ML20072V314

Responds to NRC RAI Re TS Change Request 93-19-0 Submitted by Util to Increase Spent Fuel Storage Capacity. Revised Pages for Holtec SAR & Response to NRC Structural Audit Questions Encl.Response to Subj Questions Withheld

ML20072V314
Person / Time
Site:	Limerick
Issue date:	09/01/1994
From:	Hunger G PECO ENERGY CO., (FORMERLY PHILADELPHIA ELECTRIC
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
Shared Package
ML19353C508	List: ML20072V314
References
NUDOCS 9409190305
Download: ML20072V314 (15)

Text

ct thn cupport D pirtm:nt A

T[

PECO ENERGY

• w e c ~ e"v Nuclear Group Headquarters 965 Chesterbrook Boulevard Wayne, PA 19087-5691 i

September 1,1994 Docket Nos. 50-352 50-353 j

License Nos. NPF-39 NPF-85 U.S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555

Subject:

Umerick Generating Station, Unit 1 and 2 Response to Request for Additional Information Concerning Technical Specifications Change Request No. 93-19-0 Gentlemen:

This letter is being submitted in response to an NRC request for additional information concerning Umerick Generating Station (LGS), Units 1 and 2, Technical Specifications (TS)

Change Request No.93-194. PECO Energy Company submitted TS Change Request No.93-194 by letter dated January 14,1994, requesting that the TS (Appendix A) of Operating License Nos. NPF-39 and NPF-85 for LGS, Units 1 and 2, respectively, be amended to increase the spent fuel storage capacity in each Spent Fuel Pool (SFP) from 2040 fuel assemblies to 4117 fuel assemblies. This proposed TS change is necessary to facilitate implementation of a plant modification to install new high density spent fuel storage racks in each SFP at LGS.

By letter July 14,1994, we responded to an earlier NRC request for additional information conceming this proposed TS change. In our response, we addressed 26 questions relative to the SFP reracking effort which involved issues pertaining to Plant Systems, Structural Engineering, and Radiation Protection.

Subsequently, on August 3-4,1994, the NRC conducted an audit of the structural design aspects of the existing LGS SFPs and the effects of the proposed reracking activities. As a result of this i

structural audit, the NRC identified nine (9) additional questions requiring a response. Prior to completion of the structural audit, we provided Information in response to three (3) of the nine l

(9) questions (i.e., Questions 1,5, and 9) which satisfactorily addressed the issue identified.

However, the NRC requested that we respond to the remaining questions by September 2,1994, in order for the NRC to continue its review of this proposed TS change. Therefore Attachment 1 to this letter provides our response to the questions identified during the audit. Each of the questions, including the three (3) previously resolved during the audit, is restated followed by our response. The information contained in Attachment 1 of this letter is of a proprietary nature and should be withheld from public disclosure in accordance with the requirements of 10CFR2.790, since this information is considered " confidential commercial information." An affidavit attesting to the confidentiality of the information is contained at the end of Attachment 1.

9409190305 940901 l

PDR ADOCK 05000352 P

PDR g

I l

September 1,1994 Page 2 in addition, Attachment 2 to this letter contains revised pages to Holtec's Safety Analysis Report originally submitted by letter dated January 14,1994. These revised pages are being resubmkted to correct typographical /edkorial errors identified on several pages d Holtec's Safety Analysis Report. Also provided in Attachment 3, is a Technical Brief paper published by the Joumal of Pressure Vessel Technology entitled, " Design Strength d Primary Structural Weids in Free. Standing Structures

• The NRC requested a copy of this Technical Brief paper for review since k was referenced in Holtec's Safety Analysis Report (i.e., Reference 6.7.1).

This information in this letter is being submitted under affirmation, and the required affidavt is enclosed.

i

)

if you have any questions or require additional information, please do not hesitate to contact us.

Very truly yours.

e G. A. Hunger, J.

Director - Ucensing Attachments Enclosure cc:

T. T. Martin, Administrator, USNRC, Region I, (w/ attachments, enclosure)

N. S. Perry, USNRC Senior Resident inspector LGS (w/ attachments, enclosure) l l

R. R. Janati, PA Bureau of Radiation Protection, (w/ attachments, enclosure)

I i

l l

1 l

l

[

l

}.

2 COMMONWEALTH OF PENNSYLVANIA ss.

i COUNTY OF CHESTER 1

W. H. Smith,111, being first duly swom, deposes and'says: ~

~2 i

i That he is Vice President of PECO Energy Company; the Applicant herein; that he has read the foregoing response to the request for additional information conceming Technical Specifications Change Request No.93-194 for Umerick Generating Station, Units 1 and 2, Facility Operating Ucense Nos. NPF and NPF-85, to increase the spent fuel storage capacity from 2040 fuel assemblies to 4117 fuel assemblies, and knows the contents thereof; and that the statements and matters set forth therein are true and correct to the best of his knowledge, information, and belief.

i

/

i u

Vice President l

~

Subscribed and swom to i

Sb before me th!s 6 I day 1

of 1994.

/

//

N 4

Notaiy Public l

Erica A Santori,NotaryPLklic Myco k

995 9

-en eeme.-,,

(

v

wr

-r4

-ee-

.=.

M r

. i t

zi i

ATTACHMENT 2 l

Umerick Generating Station l

Units 1 and 2 Technical Specifications Change Request No. 93-19-0 Spent Fuel Pool Rerack' Revised Pages for Holtec's Safety.-

Analysis Report e

I l

1 l

i 6

de e.

l

._.e.

l

a 7.3 Overhead Platform Droo The overhead platform (1900 lb. dry weight) has four legs which are inserted into four empty 4

cells. The platform may be carried over stored fuel as-long as the distance between the top of the rack and the bottom of any leg does not exceed 18". The insertion of the platform into the pool, or removal of the platform from the spent fuel pool, can only be carried out over an empty

)

rack.

Calculations were carried out to assess the effect of dropping the platform during its handling

! in the fuel pool. It is assumed that the platform lifted 36" (or less) from its normal storage elevation during its movement in the fuel pool. At its point of maximum elevation, the bottom of the platform legs will be 18" (or less) above the top of the rack. Postulating an uncontrolled lowering of the platform from the maximum height is found to lead to local cell deformation in the top region of the rack. The plastically deformed region, however, is less than 4" in length (height) and the active fuel region (equipped with the Boral neutron attenuator) is unaffected.

7.4 Conclusion The postulated handling accidents for the Limerick Generating Station have been considered to determine whether the proposed racks meet the essential criteria of subcriticality and structural ruggedness. The subcriticality criterion requires that the center-to-center spacing and other design basis parameters in the active fuel region of the racks are not altered due to a postulated fuel assembly drop accident. Analyses conclude that, under both " shallow" and " deep drop" scenarios, the stored spent fuel array remains subcritical. These conclusions are obtained considering the " heavy fuel" in the drop simulation which is considerably heavier than a typical BWR fuel assembly. Analyses also show that the shallow fuel drop scenario bounds the consequences of a postulated platform drop from 18" above the top of the rack.

6 7-3

l l

e

• k l

I Table 6.8.2

.(

COMPARISON OF BOUNDING CALCULATED AND CODE ALLOWABLE LOADS / STRESSES AT D.?ACT LOCATIONS AND AT WELDS FOR S A ACT FUEL LOADING l

VALUES ITEM / LOCATION CALCULATED ALLOWABLE Fuel assembly / cell wall 1412 2587 impact, Ibs.

-i l

Rack / baseplate weld, psi 13030 29820 j

j Pedestal / baseplate weld

.574 1.0

~

l (dimensionless limit load ratio)

Cell / cell welds, Ibs.

2824 7906 1

1'

~

b 4

e w

i Table 6.7.2 i

SUMMARY

OF WORST RESULTS FROM 62 RUNS OF SINGLE RACK ANALYSIS

^

FOR HOLTEC RACKS IN LIMERICK POOLS OF. UNITS 1 AND 2 LOADED WITH 700# REGUL d FUEL ASSEMBLIES; Excitation Loadings:

(SSE-2)x1.1 OR SSL=(SSE2+5RV3+ LOC 3)il.1 OR (OBE-2)x1.1.

3:-

Item value

.Run I.D.

1. Maximum total vertical pedestal load: 630,869 lbs. dc2sseis.rf8

2. Maximum vertical load in any single pedestal:

266,191 lbs. da3ssei2.rf8

3. Maximum shear load in any single pedestal:

196,735 -lbs. dc2sslil.rf8 3

4. Maximum fuel assembly-to-cell wall impact load at one local position:

1,040 lbs. da3sseo2.re2

5. Maximum rack-to-wall i

impact load at baseplat level:

0 lbs.

6. Maximum rack-to-wall

[

impact load at the top of rack:

0 lbs.

7. Maximum rack-to-rack impact load at baseplat level:

1650.9 lbs. da3ssio2.rf2

8. Maximum rack-to-rack impact load at the top of rack:

113.9 lbs. da3ssio2.rf2

9. Maximum corner displacements i

Top corner in x direction:

1.0482 in, dc2sslis.rf2 in y direction:

1.0149 in.-

dc2sseis.rf8

(

Baseplate corner in x direction:

0.9445 in.

dc2sslis.rf2 l

in y direction:

0.6037 in.

dc2sseis.rf2 t

10. Maximum stress factors Above baseplate:

0.423 (R6) da3ssei2.rfB Support pedestals:

0.650 (R6) dc2sseil.rf8 "h

~

l l~

i l

l r

~

Table 6.7.51

~

SUMMARY

RESULTS OF 3-D SINGLE RACK ANALYSIS FOR RACK MODULE: RACK-A3,

~

Holtec Run I.D.: da3sseol.rf8 Seismic Loading: SSE-SET-2 Fuel Assembly I.D.

and Weight:

Chan'd 700.0 (lbs.)

1 Fuel Leading: 252 cells loaded; Fuel centroid X,Y:

.O,

.0 (in.)

Coefficient of friction at the bottom of support pedestal: 0.8 DYNAMIC IMPACT LOADS (1bs.)

\\

(1) Maximum total vertical pedestal load:

394079.7 f

(2) Maximum vertical load in any single pedestal:

-135359.2 (3) Maximum shear load in any single pedestal:

84824.2 (4) Maximum fuel-cell impact at one local position:

861.3 (5) Maximum rack-to-wall-impact at baseplate:

.0

.0 j.

(6) Maximum rack-to-wall impact at rack top:

.0 (7) Maximum rack-to-rack impact at baseplate:

(8) Maximum rack-to-rack impact at rack top:

65.1 MAXIMUM CORNER DISPLACEMENTS-(in.)

Location:

X-direction Y-direction Top corner:

.1609

.1399 Baseplate corner:

.0166

.0233 l'

MAXIMUM STRESS FACTORS

• Stress factor:-

R1 R2 R3 R4 R5 R6 R7 i

l Above baseplate:

.060

.093

.097-

.089

.154-

.174

.065 Support pedestal:

.173

.119

.126

.092

.244

.268

.162

• See Section 6.5.2.3 of the Licensing Report for definitions.

~_

f l

l

[:

l l

~

-~

l l

Table 6.7.52 RACK-A3

SUMMARY

RESULTS OF 3-D SINGLE RACK ANALYSIS FOR RACK MODULE:

Holtec Run I.D.: da3sslol.rf8 Seismic Loading: SSE2SRV3 LOC 3

~{~_

[

Fuel Assembly I.D.

and Weight:

Chan'd 700.0 (lbs.)

Fuel Loading: 252 cells loaded; Fuel centroid X,Y:

.0,

.0 (in.)

Coefficient of friction at the bottom of support pedestal: 0.8 1

DYNAMIC IMPACT LOADS (lbs.)

407029.9 (1) Maximum total vertical pedestal load:

l l

133835.8 l

(2) Maximum vertical load in any single pedestal:

84036.0 (3) Maximum shear load in any single pedestal:

(4) Maximum fuel-cell impact at one local position:

860.3

.0 (5) Maximum rack-to-wall impact at baseplate:

.0 l

(6) Maximum rack-to-wall impact at rack top:

.0 (7) Maximum rack-to-rack impact at baseplate:

74.8 (8) Maximum rack-to-rack impact at rack top:

MAXIMUM CORNER DISPLACEMENTS (in.)

Location:

X-direction Y-direction Top corner:

.1628

.1379 Baseplate corner:

.0164

.0190 MAXIMUM STRESS FACTORS

• Stress factor:

R1 R2 R3 R4 R5 R6 R7 Above baseplate:

.065

.100

.098

.087

.155

.175

.080 Support pedestal:

.170

.128

.105

.099

.271

.298

.135

• See Section 6.5.2.3 of the Licensing Report for definitions.

e I

e

t l

..c.

l l

l l

ATTACHMENT 3 l

l Umerick Generating Station Units 1 and 2 Technical Specifications Change Request No. 93-19-0 Spent Fuel Pool Rerack Technical Brief Paper Entitled, " Design Strength of Primary Structural Welds in Free-Standing Structures" l

[

.]

I 1

_ i

- 9 i

l e

l r

~ TechnicM Brief i

j

~~

~;. \\

Design Strength of Primary Structural Welds in Free. replacement racks without great expense, disruption in plant operation, and a ceruin concern for safety of plant personnel.

Standing Struernres These considerations led the industry to adopt the so-cr.lled

" free. standing" racks. These racks are essentially cellular i

structures with four or more support legs. Figure I shows the j

schematic of a rack module used in the Diabio Canyon Nuclear t

Power Plant of Pacific Gas and Elecric Company. This rack, j

K. P. Singh,' A. I. Soler,' and S. Bhattactarya like all others, is designed to withstand seismic excitations i

stipulated for the power plant. The mathematical models and I

requirements of the analysis for this rack (and others) are A rational analysts technique to evaluate structuralintegrity governed by USNRC guidelines (1). However, USNRC reg-ofprimary welds infree-standing struc: urn in accorcosce with ulations are intended to provide a general framework, not a the ASME Code is pre:ented. *lizis paper is intended tofill the specific code for design. Spe:ific cite:ia and design require-

\\

1 void in the ASME Code rules for analy:ing welds under met:ts are invoked by the USNRC documents by referring to l

" faulted" (level D) conditions in nonlinear free-standing other established sources, such as AISC Standards (21, and structurci components used in safery-related applic:tions in Subsection NF of the ASME Code, Section 111 (3). For struc.

tural analysis and citeria, the analyst typically prefers the nuc!cer power plants.

latter, which provides a rather complete set of citetia for structural qualifica: ion. Sucsecion "NF" has separate tules for three classes of structures, which are refe red to as cla.ss Nomenciature 1, class 2 and cbss 3, respectively. The fuci racks have been d = distance of neutral axis from centerh.ne designated as a class 3 strucure.

(E E. D Rules for class 3 strucures in subsecion "NF" of the cede F,, F,, F, = applied force loadings (Fig. 3) have provided a reasonably complete basis of structural qual-1 M,, M, = applied moment loadmss (Fig. 3)

Ification of racks. However, gaps in the code tu!cs exist, pre-Ms = limit moment mean radius of Ah weld ring r4 =

1 t, = mean width (tsdial thickness) of ith wc!d ring x-- ~v- - -

rp = force facer of sa.fety f f ","_, % ",% ~_~_',~,~,~/.-(,.

P.j ru = moment factor of sa.fety e = hybrid factor of safety e- -

a" --~ ~ ~ -- w '..,,t,

^ - - "

r l

r, = shear facor of safety f__.__

,~

1. 1. #""#~^-"' -~' " "" JE p ' fl ca.ie

~

A 38' 8 = angular orientation of neutral axis with re-j e.-~~~~~~

p g

spe= to weld h.oe.

e_

__v _._

,7,.,'ip,p q

f (1

1

.,,,,,,, ! p Uj

< l l

a, = " maximum stress" for ith we!d patch l

,r, n n o n n ti l

n

/

0 j

I j

' [9 l

l i.

l

1 Background

With perhaps the sole exception of spent fuel storage racks, l ll j

l i {-

l l

all equipment, components and appurtenances in nuclear power i

plants are firmly anchored to appropriate foundations The i

j j

l 1

i t

i early generation of fuel storage racks were also anchored struc.

{iI

[

j tures. However, in the late 1970s U.S. nuclear power plants i

faced the prospec of indefinite ensite storage, and moved to ij )I l$

i I

^;

retrofit existing fuel pools with racks which maxi =l: d the r

I

[ g quantity of available storage. Existing fuel pool slabs could O

i l

f,

, f not be reconfigured to provide foundation anchors for the new i

)W l

j'

'Honec laternauonal Chc7y Hilt. NJ C800s-1666.

l h'

g[',

8 Penc Cu and Cee:.nc Company. San Fra.ncuco. CA.

2 J.._ _

__:.s_ _ _ _.

Contnbuicd by the P surt Vcueis and Pipsns Division, et Txx Aussacan Y

soc-fy or MrexAnscAL facenzas. Manu.sc tyt received by the PVP Division.

septemoer !.1989 revued manuscnot retzived March :: 1991.

Fig.1 Scfiematic ci olacio Canyon ract rnodule AUGUST 1991, Vci.1131 &1 Journal of Pressure Vessel Technology

itRitcoL asts t%8 r1 rnun ru I

.W' pn assu tait Fu'T g

v r

IC2FWI

)

{

m stFruc irtaid 1

g, sp r,

mm

/

1 s

I l

/

i

- f..

l

/

~

r Pg. 3 Appnad loadings at the support legtbase plate interface Ag. 2 Typical raet support geometry formation are acceptable. Subse. ion "NF" of the Code pro-sumably because "NF" rules [3] are centemplated for anchored ddes explicit stress limiu on pnmary stresses under level D strucures, not for free-standing strucures. The design c-iteria " conditions for linear membes." but fails to provide specific

~

of "NF" (3), which we will hereinafter refer to as the " code " rules for treating the welds which join such linear members to do not envisage short duration high amplitude loadings typical other portions of the strucure. The weld joining the support of free-standing strucures undergoing rocking and sliding dur. legs to the base plate in Fig. 2. for example, is such a weld, ing a seismic event. The "NF" suess limits are focused on which we will refer to as " primary welds." Recalling that the analyses performed by the response spectrum (4), or by static core require:nent of the Code is that a total couapse of the analysis techniques. Such analysis techniques are fundamen. structure does not occur, even as permanent deformation is tally inappropriate to analyze free standing nonlinear strue. pennissible, it is possible to develop an NF-consistent proce-tures. Aside from the nonlinearity introduced by lack of dure for analyzing pnmary welds subje= to leve! D conditions.

anchored conne= ions between the rack and the foundation. The purpose of this paper is to present the concepts evolved similar lack of fixity between the fuel assemblies and storage to develop such a proccdure, and to aniculate the concepts ceu makes the fuel rack strucure a highly nonlinear one. The e.

which underlie the proposed medodology. Much ef the taa-fore, the seismic analysis of racks, of necessity, must be per. terial presented herein was developed during authors' studies formed using one of the various time history integration of the Diablo Canyon high-density racks in the period pre-techniques [6] which accurately captures hign and low har. ceding the Atomic Safety Lic= sing Board hearings in 1986-monics of the responses and provides the anaJyst with a com. 1987.

plete time domain profile of the impac and impulse loads with all their sharp peaks and deep valleys.

2 Weld Configuration A large portion of the peak load stems from "impac" effecs and, as such, are strictly of very short duration. A stress anal.

A typical geometry of the primary structural welds in a spent ysis com:nensurate with the dynamic analysis would call for fuel storage rack is illustrated in Fig. 2. An internally threaded treating these impulsive leads using wave propagation theories. suppon spindle is attached tn the baseplate (Fig.1) of the rack For class 3 structures, the Code (7] circumvents this necessity module through groove and fillet welds. Figure 2 shows the by posing stress limits on " primary" stresses only. In other weld configuration utilized in the Diablo Canyon racks which,

~

words, the Code [7] limits the strucural integrity assessment due to high seismicity of the c=stral California coastal region.

~

to ensuring that gross strucural collapse will not occur. High required large cross sccion wc!ds. Considerations of warpage locr.! stresses, defined in the manner of the classical ASME of the threads in the suppon spindle due to the heat of welding Code for unfired pressure vessels [8], are considered accom-prompt the designers, where possible, to dispense with the modated by local plastic flows. In othe-words, when evalu-internal groove welds, and We the size of other welds as ating stress limit cc:npliance, the stress field in the serien well. However. the fi::al weld configuration must provide suf-under consideration is assumed to be fuUy developed and stress ficient strucural connecivity to transfer the steady-state and gradients due to St. Venant effe=s or stress concentrations are dynamic reaction loads betwe=n the suppen legs and the body negleced. These fully developed se= ion stresses, denoted as of the rack strue:ure. During the seismic event, these reacdon

" primary stresses" are required to meet specific limits in direc loads develop at the suppon leg / liner interface and are present tension shear and flexure.

only when the support pedestalis in contac with the fuel pool The foregoing concept for class 3 NF stru=ures (such as the floor, spent fuel racks) establishes a clest basis for assessment of The reaction loads are primarDy horizontal shear due to structural integrity under design basis loads. The crite-ia, how-Coulomb fricion between the support and the underbase struc-ever, do not carry through in respe to " primary" structural ture (pool liner for racks) and venical compression force due welds, where funher a=;11fication of the cesign bases is re-to dead load and the venicn! motion c f the rack leg (including quired. However,in addition to the design basis loadings, there impacts). These forces are equilibrated by five reac. ion load-are other operationaj leads which also require strucural eval.

ings at the weld plane (rig. 2). In Fig. 3 they are denoted as E

ustion. These are refe red to level A (normal), level B (upset).

E,, F, E,, M, and M. The oojen of this paper is to present 7

level C (emergency), and leve! D (faulted) conditions in the a method to evaluate the design strength of the weld connenion Code (4]. Level D, the so-called faulted condition loacing, and to introduce the related concepts of "facor of safety" 1Q arises from postulated events of extremely low probability, for different categories of seismic events. Details are presented such as the safe shutdown earthquake. A nucicar structure is here for pedestals of circular pianform: the same design phi-n.erely required to be safe from catastrophic failure or collapse losophy can be applied to pedestal weld planes of noncircular under level D loadings; extensive strucural yielding and de-cross sections.

472 i Vol.112. AUGUST 1991 Transactions of the ASME W

I I

!able ! Maxit' sum tinsion and shear stresses Ier austenitic

" C'('1I S ' % '

ut t>v ainless steel at 150*F mas riu.n sta C c2cott stu value Itcierence ASME I'

Ouanutv

[ksi)

Coce taole f Utumste strengin (ksi) 68.1 Tacle 13 *.

Max 2 mum tension stress (ksi) 42 Taole NF-33

4.5faH and AppencLa F r

,-"-- - It'

• Ftat s

in parual penetration groove

! MaJumum shear stress (ksi) 23.6 Table NF.3523(bkl l

, wc!d "e

l Maximum tension or compression

]

~l f

and Appendix F str:ss m fillet weics

!!.6 i

) Design Strength i The notion of design stre::gth of welds has a direc parallel

'n de reinforced concrete literature [81. In essence, the load-carrying capability of the structure is calculated by postuladng 5 hat the " maximum stress"levelis reached in the entire load-Fig. 4 Equavaient pianar weed nngs for Fig.2 structure baring region and that the stress distribution satisfies force sd moment equilibrium. In this manner, the design strength uNB.,, must deal with maximum soment and shear of a secion are computed. To determine "O*E

"*"*3 P*# '".bsen:en wruch, volve combination of direc and Code compliance, the analyst is required to incesse the in-

" stress intensities, m

dividualloading comoonents by prescribed factors to octain spear strmes. It is felt dat takmg the minimum charactedstic the facered load on the scucure. For example, the multiplier cmension, such as throat of the fillet wc!d, rather than its on the op2ating basis earthquake is 1.25, and that on the safe 32d8 dimension, to dettne the weld cross secion introduces tuonal (des,rable) conservatism in the evaluation of the Ishutdown earthquake is 1.0 [81. The secion moments and add.

t

! shears due to the factored load combinuions are compared to design strength. We would, therefore, re=ain consistent with their respecive design strength values.

d' E'U".isi ns of the "NF" and treat shear and direc loads i

I Similar load comoinations are prescribed by the USNRC for * *E *T***2Y '

"NF" structures as well (81. It is, therefore, logical to develop a concept of design strength of a primary weld secion along 4 Formulation for Circular Sections I

the lines of the parauel concept for reinforced concree struc-Before setting down the governing equations to evaluate the design strength. it is nec=sary to define the mathematical mode!

' tures.

Although the reinforced concrete design code prescribes the for ce weld connecion. Consistent with the guideli::es of the ultimate strength of the re-bars as the " maximum stress," it ASME Codes, the weld connecion is simulated by planar dngs

! need not be so for we!d plane design strength evaluation. The of widds equal to the minimum charactedstic dimension of

. logical and conservative value is the allowable Code stress the weld cross secion. The weld planar rings for the ciretlar corresponding to de so-called level D (faulted) condition of support joint of Fig. 2 are illustrated in Fig. 4. The u:twelded the ASME Codes. For example, the maximum stress values centac surface between the support leg and the baseplate is consistent with "NF" for a common austenitic stainless ste:! assumed to carry only compressive leads, and de " maxi =um" material-SA240-304L-are presented in Table 1.

compressive stress on this interface is set equal to the tension To compute the design scength of the weld connecion. the

" maximum scess" is assumed to develop throughout the weid and compression stress value of the adjoining groove weld, A comment on the potential of metaj-to-metal contact at plane, although its sign may be tensile or compressive to satisfy the inte-f ace between the baseplate and supportleg is warranted moment and force equilibrium. The resuhing stress distribution is of recangular shape, akin to the fully developed plastic at mis point. Fillet or groove weids of the type shown in Fig.

stress ficid in an ciasde-perfec Jy plastic material. The scess 2 require addition of filler material during welding.The distribution is charactericed by a neutral axis. The location of age of the weld puddle during cooling of the we!d produces a the neucal axis, defined by its offset from the axis of symmetry tight metal-to-metal contac between the two welded parts.

and its angular orientation, depends on the applied direc loacs Such contact cannot be assured. for example, if the join and the two applied bending moments. However for a given were done by elecric resistance weld. However, elecric re-i axial thrust, the maximum resultant moment Mcwhich would sistance, or another type of confiller matedal weid, is not equilibrate the scess distribution corresponding to the design pracical for this application. One can,in general, usu a metal-to-metal contact at the support leg-base plate interface strengd can be computed. Thus, a thrust-versus moment "in-e.ists, unless the geomecy of the welding detailis unusual, te-action curve" can be generated. nis interaction curve quan-indicating the possibility of lack of such contact.

tifies the design strenge of the weld plane. It is a derived The welds in Fig. 2 are shown in the we!d plane in Fig. 4 physical property for the specified weld plane geometry and Radii r, extend to the center of each weld annuius. The width can be viewed as a limit which must be sansfied by the applied of the weld nngs is taken equalto the" throat" of de respecive I

toads.

The foregoing discussion pet ains to the interacion of direc filled and groove we!ds.

j To compute the design scength of a we!d secien in a support and bending loads. Consistent with the spirit of "NF." the leg it is necessary to specify an axial compression F: and 3.

sher design suength is rested separately and no interacdon compute the associated limit moment M. The wc!d rings are i

t analysis is done. The shear design strength is =erely the gross shear which can be carr:ed by the weld plane if the maximum modeled as thin rings of dickness t.; and rnean radii r,(F 7

shear stress is assumed to develop throughout the weld plane 5). Radii ri, et, and r3 denote the mean radii of tne inner groove.

Y ;I secdon. This pracdce of separate consiceration of direc and the outer groove, and the fil!ct weid rings (Fig. 4), respec The fourtn element is the spindle to baseplate interface (mean shear stresses is the standard praedce for class 3 components radius r ). Element 4 is ineffecive in tension, but can carrv in Senion III of the Code. On the other hand, equipment desigt.ed to higher classes of the ASME Code, such as class I the compression load.

AUGUST 1c91, Vol.113 I 473 Journal of pressure Vessel Technology j

I AP7 tits react us Table 2 Weld data for the eaample problem

{'*

80'M '

s,m_ :st as**

Mammum M tt :2m

}

W CMt1L!st,

f charactenstie Shear

+.

/

dimension or

' EHee:ive area

-4 Weld line enroat (in.)

radius fin.)

(ini or 0.44 3.J3 9.25

/

s-2 E.nenor 0.44 4.29 11.91

\\

CDNTAC 2tNe

/[\\

stoove i

CENTUList

//

i J E terior 0.442 4.71 13.ca f

l / / q,.

i

.i j

i. i \\ \\

7 i;

i.co.,.

D Y

sai

'. : i OCr3 020 cit g\\

/., '

TEL2 CryrtaL!5I

\\

//

2.co

/,

i

/

8 J

)

!!t f7 * *,8 f"I't eCTEI rlLtI 7 ELD cDit1t!Nt ~

. ?

2.00.O

~

}

f' Ag. s Weed plane section and no,, tral ans

\\

A w

Figure 5 shows the center!ines of the four rings in the weld - y f-- /Y --------- '

plane, and location of the neutral axis. Force equilibrium yte!ds y

i.co g (Fig. 5) s 3

/

E. = { [(r - 23 )r,t,aj - [ (1-28,)ar,t, g*

3 c

,.,e 0.00

,..g.

o,,3...

or 8"

3 THRCST X 10t ib.

F: = 8 [ 8/,a.t - ( r 2B )a.r.t.

(1)

Rs. s Thmsumoment inimenon curv.

i 4

..i where Figure 6 shows the thrus:/ moment interacion curve ob-e, = sin ~ ' d; l=1,2,3,4 (2) tained for this geome:.ry by solving Eas (1)-(3). The ac:ual load point corresponding to the foregoing data is denoted as For a given F: d, the location of the neutral axis. can be point A in Fig. 6. The fae:or of safe:y can be defined in three computed using the/7ue nonlinear relations expressed by Eqs. distine ways:

(1) and (2).

(d m ment fac r f safety, ru; rati of limit = ment to 1,

Moment Equilibrium. Taking moments about the cen. applied moment for a given compression lead, t.e., EC/EA i

troidal axis, we have after some algebra, m Fig. 6;

~

y, sk (b) force factor-of-safery, rf: ratio of limit thrust to acual a, dt, cos #, i L,a, rl t, ces 8,= M'-

thrust for a given applied moment:i.e. FD/FA in Fig. 6:

(c) hybrid factor of-safety, r,; ratio of OB to OA in Fig, 6.

i or 3

I a T a, d t, cos 8,-2a, ri 14 cos 8,= Mt (3) In numerical computations, it is most convenient to camp,ute i"

rw since Mr. is casily compared with the problem input (Mi -

Mif'* where M, M: are the reacion moments about :wo ces e,=cos (sin-'

i orthogonal axes with origin at the center of the pedes:al.

where The load point must lie inside of the interac: ion curve for For a series of values of F:, the corresponding limit moment level D leading condition. The Code prescribes a facer of 2.0 M can be computed using Eq. (3) and a force / limit mornent for austenitic stainless material (Appendix F of Secion !!!)

t interac: ion curve for the semion can be generated. This is best be: ween stresses due to level D and level C loadings. Therefore, illustrated by considering a numerical example.

for level B or C condition (operating basis earthquake), at least For illus: ration purposes we use the design and load data one of:he foregoing saft:y facers must exceed 2.0 in order for the Diablo Canyon fuel racks. Table 2 gives the weld to account for the lower limits that apply to these load levels.

geomery. The following instantaneous peak reacion loads For the example problem, the ratio OB/OA = r,is equal to were obtained from time history analysis of the racks under a 1.92.

faulted condition load case:

The shear load-carrpng capability is readily calculated sep-arately by multiplying the maximum shear stress by the effe:.

Resultant lateral shear *

• • 6 kips tive weid ring areas. The shear fac:or of safe:y is simply the

~ ~ -

Axial thrust. F,:

290 kips rano of this limit shear load and the acual(applied) ne: shear Resultant bending moment 129.s kip-in.

load. Referring to Table 2, the total weld plane area is 34.2s (venorial sum of two sq. in. This leads to a shear design strength fae:or of safe:y r, orthogonal moments)

= (34.24)(23.6)/126 = 4.33.

474 i Vol.113. AUGUST 1991 Transactions of the ASME 4

g a-

-4 7-y

w.,, - -,

+

w

-t

5 Closure References a cnmcs. AJt "oT Pondon for Renew and Acceounce of Soest Fud A computational procedure for evaluating the design strength storage and Hanen:Anmns." usNRC. w sseion. D.C.193.

u of primary strue: ural welds hu been proposed. The method 2 Maausi of Seeev Conursenon. American Insutute of Stret Construcuen.

of analysis is consistent with the spirit of the ASME Code wo. tn 19so.

~.i...

nales for component support strucnues, and bridges the dis.

3 AsME Boiler and Preisure Venat Code.Section t t!. 5ubsecuen N F. ASME.

connect in the Code rules when they are applied to geomet.

Ne* Ya's.1983.1986.

j

~

4 ASME Baler and Pmsure Venei Code. Section it!. Subec:2cn NCA.

y91515 presented. free-standing) structures. Although the. anal-rically noniinear(

ASME. New Yort. 1983, 1986.

i' in the context of annular wefd patches.11 Can 3 Sinsa. K. P., and Soter. A. I.Weaanent ocnga of Neer Ercasarrrs be directly extended to other shapes.

sad prerrwe Vener Compoaums. Chao. :1. Arcturus Punusmers. Chmy Hill.

NJ 1984 6 Soner. A. ! and Sinth. K. P "Sastmc Response of Free Sanding Fud RxJE Conarucuons to 3-D Motsons." eWclear Engmernar and Centa. Vol.

I 80.1984. pp. 313-319.

7 Bernacia.M. D " Design C:ncria for Boilers and Pmsure Venets in trie USA." Procredmss of the Fifth (marnanoaat Coafereace on P estwe Yesset Acknowledgment Tecanosoty. Bajing China.1988.

8 ASME Bauer ana Pmsure Vessel C:de.Section V!!! Div.1. ASME.New The authors are thankful to Mr. Hans Ashar of USNRC Yort.19:6.

for hiS input during this work effort.

9 "NUREC 0800. Secuan 3.s.4." USNRC. Wasnington, D.C 1980.

l i

4 l

l l

ee.

w I

l