ML19282C782

Forwards Response to NRC 781221 Comments Re SAR for Shipment of S2W/S2Wa Spent Core Cartridge Shipping Container

ML19282C782
Person / Time
Site:	07109785
Issue date:	04/05/1979
From:	Brodsky R ENERGY, DEPT. OF
To:	Casey Smith NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS)
References
G-6262, NUDOCS 7905020264
Download: ML19282C782 (12)

Text

• /

~

jfG.'j%%

6 Wh

'%a >

yj - 97CSf Department of Energy 7

i Washington, D.C. 20545 7^v 4

it NR:Rit:LMWissel G#62G2 Dr. C.

V.

Smith Division of Fuel Cycle and Material Safety Nuclear Regulatory Commission Washington, D.

C.

20555 Thru:

Dr. W.

E.

Mott, Director, Environmental Control Technology, DOE S2W/S2Wa REACTOR SERVICING - NUCLEAR REGULATORY COMMISSION COMMENTS ON SAFETY ANALYSIS REPORT FOR SHIPPING SPENT CORE CARTRIDGE; RESOLUTION OF Nuclear Regulatory Commission (NRC) memorandum FCTR: RHO 71-9785 dated December 21, 1978 provided NRC comments on Revision 3 of the Safety Analysis Report.for Packaging (SARP) for ship-ment of the S2W/S2Wa Spent Co7e Cartridge Shipping Container.

The Naval Reactors resolution of these comments is provided in the Enclosure to this memorandum.

The Enclosure is in the form of point-by-point resolution of the NRC comments rather than in the form of revised pages for the SARP.

Cr

.,\\(ayhv l

g.

R.

S.

BRODSKY Division of Naval Reactors

Enclosure:

Naval Reactors Resolution to Nuclear Regulatory Commission Comments on Revision 3 of the S2W/

S2Wa Spent Core Cartridge Shipping Container Safety Analysis Report for rackaging.

1 Copy to:

D.

A. Nussbaumer, NRC gj C.

E. MacDonald, MRC (w/ Enclosure)

1. ~~

f Dr.

W.

E. Mott, ECT, DOE O.

R.

Rieley, Bettis THIS DOCUMENT CONTAINS '[ w y,-

P00R QUAUTY PAGES

'Q;g "

"'a_ N' [

o,,

7s05026;2U/

4

-;x

..,: s ; :.

x XY,E: b. w:

. ~,_2

e t

I C. _ ;yi1 % N.',*

[

h d 9 ' n !,

llenolution of IllC Commento dated December 21, 17/8 on the S2W/S2Wa SAftp The following presents the Ni!C comment followed immediately by the resolution of the comment.

Comment 1 Nonductile Fracture

~

Show that the carbon steel components (impact limiter, chield container and its cover, including velds and bolts) are adequately designed against nonductile failure under the hypothetical accident 30-foot free drop and 40-inch puncture loadings at lov service temperature environments.

The present evaluation of the packaging for nonductile fracture failure is incomplete.

Ilesolution The shipping shield cask is made of MIL-S-15083 cl. cW cast carbon steel. The impact limiter and chipping nhield closure are mde of ASTM A516 Gr. 55 and 70 structural steels for intercediate and lov temperature applications. The shipping chield cask closure bolts are made of high stren6th alloy steel.

'Ihe A516 Gr. 55 steel has a 50-ft-lb Charry V-Notch transition temperature of -60*F for a 21/2-inch plate.

Thin mterial vill definitely behave as a ductile metal at -20*F vhich is the lowest temperature to be considered, according to the proposed NIC llegulatory Guide 7.8.

Th e A516 G r. 70 h a s a 50 f t-lb. Charpy V-Nutch transition temperature of -30*F in the 1

transverac direction for a 4.0-inch plate. The A516 Gr. 70 plates used range from 1.0 to 3.0 inches thick. The thinner plates are expected to have lover transition temperature than the 4.0-inch plate. MIL-S-15083 C1. cw is a rather broad range specification.

Accordirg to steel mnufacture, the nominal carbon content used in the cast carbon steel is 0.22/3. With this carbon content the Charpy V-Notch fracture energy, interpolated fron published data on norr.altzed cast carbon steels of 0.16 and 0 30/, carbon, is estimted to be 20 ft-lbs at -20'F.

'lhe structural design specification for the Naval Reactor Program calls for a minimum Charpy V-Notch value of 20 ft-lbs for the low carbon cteel.

'Iherefore, nonductile failure is not expected The chipping s'hield g

cask and closure bolts have, according to the nanufacture, a CVN of 25 f t-lba at -20*F; and, therefore, no brittle fracture is expected to occur. By above reasoning, a complete brittle i

fracture evaluation of those carbon steel components is not necessary.

page 2 L-Comment 2 30-Foot Flat Top and Top Oblique Drops - Core Cartridge Holddown Studo llevice the stud stability evaluation to show that the applied loado do not exceed either cruch ultimate load limit or inelastic (clactic if stress ic leco than ' tc yield value) buckling load limit. According to the analycic chovn in pagcc 199a - 199c of SAll, the studa would begin to yield at about 18g (axial load = 56 kipc) and buckle at 32g (axial load = 117 3 kips). Which are much less than the predicted g-loads.

In the inelactic buckling analycis, the eccentric loading effect, which in particularly important in an inelastic cace chould be considered. In the control rod withdrawal dictance analycic (pp 1.103 - 1.10'(), ceran chaft buckling analysis chould be made by taking into account of additional load impoced on the chaf t due to holddown plate /

P&S container top cover contn. cts either due to excessive cruching or buckling of the core cartridge holddown studo.

(It chould be noted that if the sleeve does not buckle, it can cuotain strecc r21ch higher than the static yield stress (or limit load) due to material strain hardening, ctrain rate and muchrooning effects. Hence, the conclusion in the applica-tion of no chaft buckling in a post-yield state based on the fact that the critical inelastic chaf t buckling load in greater than the cleeve yield (limit) lond is not justified). The dynamic model uced in finding dynamic a plification factoro is not realistic because the core cartridge and ceram chaft mances are mostly trancnitted to the impact limiter through top cover acccmbliec of RLS and chieldi:g containers rr.ther than through the bottom ends of these containers.

Resolution Additional evaluationc of the core cartridge holddown studo for the cases of the top and top corner drops have been made.

'Ihece evaluations, which included dynw.ic amplification, the effect of strain hardening, strain rate and raichrooming, eccentric load application and the presence of the new energy abcorber,

!g chov the following-Flat Top Drop Top Corner Drop Peak Stud Cruch Ioad 63,2500 88,!>oo Critical Inelactic Buckling Load 11t9,2500 l!49,1500 g

9 The mode of deformation is determined b;. the lever of these two values and hence, cruching rather than buckling occuro.

In both orientations, crushing stops when the core cartridge

l' age 3 lif ting logo, guide tubca and nerem charta begin to bear lond, limiting crunh of studo to.298 to.922 inchen. Since the stud la 19 2 inchen long, the ct. rain is 4.8/> vhich in within the ultimate strain of 5 0/, and failure does not occur under the worst case clearance condition. The identified crush depth may result in rod movement, but the total rod movement is less than the six inches found acceptable in the criticality evaluation of Chapter 5 The ceram nhnft/holddown sleeve has been reevaluated considering the presence of the new energy absorber in the flat top drop vith the following results:

Flat Top Top Corner Applied Capability Applied Capability Load Load (lbc)

(lbs)

(1bs)

(lbs)

Ifolddown Sleeve 6lho 18,160 13,340 18,160 Scram Shaft 6140 110,420 13,3h0 110,420 Scram Shaft Critical 25,680 15,930 Buckling Load Since in all casec the applied load is less than capability, there is no inelastic deformation. Since there is no deformation, the valuco thown above are based on clastic r;tterial properties.

Thece evaluations use dynamic amplifications as cuamarized belev:

Flat Top Top Corner IIolddown Studs 1.22 1.22 Scram Shaft 2.17 2.17 The dynamic model chown for the flat top drop in the SARP is realistic for the steel energy absorber (i.e., withcut the new energy absorber). During the flat top drop, the energy absorber impacts the unyielding curface and beginc to cruch. The lond in transmitted through the chipping shield cover to the cover of

~

the IGS container and to the core cartridge through the holddown studa and eventually the rod holddevn plate, litting lugs and funnelc. There is no force trancmitted through the bottom of either container to the core cartridge becauce there is no bottom attachment between the chipping chield and the PAS container or between the IMS container and the core cartridge. The nprings included in the model represent the cruching memberc and the masses are the masses directly supported by the corrcoponding crunhing member. Thus, the dynamic model is a realictic one for the case of the steel energy absorber. Dynarite naplificatl.a factora of 2.17 between the energy absorber and the scran chaft and 1.22 between the energy absorber and the core cartridge holddown studs are chown for the top drop with this model.

The amplification factor used for the top corner drop is obtained by considering that the lateral and axial components of package

l' age 16 deceleration are independently cubjected to the interal and nxial amplification factoro obtained from the SHOCK code analynin of the nide and flat top dropa recpectively, and recombining the resultant componento. 'ihin appronch 10 valid since there are independent force trannmicolon Imths in the two directionn. On thic bacio, and since the aide drop dynamic amplification 10 essentially J

1.00 (i.e. 0 93), the same valuen of dynacic amplification are used for the +op corner drop as were used for flat top drop.

No explicit dynamic calculationa vere mde considering the new energy abcorber.

It will reduce the amplification becauce of itu low natural frequency (fm -

1

[^ ' n 1.8 cps ).

2 Tr y a Therefore, uce of the amplification factor calculated for the steel energy abcorber applied to the g-load set by the new

~

energy absorber results in a conservative 38 x 2.17 = 82 which is cubstantially less than the scran chaft capability of 281 g's for the flat top drop.

Comment 3 3_0-Foot Stable Bottom Oblique Drop The analysic should be revised or supported to concider the following:

Revice the g-load analysin to include the effect of a.

additional work energy contributed by the lead cruching forces.

b.

Denonstrate that the g-load value of 360s used in I:AS container chcIl ntress analynis adequately taken :are of possible dynamic aaplification effects by taking.nto accoint effects of flexibility of ctructural elen ents of the system.

Include effects or lead crushing foraes and core cartridge contact forces at the nupport choe assemblies.

Resolution The inclusion of the contribut * *>n of lead entshing fr rces and consideration of poscible dynamic amplification effects has resulted in minor changes to the calculated g-loadin; s on the structures and crunh depthc. Tbc. dynamic amplificatt on factor fc; the bottom corner drop in found to be approxinately

..O uning the octhod dencribed in the recolution of Comment 2 and Ge lateral cr -l bottom end axial amplification identified in the res >1ution of Comments 2 and 1, respectively. The follovirg table conpares the 5

renults obtained concidering the contribution of the 'ead crushira forcen with the exinting issue of the DAW.

Shipping Shield C ruch Depth, inchen Drop Orientation Revised SARP

__ G -load lieviced SAllp

~ ~ '

Stable Bottom Corner (46.9*)

2.h5 2.23 229 212 60* oblique h.oo 3.83 218 191 75" Oblique 3.60 3.h2 2A 218 The reviced crush depthe continue to be lenc than 5.lh inches, the vabte uned in the chic 1 ding, analysin which choved acceptable reculto.

The evaluations of R&S chcIl at.recccc in the region of the mmrort nhne nnsemh11en are rhovn startinc on pn. 1.175M (for

Page 5 the bottom corner drop) and thov ctrecoca of 68,200 pai which

[

are high enough to enune local yielding, but not great enough to cause rupture. Thene SARP atreso values are based on a g-load of 356 (rather than 360 g's cited in the NIC co =nent which renults in 260 g's lateral). Since the maxirx: revloed g-load in 25I4 g's (215 g's in the lateral direction) and includes the effects of lead 4

cruching forcen, the strena values chown are conservative. The conclualon that local yic1 ding of the shell accura, but not rupture, in valid, f

Comment c Appropriate measuren chould be taken to limit the IWS container vall strences to that belev the material ultimate strength (Su) at temperature. Note that the current standards on design criteria are contained in Reg. Guide 7.6 (alco Appendix F of ASME Code III) which limit strena intensity of primary membrane stress not to exceed 0 7 Su and 2.15 S where S design strena

=

intencity value given in Appendix 109, AS!.1E Code III; strecs intensity aricing from primary membrane streco and primary bending strences chould not exceed Su and 3 6 S=.

In obtaining the strecs intensity values, in addition to longitudinal and chear strecocc, other streco componento (e.g., circumferential strecces on page 1.02TT) chould alco be combined.

RecoJution The following table chova a typical comparicon of actual ctress and allovable evaluated by the existing SARP criteria and evaluated by Reg. Guide '(.6 criteria:

SARP S_trecc Allowable Shear 39100 ps! (35800 psi)*

38600i poi Principal 77600 pai 81100 poi Reg. Guide 7.6 Primary Membrane 17800 pai 15000 pain 4

Primary Membrane +

75100 poi 64200 pa iw x

• Dending

• peak value at outermost fiber; (t verage value through the vall thickness no.'( Su

• NkSu Shece values are for the stress in the pas container chell im.ediatel,y adjacent to the lead cupy rt platea during a bottom corner drop. These ntreca values have been adjusted to include the effect of ciretmferential ctreca dtw to the moment on the lead support plate, and are in excess of the SARP criterion in

pnge 6 a localized area.

It should be noted, however, that thene

?

etrecaea only exist over 3% or the thell circumference and are nelf limiting (by deformation or tearcut of the lead support platen) and, therefore, cannot recult in a cheur rupture. Secondly, at locations between the lead cupport plates the primary membrane plun bending streca intencity value in 17,800 pai and thic meetc the allovable

{

value of C4,200 p31. The average chear strecc is lean than fl the SARp allovable value.

l c

Comment d Une of lead flow ctreca of 3,200 rot chould be juctified noting that the static ultimate tennile strength of chemical lead may be no lov as 600 poi at 250*F.

11ecolution The S2W/S2Wa las container uses pig lead confoming to Federal Specifiestion QQ-C-171 Grade C requirementc. This in exactly the come kind of pig lead which was specially tected by Southwest Ilecearch Institute in 1977 at varicus compreccive otrain rates and at different temperatures. The use of 3,100 poi for dynanic flow stress for lead was chocen for the estimated at, rain rate j

of 12 in/in/ cec and at a temperature of 250*F.

Unen maxir.:un deceleration of the IGS container occurs, the lead was clumped approximately three inchen and localized strain exceedo 0 5 in/lu. The 12 in/in/ cec strain rate in based on an average velocity of 263 5 in/sec (i.e., one-half the impact velocity) lead decelerrition and a deformation dictance of 217/16 inchec (after which the lead in assumed free of the support platea).

Thic given a strain rate of 263.5/217/16 = '2.29 in/in/ cec.

A variat, ion in ctrain rate by a factor of two reculta in approximately a 200 pai chance in the flov strecc which in turn resulta in about a 1.4 g ch 2nge in deceleration.

5'igure 6, attached, has been extracted from tne above listed report and dicplays the eencitivity of flow strecs ao a function of a otrain and strain rate.

Coment e y

In the ceram chaf t deceleration capability analycia, juctify neglection of the portion of holddown plate impact loading that is trancmitted through ceram chaft; also include therml streco effects in the evaluation unleco they are chovn to be incignificant.

llecolution The impact loading of the control rod holddown plate vao not included in the neram chaf t deceleration capability analyclo contained in the SARp. A revised evaluation has been made including the plate loading imposed on the ends of the ceram chai'tc, and the capability is 300 g's for stable corner imIcet which exceeds the impact g-load factors for the oblique bottom corner drop accidents (i.e., 218 g's for 60*, 25h g's for 77, 229 g's for the ntable condition (46.9 )).

= 21 7/ o inchen l

6 f, -- lead ctrain over h2 7/8 inch length

.q

= 0 5 infin.

a v -- average velccity

= 527 2/2 = ~ 265 in/sec.

l, 5

~

E,-- ctrain rate i

= v/6

= 12 36 cec-1

..t i,,

• C -- le d. fl v stress jj 250*F

/

Pb m7

.a..

..y^

3100 "o*

=

r 4

_.Gy t

-, p-

_ El

+

_s

,,t. p n

g y

L_

E

',,t-tl

.. ;-t-4 /,

,tn

. r*

/. '. ;

n

,1--

i,

cn f

M T I'

.f / i M

i,,

.T IIT t,,

ie i

h v

1 3

U

,l!

+ --

?

_.[-

d.

l l,;r.

i m

r-r.p.

__ti

',i.

w;7

.-- - " ; [

i

' -U r-/--'y A

l

-t-b m

a g

-~}

. e..1 ". - -

o

4 mi

,,.T' 1

-..-.a..

.LH

,2

._(a_ip 2

! e, yg

_ __1__ 4_ _, _r.

__u _, _.. i AL, a.,

..____..__,_q_..

u j _.L_ t 4 ' 4.

g _l; ; '.

t

.,_..t i

I

.....__4

' i _L.-. L

. _, _G __ _i.-,...

.. +.

.l

.___.,--.__...p,gI,..

4 4-.

,' __.7_7. 4 1 y a j

-.R.....

..H 7

'.M.

.__._._o

.-_..,21

,u,

'_A_._

-q -.. a.....

__._..g_..

,7+4

' _. Hy.

4_.

A__ u.

-_y

.;+

.y.

~,.

1

, e-O r- - ~ 4

,1

,. :t

^'

,7-*?**

?!7:

, d. -'+ ?

'-+ T" i

i,

,, t*

i 6

• • T i

t?

i ttt i,i *~t t - - *h

+-+-r-+-+-~.

n t-

-t t h i

. -+ t i i

r r

L L- ;-....,

.._ L-. L._ L..,,..

.. i. y.. {. y.. '... _ 5 __... t..t-

• f.f *

'- '--- -- k __.,E - *

• t i + *t!*!'--M-t'~kf4{'

--t--

r-4 ' 4 * *

• -* l '

.-.A_..

.-.-.w.-

t

._ _ L___ L L _... l.

-l1,I

',i'

-* i "

._____7_...._.

.7

_._.'-*t t !

t-+-+'

t-t---- r-~[ t-t

- t-' - r-i

.Tt;

'i*-

,--t-**-**

m----

t n-t-r -r-t t.

i

• t-

-3

-2

-1 1

2 3

4C 10 10 10 1

10 10 10 10

~

• Page reprcduction of test True Strain Rate (sec-1) data given in attachment to KAPL Letter "CGN 855h2-ll, dated May 18, 1977" blGURE 6.

KAPL LEAD AT 250*F 4

Page 8 Germal strences within the 43 56 inch long,1.248 directer section i,

of the nerum chaft are negligible cince the therr.al coefficient of f

expansion of the certta chaft c.aterial (LT h I}{, 10-6 x 6.4 in/in/*F) in 50rf crn11er than the core cartridge upper assembl,y structure caterial (3h8,10-6 x 9 7 in/in/ *F). On this bacio, the temperature difference between the upper annembly and the neran charto would have to be more than 300*F before the ceram chaf t could be thermlly constrained. A 300*F temperature differential la incredible becauce the tr txinum temperature differential from Ieak fuel temperature to core perimeter in 378'F, based on conservative calculation and including the decay heat generation. The ceran charts and upper assembly are axially displaced frcn the region of decay heot. The large air space around these com;.onenta provideo better heat transfer from then due to convection and radiation than directly opposite the fueled region. Thus, there in a tendency to equalize the teuperatures of the scran chafts and the core cartridge upper asnesbly structure.

Stuce the upper end of the fueled region is lesc than 500*F, the temperature difference between the scran shaf t and core eartridge vill be leca than 300*F.

>n:nt. 4:

30-foot Flat Bottom Drop Tne analysis should cons ider the follusi ng.

a.

R&S contain3r shell g-load and a tresa analysis - Fevise the analysis to include effects of forces ard eor':

i ener6 es contributed by support, shoe housing an! lead crushirg forces and the portion of core cartridm lcads (trarenitted to the F5S container through the support shoe housing assemblies ),

olution

A revised analysis, including effects of for es and work energies contributed by the support ch0e housing,. lead crushirc forces and the portion of the core cartridge loads transmitted to the F13 container through t he support shoe housing asser.blies has t'een performcd.

The c ruah damage to t.he lower end of the container shell increased, due to the additional effec ts, from 0.57 t.o r.75 inch ; but the g-load fac tor was reduced from 637 to 634.

The revised analysis coraidered the additianal forces on the shell du7 to the yielding of the core cart.ridg' t.rar.S i ti on asserbly and lead slunping which increased the ava ilatle - srcy f or crush by 36 g

forcent. Ilowever, when the unbalanned fc: ;es

c. Lirc on the R&S con t a i ne r f rom the lead slump (93,000 pou ads ' ani the ccre car tridge (3.74 rillion pounds ) are coruidered in c"pcr i', ion to the shell crushirg i-fmt ea (l?.44 nillion pounis ),

the g-loai ac tir6 on the shell of the R&3 container is reduced relative to the exis 'ird resul t.

S ince the published stress analysis is t'ared cn the higher g-Icad fac tor the exis tirc results are conservat ive. Subse quent annlyneo in the S ART, e.g. shielding assesrmant., have corz> ide red a crush damage of 0.94 irth which is also conservative relative to m vised c rush damage.

Vav,o ')

i!'

Cor.:nent b Show that the dynamic amplificat. ion on the <!ynamic inpact r esponses of MS container is not, sis;nificant and nay be ignored.

vesol uti an:

Dynamic arplificat. ion could be signi ficant to the resporre of the i

RSS cent, airer during a flat botton drop if there were insufficient separation be tween the natural f rnquencies of the shipping shield base and the R&S contaitur. Howe ver, i t is calculated that the nat. ural frequency of the shippirt shield is nore than a factor of 20 greater than that of the MS contairer chell. Considering the cottom of the chipping chield as a single opring, ningle cuco syaten frequency componento of the inpact pulse below half the natural frequency of the chipping chield are trancrtitted to the 16S container with no attenuation or amplification while components above about 15 times the natural frequency are attenuated. Since the IWS contatner natural frequency in about one twentieth that of the rhipping chield, the frequency components of the impulce near the natural frequency of the 183 container are transmitted through the shipping chield with no amplification or attenuation and it 10 just an if the IRS container impacted the unyielding curface directly.

'lhe load transmittal pat.h for a flat bette-drop is as felle:e:

'lhe bot,Lom cf the shipping shield contac ts the unyielding surface anl decelerates quickly because of the _arge area of solid steel i nvolve d.

'Ihe door of the R&S containe: bears direc tly on the bo tton of the shipping shield and the decelerat ion loads are t.r a ns ni t t ed through it to the shell of the RSS container.

Conside ring the botton of the shippirg shield as a steel dist and deternining weight, and clastic sprirc; constant of it results ir. a weight, of 27,750 pounds and a spring cons tant of 2.04 X 1010 pom.Je per i nch.

This correspords t.o a natu al frequancy of 2680 Hz.

Tne crmhing area of the MS container shell supports a weight of arr -ex,in.ately 47,E00 pounds ar:1 has an elast.ic crrirg cons tant of 6.83 10 pounir per i r, h w i th a corresponding na tu-al frequery:y of abou' 112 he r tz,

Tapa10 1

Cottr e rit c

% Cormat 3.c above.

e "es o l ut. i o n :

'Ihe conbirnd container shell s tresses fo r press ure, thercal and irr pac t, load ing are given in the S/J1P s tar ting on page 1.2070 ard j

when ruvised to a 460 g inpact, loading rotri in the respcme to NRC Corr: cent, 4 A, are less tnan the material ultimate str ergth at te mpera t ure (253 F).

This cemparison i' ra le in the fol lowing tabulation at, the sann locatiors analyzed in the SARP.

S tress Ultima te Strength Shell Location (kni)

( ks il 1.

Adjacent Top Flange 8.82 64.0 2.

Adjace nt. Suppcc e 3i00 5 9.94 64.0 flous ings 3.

Adjacent Inad SupporL 37.13 64.0 Plates To show comp 11arce with the Regula tery Guice 7.6 cri teria at these s a rre loc a tions, the s trees int.ereity valuer vers us allewable s trersths are as follows :

Stres:

I n t e ns i t.7-0.7 Su or Su (ksi)

( ks i) 1.

Adjacent, Top Flange a.

prirr.ary rentrano 3.04 44.8 (0.7 S )

u b.

prir-ary nen.brane plus 8.59 64.0 (S l u

bending 2.

Adjacent Support Shoe Hous i ng a.

pritrary nerbrane 33.0 44.8 (0.7 S Iu b,

pr)n.ary rembrane plus 43.2 64.0 (S l bending u

3.

k!jacent Load Suppe.r t. Plain a.

primary nombrano 32.5 44.S (0.7 S l u

b.

primary r.embrar,e plua 35.6 64.0 (S l u

t be n'!i tr

~

J'ar ' 11

enent d :

Shus that t he It&S contairer shell d:es rol buckle unde r the comb ined axial impac t load ani ra d ia l lead ;ress ure.

Fesolution:

The.,tability of the li&S container shell, under the corbined ax ial impac t load and radial lead pressure hu been assessed by corisidering j

the shel] to be closed at the ends and subjec ted to ex te rnal pressure 3

and axial loadir,s.

These cases aru treated in Flugre's " Handbook of 3

Engirecring !!cchanics", paces 44-3 9 and 44 40, anl the analysis is outlined below.

The eritical axial s tress can be appreximated as 6cc = 0.577 ( t ) E. /r d_ _ - 2r -

2 1

which assumes inelastic properties over the full length rather t,han over the bottom one inch of shell length where plastic crushing is occ uri ng.

This conservat.icm introduces a safet.y factor cf 5 in tne elastic buckling fortgla. p'rThe critical le id prescure can be approxi-rated as(f

= k TT K/l., where pb Et = tangent modulus 3

Et K = flexural rigidity 12(1 - U2) k = from Flugge's I!andbook...

The chell vill be stable if f / T' a

cr pb/ 0'pb 1 vhere f 4 f 4.

in the a

imposed deceleration stress due tc cruching and includes the weight of the head and lead force reacted 5.hrough the nupport plates, and f in the precnure imposed by the lead. Iloark uses this same inter-b action formula in accecoing the stability of truncated conical chells under combined axial load and external precsure and indicates the equation may be used for cylinders. To saticry this equation for a 634 g lupact loading, the lead pressure rnct be less than '(,500 pai and since the dynamic flow otrecs is less than 3100 psi, the R'eS chell vill not buckle.

r I'

t i

k I

i

<a