Elevation 541'-5" through the corrective weld at the interface and then into the ring box below the interface.

To clarify the transmission of the shear and the redistribution of forces that occurs at the interfacei a

sketch showing the forces acting on the ring channeL and on the ring box for one 15 degree panel of the waLL is enclosed as Figure 130.461.

The redistribution of Loads at Elevation 541'-5" is shown in the sketch.

As shown thereon'ymmetrical tangential Loads from the upper waLL are applied from above equaLLy on the interior and exterior fLanges of the channeL.

These symmetricaL tangentiaL forces on the channeL are equilibrated by a set of forces along the Lower exterior edge of the channeL applied through the corr-ection weld.

This equilibrating set of forces consists of a uniformly distributed Longitudinal force and a distributed radiaL force whose magnitude varies Linerar ly between extreme values at the ends of the weld.

In effects the equilibrating set is statically equivalent to the applied symmetric forces acting in the reverse direction.

II

WNP-2 The preceding redistribution of upper symmetric forces on the channeL into asymmetric Lower forces at the interface occurs in reverse for the ring box.

Thus'he asymmetric forces acting at the top of the box are converted into symmetric shears in the side plates of the box member; these symmetric shears are transmitted to the Lower waLL.

The asymmetric set of forces from the correction weld consist of the same uniform Longitudinal force and varying radiaL force as above but acting in the opposite direction on the upper exterior edge of the box.

The equi librium of the box member is then maintained by uniformly distributed Longitudinal forces of equal magnitude acting on the interior and exterior side plates of the box.

The preceding description and Fig. 130.46-1 show that the channeL and box members above and below the inter-face are each subjected to bending moments about the verticaL axis.

Analysis has been made to determine the resultant stresses in the members; the analysis is done ignoring the continuity of the members beyond the panel in question.

Since the box member has substantiaLLy Larger sectional properties than the channels ca lcuLations were made for the channeL member only.

These calculations together with the associated calculations for the equilibrating forces in the correction weld were transmitted to NRC in Attachment 4 of Reference 1

and are included here as Figure 130.46-2 for convenience.

As indicated in the calculationsi the resultant bending and shear stresses in the members (shear 9.9 KSIr bending 0.8 KSI) i are Low and are weLL within permissable values.

Consistent with the Load transfer path through the joint at EL.541'-5" assumed in the orginal designi where aLL horizonta l shears are resisted by slot weldsr the proposed corrective weld is designed to resist these horizontal shears.

The design margins resu lting from this design are discussed in Attachment 4 to Reference 1.

Ifi alternatelyi one assumes that all the, loads to be transferred through the joint at El. 541'-5" will be jointly resisted by the corrective weld and th,e splice plates'ncreased design margins wiLL result for the corrective weld.

WNP "2

References:

1.

WPPSS reports "Engineering Evaluation of the WNP-2 SacrificiaL Shield Wall".r transmitted to NRC on WPPSS Letter G02-80-168'ugust 1i 1980.

~

o

~

~

k

~ ideal!~

~

~ ~

~

~

~ ~

~

~

~

~

~

~

4 '

~

~

~

~

~

Cl+cl C30rg 4 VJ'0 t4la I 00 Drawing No.

By i> t~recA. I ter Rcr e~PPVI 5-HALF Or?0 alc. No.

Checked i4b, l-o. C,,'Cl (if'Z

~I Book No.

I.. <'f.b

.5v Aoproved 1 L06, SAC R i F >c wu'$

Q gioi i O4 i Ihr itrVI I P /

Page No,

~ Sheet

'o of

'r ~iV )l~f:.-i I>>rory C D Ei 5'.+ I -

5'P HEWSueF.

2 vgbtZ Co'REF CPoU

>0'i 9 LoADi46 t.'opT)gV ITV 'F CHAL)]JEL ic

'LCQoPEO

~ Sus JEcT: ctcescT tot)

ME/teuRR; AT M>RFAcc H, AURAL Roi>>

0 f-Pk'o Poi~E0 i." >RpEc i, c HKi.r cHAPI4EL H% 48%.R e.

~ LOAO ft)r LAYtoT

~ CoaSaRVATiSH' t

C g6.f << "f 'i)

C

)

I I

t CC

~ i LoAoigG HA<a>Tpr Rc q )

C)r'ic CI QeT ~ Tc c-

$. l-4'tcf, = 4'I)$

FoF HAY. N

'S./

I I

CI" Q E:g Qi P(T ltt, X

• 'iS t

Ju

= ~hC X.-.t.54X.

llew% =0

~ Av t

C i

t ~

I I

'I 4 I)S(SC) R. = CI,T)C <iffy. lli,i c roe X 8'3.o'> < +

1F lIC 7 "- y Z

S3o~ g

~~,~

~

1>< i<

. -...~

W%oZ lI O1

<tee, -- fo.t)T (C.CT).C.)C(I..CTj -ilk)f(C.ICT)=

0 I

wC,oL" d.

S~zeA Hpx V =

tI

. 7 a Slog

~

ilg.2.

'L$)r tK'

!Cg g,4 4

- --OIC. ~

~

't Form BR 0002 2

I

WNP-2 Q.

130.047 You indicate in the Enclosure to Attachment 4 of your report that the seismic Loads used in your ana lysis of the partial penetration weld're based on your current definition of seismic Loads and not on the originaL seismic Loads used to design the sacrificial shield waLL.

Provide a complete List of both your originaL and current seismic loads.

Indicate the differences in your assumptions and your methods of analysis for these two sets of seismic Loads.

Response

The current and originaL seismic Loads in the controlling panel used in the analys'es and design of the correction weLd are tabulated below:

Current Original Tangential Shear Radial Shear 13.6k 3.8 41.3k 12.0 The assumptions and the method of analysis used to obtain the original seismic Loads are presented in Sections 3.7.1i 3.7.2.1.1 through 3.7.2.1.8i 3.7.2.3 and 3.7.2.3.of:~the FSAR.

The current seismic Loads are obtained from finite element soi l structure interaction analysis in accordance with NRC Standard Review PLan Section 3.7.2.

V

WNP-2 Q.

130.048 In Technical Nemorandum No. 1188'ou state that the design response spectra at all mass point of your mathematicaL model of the reactor buiLding were generated in your soiL-structure interaction analysis.

Our current Licensing position requires that you analyze the soiL-structure inter-action using both the half-space (Lumped spring-mass or compliance function) methodology and the shear beamed finite element representation of the soi l foundation.

Our concern in this matter is -that the results obtained by the finite element approach using the deconvolution procedure and the FLUSH computer programs may not be acceptable.

Accordinglyr we require that aLL seismic Loads and the corresponding design response spectra be recalculated by employing a half-spacei soiL spring model and that you adopt the 'higher values calculated byeither of the two methods discussed above'esponse:

The originaL seismic analysis utilizes Lumped spring mass and damper representation for soil structure interaction effects.

The soiL spring constants are calculated using elastic half-space theory.

However~

the soil damping values are conservatively assumed to be Low values as presented in Table 371 of the FSAR instead of those calculated from half-space theory as recommended in Section 3.7.2 of the Standard Review PLan.

Consequentlyr the originaL seismic Loads are substantially higher than those which wouLd have been obtained from half-space methodology as suggested in this question.

In connection with the sacrificiaL shield waLL correction weld addressed in the responses to questions 130.45m 46'7'nd 49'nalysis of the correction weld was made usingi in turns the results from each seismic method in the controLLing combination of Loads.

Design margins are calculated which are equal to the ratio of the permissible stress in the correction weld to the maximum stress using the controLLing load combination.

With the current seismic definition based on finite element soiL structure interaction analysis the design margin afforded by the weld is 2.4.

With the seismic Loads as originally defined (Lumped mass-spring modeL) the design margin in the correction weld is 2.1.

Thus'elativeLy high design margins result with both methods of seismic analysis.

This follows because the seismic Loads are relatively smaLL compared to other Loads in the controLLing Load combination~

namelyr annulus pressures and pipe break reactions.

WNP-2 In connection with the structuraL design of the balance of the plants the current regulatory position referred to in this question wiLL be conformed with.

The FSAR wiLL be revised in the future accordingly.

WNP-2 Q.

130.049 In the Last part of Attachment 4i "Analysis and Design of Sacrificial Shield Wallr" you state that the inner and outer skin plates are treated as a single plate having a

thickness equal to the combined thickness of both plates.

Such an assumption is incorrect for out-of-plane bending.

Additionallyi it is not clear in your report what type of plate element and the associated boundary conditions for these plate elementsi are used in your computer modeL.

Indicate whether the plate eLements at the interface are treated differently for the free edges of the as~bbiLt structure.

Response

This request is concerned with the structural modelling of the skin plates as part of the overaLL structuraL modeL used in the analysis and design of the Sacrificia l Shield Wall (SSW).

Information pertinent to this request is furnished beLow:

The paper entitledi "Analysis and Design of Sacrificia l Shield Wall"r part of Attachment 4 in Reference 1i which is referred to in this requests respresents a

condensation of the information on that subject in WPPSS Report No. WPPSS-74-2-R2-B (Reference 2) which was submitted to and approved by NRC in 1975.

As noted in the approved reports the methods of finite element analysis are used to include the effect of the skin plates as part of the framework.

The inner and outer plates in a

paneL are treated as one plate having combined thickness of both plates and joined to the

,framework at the framework paneL joints.

A plane stress type of finite element of rectanguLar or tri-angular shape is utilized in the computer program.

The curvature of the plates is considered in evaluating the permissible stress in the plates.

The membrane stresses (normaL stresses and shear stresses) are funished as computer output for each plate element.

These are used in the design of plate thickness and of attachment welds which are based on Part 1 of the AISC Code.

l A

4

WNP-2 Based on preliminary analysis for the SSWi it had been determined that the principaL structural function to be served by the skin plates was to afford additiona l capacity against in-plane forces carried by the struc-turaL framework made up of the structuraL columns and ring beams.

The structural model of the waLL consists of 378 members and 136 plate elements.

In order not to overcomplicate a

modeL which was already fairly complex't was decided to adopt the conservative approach of assigning all out-of-plane bending capacity to the structuraL members (columns and ring beams) and utilizing a plane stress type of element for the skin plates.

With this models all necessary out-of-plane bending capacity required by the Loading is provided in the design of the structural members and the skin plates transmit membrane forces only.

It was noted above that in the plate element computer models one plate having the combined thickness of the inner and outer plates is used-Howevers the permissible stress in each plate is calculated on the basis of the thickness of one plate only.

As was previously stated the plate elements used in the computer modeL are of the plane stress type.

The plate element boundary conditions used in the computer analysis'hich utilized the computer program STRUDLi derive from the finite element methodology in the program.

Although the plates are physically connected a LL around the peripheryi in the finite element methodologyr the elements are taken to be, joined to each other and the frame work only at the nodes; equiLibrium and compatibility of displacements is enforced at the nodes.

Thus'.,in, the analysis'lL elements have similar boundary conditions in that they are treated as connected only at the nodes.

Howevers in actual designs the difference which applies to the plate elements at the interface relative to the transfer of plate forces through the nodesi is taken into account.

In generalr the skin plates away from the interface are connected to peripheral structuraL memebers which intersect at the nodes so that transfer of plate forces through the structuraL members to the nodes is automatic.

At the interfacer the skin plate forces are transmitted to the ring channeL and the W24 columns which in turn must transfer the Load across the interface.

This is done via the correction weld and the splice plates.

'E

~

P

WNP-2

References:

1.

WPPSS reports "Engineering Evaluation of the WNP-2 Sacrificial Shield WaLL"r transmitted to NRC on Letter G02-80-168'ugust 1i 1980.

2.

WPPSS Report No. WPPSS-74-2-R2-B entitled'Sacrificial Shield Wall Design Supplemental Information" submitted to NRC by Letter G02-75-240'ated August 19'975 and approved by NRC byletter dated October 15'975.