ML20211E280

Nonproprietary Castor V/21 Borated Stainless Steel Basket Evaluation

ML20211E280
Person / Time
Site:	07100510
Issue date:	01/31/1987
From:	GENERAL NUCLEAR SYSTEMS, INC.
To:
Shared Package
ML20211E273	List: ML20211E268 ML20211E280
References
REF-PROJ-M-37 27854, NUDOCS 8702240279
Download: ML20211E280 (110)

Text

g 8 37 l

RETURN To 395_gg I

T I u n, l

pggg {

CASTOR v/21 80RATEo STAINLESS STEEL BASKET

( f U.s.ugg[IS87 klp}

pgg 2

i

~

EVALUATON

\\, %

1. 1. E5ll3p[/

JANUARY 1987 1

L'd:

I e:t;os, E

NON-PROPRIETARY COPY

\\

W lI I

1. cerrco 0

9 USNRc FEg S

O IS cs 3:a - - S7 ), :i';

Submitted by:

Doc rfoff s

%/

M

/

GeneralNuclearSyftems,(nc.

~

220 StoneridgiDrivez Columbia, South Carolina j292102 m

=

?

5 W

I I

RM 2228 P 87 * "

F-K0" REQl3 D M-37

~~

PDR

I I

I 1

I I;

I l

l

~

$drEI" g

. ::sfilg 7

E

/-

/

I lJ;ibp1 e. M g

i e

g l

l l

I l

I I

u

T GENERAL NUCLEAR SYSTEMS,INC.

A Chem-Nuclear Company I

February 5,1987 611-0029-87 I

Mr. Leland C. Rouse, Chief I

Advanced Fuel and Spent Fuel Licensing Branch Division of Fuel Cycle and Material Safety U.S. Nuclear Regulatory Commission Willste Building 791F Eastern Avenue Silver Springs, MD 20910

Reference:

Project M-37

SUBJECT:

CASTOR V/21 BORATED STAINLESS STEEL BASKET EVALUATION This submittal contains the results of the engineering evaluation and testing I

program which culminates our efforts to validate the use of borated stainless steel, in particular nominal 1.03 weight percent Radionox A 18, in the fabrication of the welded basket for the CASTOR V/21 spent fuel storage cask.

This cask has previously been evaluated and approved using an all stainless steel basket employing the current basket design.

The Radionox A 18 basket is

' subject to the same mechanical design, fabrication and inspection controls as the all stainless basket.

This submittal incorporates several changes to our previous submittal based upon our meeting with NRC and Lawrence Livemore National Laboratory personnel in October 1986 All coments from this meeting have3een.. addressed in our report.

These changes involve additional structural analyse.s-reflecting the

, 'as-built basket design and a material testing program which 'was' implemented to establish minimum material properties such as strength, elongation and n

reduction in area and to verify Radionox as a ductile material.

Data from tensile testing was statistically analyzed using the methods delineated in Military Standa rdization Handbook, MIL-HDB K-5C.

The A-basis lower bound values (99% of population with 95% confidence level) were detemined by these statistical analyses ud were then used in establishing the minimum reouf red j

property values speci fled in the material procurement speci fica tion.

This speci fication is also incl uded in this report.

Dynamic tear testing was perfomed to verify material duc tili ty.

This testing was perfomed in accordance with ASTM E-604 over a temperature rance of -60 to 350*C, All specimens tested exhibited fracture surfaces of 'l00% shear, as would be evidenced in austenitic stainless steels.

The results from this testing shows that Radionox A 18 is a ductile material throughout the operating temperature range of the basket.

I

~

As discussed with Mr. John Roberts of your staff, copies of this report will be transmitted to Mr.

Martin Schwartz of Lawrence Livemore Na tional j

Laboratory in parallel to this submittal.

i 220 Stonenc;;e Dr.<e < Como a Soum Cwu ~a 29210 i n,2 % G W.o Tee. m w

I 611-0029-87 Mr. Leland C. Rouse February 5, 1987 Page Two This submittal incorporates the following documents:

Document 1 - CASTOR V/21 Basket Analysis Comments and Responses This is a summary of the responses to each of the comments developed as a result of the February 1986 and October 1986 meetings.

Document 2 - Material Evaluation Report (Revison 1)

The report presents the resul ts of the material tests perfomed to the Radionox A 18 in the base and weld material.

Also included is a discussion of the material strain acceptance criteria and background on the ASME N 47 Code Case.

Document 3 - CASTOR V/21 TSAR Revisions The TSAR calculations and write-up have been revised to reflect the changes from the INEL test basket to the current design in the areas of mechanical I

design, inspection protocols, and Radionox A 18 material specification.

The specific areas of basket structural evaluation include:

I 4

o Normal Handling Section 4. 2.1.4.1 o

Thermal Stress Sec ti on 4. 2.1. 4. 2 o

Accident Handling Section 8. 2.1. 2. 3. 3 Also included:

I o

Material Specification (BS 05)

Appendix 3 i

o Summary of Basket Dimension Appendix 8 l

Measurement Technique i

Document 4 - CASTOR V/21 Demonstration Test at INEL (Revision 2)

This report evaluates the conditions leading to the basket indications I

' discovered following the DOE demonstration testing of the CASTOR V/21 at INEL.

The thermal mechanisms experienced are directly related to the basket joint and ligament strains.

Very truly yours, C

ht4CO Robert T. Anderson Director, Cask & Transportation Systems

~

cmb/0016W f

Encl osures: CASTOR V/21 Borate'd Stainless Steel Basket Evaluation 10 Proprietary Copies 10 Hon-Proprietary Copies I

5 Proprietary Copies c: Partin Schwartz

I AFFIDAVIT SUBMITTED TO THE U.S. NUCLEAR REGULATORY COMMISSION BY GENERAL NUCLEAR SYSTEMS, INC.

CONCERNING CONFIDENTIAL INFORMATION AND TRADE SECRETS I

STATE OF SOUTH CAROLINA)

) SS COUNTY OF RICHLAND

)

I I, Robert T. Anderson, depose and say that I am the Director, Cask and Transportation Systems, of General Nuclear Systems, Inc. (GNSI) duly authorized to make this affidavit, and have reviewed or caused to have reviewed the information which is identified as proprietary and referenced in the paragraph imediately below.

I am submitting this affidavit in conformance with the provisions of 10CFR2.790 of the Commission's regulations for withholding this information from public disclosure.

The information for which proprietary treatment is sought is as follows:

1.

Arpendix to the GNSI Response to NRC Coments on CASTOR V Basket Analysis, January 1987.

2.

Material Evaluation Report for CASTOR V/21 Basket Material, Rev.1, January 1987 3.

Material Specification for Radionox A-18 (X? Cr Ni B 1911) Used for Dry Spent Fuel Storage Cask CASTOR V/21 Fuel Basket.

Speci fication BS 05, Rev. 2, January 1987.

(TSAR Appendix 3).

4 Sumary of Basket Dimension Measurement Technique.

(TSAR Appendix 8).

These documents have been appropriately designated as proprietary.

I have personal knowledge of the criteria and procedures utilize.d by GNSI in designating information as a trade secret or as privileged / confidential information of a comercial or financial nature.

Pursuant to the provisions of paragraph (b) (4) of Section 2.790 of the I

Comission's regulation, the following is furnished for consideration by the Comission in determining whether the information sought to be withheld from public disclosure, included in the above-referenced documents, should be wi thhel d.

1.

The information sought to be withheld from public disclosure consists of design drawings and fabrication details which is owned and held in I

confidence by GNSI, and not disclosed to any third party without first obtaining that party's written agreement to hold the information in con fi dence.

2.

The ownership of this information results in a substantial econonic advantage to GNSI, over its competitors who do not know or use it.

e

3.

The information is of a type customarily held in confidence by GNSI and not customarily disclosed to the public.

GNSI has a rational basis for determining the type of information to be held in confidence.

4 The information is being transmitted to the Commission under the provisions of 10CFR2.790 with the understanding that it is to be received and held in confidence by the Commission.

5.

The information, to the best of my knowledge and belief, is not available in public sources, and any disclosure to third parties has been made pursuant to regulatory provisions or proprietary agreements which provide for continuing the confidentiality of the information and only to those parties who need to know the information.

6 Public disclosure of the information is likely to cause substantial harm to the competitive position of GNSI because development of this information by GNSI required thousands of man-hours of effort and hundreds of thousands of dollars.

To the best of my knowledge and belief, other parties including competitors would have to undergo similar expense in generating equivalent information.

Public I

disclosure of the information would enable a competitor to avoid the effort and expense to develop this information and would enable that competitor to develop a similar product at a significant cost savings, thereby impairing the competitive position of GNSI.

n

/

} / _ /_ /AZ A,

Robert T. Anderson I

Director, Cask 8. Transportation Systems General Nuclear Systems, Inc.

STATE OF SOUTH CAROLINA)

) SS COUNTY OF RICHLAND

)

I On this 5th day of February,1987, before me, a Notary Public in and for the State of South Carolina duly commissioned and sworn, personally appeared I

Robert T. Anderson, to me known to be the Director, dask and Transportation.

Systems for General Nuclear Systems, Inc., and on oath stated that he was authorized to nake this affidavit on behalf of the corporatic.i.

IN WITNESS WHEREOF, I have set my hand and af fixed my of ficial seal the day and year first above written.

OWG40

![$ ikD

, L.S.

t:

Notary Public for0 South Carolina Tamara H. Jef fords My Commission Expires: August 2,1988 4

a

I GNSI RESPONSE TO NRC COPMENTS ON CASTOR V BASKET ANALYSIS JANUARY 1987 I

1 I

I Ref:

(1 )

Letter Roberts, NRC, to Barnhart, GNSI, dated 1?/20/85 (2) NRC/GNSI Meeting, February 24, 1986 (3) NRC/GNSI Meeting, October 7,1986 1

I I

,I lI I

I lI

lI

I Point 1 1(a) NRC Comment (TSAR Themal Analysis)

The documents should show the details of the ANSYS finite element model both with and without gusset plates.

I 1 (a ) GNSI Response I

Figures 4.2-25(a) and 4.2-25(b) of the TSAR are provided which show the node numbers for the finite element model with gusset plates and without gusset plates respectively.

Figures 4.2-26(a) and 4.2-26(b) show the element numbers for the finite element model with gusset plates and without gusset plates respectively.

I 1 (b) NRC Coment (TSAR Thermal Analysis)

For the joints, a description /drawin of the ANSYS finite element model should be included.

If no detailed 2D) analysis was made, discuss how the stresses at the joints were calculated.

1 (b) GNSI Response A discussion of the thermal stress analysis is presented in 56ction 4.2.1.4.2 (page 15) of the TSAR.

I',

I 1 (c ) NRC Coment (TSAR Themal Analysis)

I Discuss for the joints how the beam model - to plate model interface was made.

1 (c ) GNSI Response Although the beam model to plate model interface for the TSAR analysis

~

,I is somewhat different than in the INEL failure analysis, the method of enforcing kinemitic compatibility between one-dimensional (tieam) and two-dimensional (plate) element degrees of freedom remains the same.

Rotational I

stiffeners of beam elements and shear / bending stiffness of plate elements both resist moment loads at a node.

The cubic in-plane I

I I ' noo'8m

I I

displ acement of an edge beam is not compl etely co;npatible with the quadratic isoparametric in-pl ane displacemen t of the plane stress portion of the plate element.

The effect of these incompatibilities is insi gni fi cant.

I 1(d) NRC Comnent (TSAR Thermal Analysis)

I Tabulate and locate stresses for all joints, highest stresses in plates, and highest stresses in ring.

1 (d) GNSI Response Table

4. 2-5(b ) of the TSAR has been devel oped which lis ts these I

stresses.

Figures 4.2-28 and 4.2-29 provide a pictorial representation of the location of these stresses.

I 1 (e) NRC Comment ( TSAR Thermal Analysis)

I Include gusset plate temperature distribution.

1 (e)

GNSI Response The gusset plate temperature distribution, Figure 4.,2-27(a ),

has been incorporated into TSAR Section 4.2.1.4.2, Thermal Stress Analysis.

1(f) NRC Comnent (TSAP Thermal Analysis)

Present and define the acceptance criteria based on, strain.

Show that all calculated stresses meet the acceptance criteria.

Will probably be in the table noted in item (d).

I 1(f)

GNSI Response The strain limit acceptance criterion has been defined in Section 5.0 of l

the Material Evaluation Report.

Tabl e 4.2-5(b ) of the TSAR provides a I

list of the calculated stresses at various axial locations along with the corresponding strain.

The values of strain have been compared with I

E 3

JANUARY ~ 1987 I

(0068W)

I I

the acceptance criterion and in all cases are far less than the criterion values.

The results of this comparison are expressed as a factor of safety which is included in the table.

I 1(g) NRC Coment (INEL Report)

Develop failure analyses and criteria based on strain or displacement I

for J1, J2 and show how failure is related to acceptance criteria.

Acceptance criteria includes ASME code case for low ductility austenitic steel s.

I 1(g) GNSI Response I

The joints J1 and J2 show strains of 45.4% and 31% respectively as determined in Section 5.0 of the INEL Failure Report, Revision 1, August 1986.

This is a factor of 15 to 20 over the acceptance criteria and actually a factor of 2 to 4 over the tested values for ductil ity.

I Hence, failure can be clearly ascribed to the mechanism reported in Section 5.0 of the INEL basket failure report.

These resul ts are summarized below:

Failure Report Strain:

45.4% and 31%

3 Margin 2-4 Specificaticn Required I

Material Ductility:

10 - 15%

) Margin 15-20 Acceptance Criteria Strain:

2%

TSAR Calculated Strain:

less than.04%

s I

I I

I I

I 4

(0068W)

J

I I

1 Point 2 2(a) NRC Coment (TSAR Thermal Analysis)

Present discussion on how the 61, G2 gaps are assured by measurement and note tolerances.

2(a) GNSI Response

,I Appendix 3 to the TSAR has been developed to provide a summary of the measurement techniques used for assuring the G1 and G2 dimensions.

A manufacturing procedure which follows this technique is performed as i

part of the required QA documentation for each basket fabricated.

l i

2(b ) NRC Comment (TSAR Thermal Analysis)

Present data on basket ring and cask cavity diametrical measurements.

2(b) GNSI Response iI l

Appendix 8 of the TSAR summarizes the measurement technf aue and data which are collected on the basket ring and cask cavity diametrical measuremen ts.

l 2(c) NRC Comment (TSAR Thermal Analysis)

Present analysis.to show the effect of non-centered basket.

2(c ) GNSI Response A discussion of the analysis for a non-centered basket is presented in Section 4.2.1.4.2 (Page 23) of the TSAR.

The effect is insignificant.

I I

I I

l I

l (0068W)

I Point 3 3(a) EC Coment (send ASAP - submit to NRC)

The final design drawing detailing how the basket will be built.

Note the weld lengths and location for SN-8, and all welds, gussets, spacers.

3(a ) GNSI Response I

Detailed basket design drawings were submitted by GNSI letter C8603-3, K.

R. Kingsley to J.

P.

Roberts, dated March 25, 1986.

Based on NRC review, certain changes / corrections to those drawings were identified.

Our agreement to incorporate those changes is documented by GHS letter I

C8604-16, K.

R.

Kingsley to J. P. Roberts, dated April 23, 1986.

The final design drawings which are intended to reflect the mechanical design for both the all stainless steel and the borated stainless steel I

baskets have hence been revised to incorporate these changes.

These drawings were submitted by GNSI Letter 611-0124-86, R.

T.

Anderson to J. P. Roberts, Septenber 30, 1986.

The basket design drawing was reviewed and it was verified that there was no inconsistency in the presentation of welds SN20 and SN22.

A note will be added to the drawin g to indicate that all basket quadrants are synnetric.

I 3(b ) NRC Comnent (TSAR General)

Correct Figure 1 in Appendix 1.

I 3(b) GNSI Response I

Figure 1 in Appendix 1 identi fies the conservative dimensions of the calculation model used for perfonning the criticality analysis and is not intended to refl ect detail ed manufacturin g dimensions.

Exact dimensions for structural analysis purposes should be obtained from the I

basket design drawin g A500.11 -15, Revision e.

The following clarification note will be added to Figure l' I

NOTE: This basket sketch is used only for the criticality analysis.

For de tailed dimensions, re fer to basket drawing A500.11-15, 1

Revision e, in Appendix 1 and the Sunmary of Basket Dimension I

Measurement Technique in Appendix 8.

I I

6 JANUARY 1987

Point 4 4(a) NRC Conment (TSAR General)

Include the statement that no welds fail during any TSAR conditions.

4(a ) GNSI Response A statement to this effect has been included in the TSAR Section 4.P.l.4 (Rev. 2A).

I 4(b) NRC Conment (TSAR Dynamic Drop) l Correct the space between welds in TSAR due to error (see SER comments).

4(b) GNSI Response The TSAR (Section 4.2.1.4) has been revised to accurately reflect the design dimensions identified on the drawing for the fuel basket.

I 4(c) NRC Comment (TSAR Dynamic Drop)

Detailed structural analysis of the cask basket.under both normal and I',-

accident design conditions alluded to in Section 4.0 of your preliminary-report should be provided to demonstrate the structural integrity of the basket.

If measures referred to in Point 2 above are taken, it should I

also be made cleat whether this analysis assumes they are implemented.

4(c) GNSI Response Detail ed structural analysis for normal and accident conditions are presented in Sections

4. 2.1. 4.1 and

8. 2.1. 2. 3. 3 respectively.

The I

results of these analyses are summarized in' Tables 4.2-5(a) and 8.2-2.

These analyses demonstrate the acceptability of the basket design based on acceptance criteria for strain as well as with respect to the strain limit criteria doffned by the Material Evaluation Report.

I I

I l

(0068W)

I Foint 5 5(a) NRC Coment (TSAR Material Report Addendum)

GNS will report on results of compact tension fracture toughness tests (per ASTM E-399).

Tests will be perfomed with 20 mm plate at 350*C looking at base material, weld material, and heat-affected zone. Number I

of tests:

3.

5(a) GNSI Response Compact tension fracture toughness tests were performed on 20 m thick borated stainless steel plates denoted as Radionox A 18 with 1.07% boron.

The tests were performed in accordance with - ASTM Standard E813, which was judged to be more appropriate than ASTM E-399 in this situation I

because the specimens do not show a linear elastic behavior at the 20 m thickness.

Therefore, the tests were performed using the J-integral method defined by ASTM E813.

All tests were performed at 350*C and at a quasi-static loading rate with one exception.

One sample was tested at a rate consistent with

(

ca sk drop loading condi tions which verified that this material I

characteristic is not strain rate sensitive.

Testing was performed on base metal and welded specimens.

These tests showed about 35% lower values of fracture toughness in the welded specimen than in the base I

metal which is typical for conyentional stainless steel weldments.

This further supports the borated steel material acceptance criteria differentiation of 15% for base material and 10% elongation for welded I

material described in the Material Evaluation Report.

The qvpyage static fracture toughness values for these were 87 and 54.5 Ksi-in M.

I 5(b) NRC Comment (TSAR Material Specification)

Radionox material specification will define Charpy-V notch values at 350*C.

The lowest of three samples will equal or exceed 15 ft.-1bs.

Reference DIN 50 155, 50 122, 5(b) GNSI Response Charpy V-notch testing will no longer be used to qualify Padionox per the Material Specification.

1I I

I l

I 8

JANUARY 1987 -

(0068W)

E 5(c) NRC Comment (TSAR Material Specification)

Material specification will delineate minimum elongations at 350*C as follows:

Base Material - 15%

Weld Material - 10%

Reference will be made to the DIN standards 50 145, 50 120.

5(c) GNSI Response Section 6.0 of the Material Specification (BS 05, Rev. 2, January,1987) l establishes the requested minimum elongations for both the base material I

and weld material.

I 5(d) istC Comment (TSAR Dynamic Drop)

A discussion of the applicability of the basket material should be made with particular attention directed toward demonstrating that either brittle fracture is not a failure mode or if it is, that safety margins against brittle fracture are adequate under dynamic loading conditions.

5(d) GNSI Response The Material Evaluation Report presents the results of Radionox A 18 base me tal and wel ded me tal material testing for the bounding temperatre range of interest, including yield and tensile strengths, I

el ongation, reduction of area, and dynamic tear tests.

These results show that both base metal and weld metal demonstrate good ductility sufficient to preclude brittle fracture as a credible failure mode.

The I

dynamic tear test results show that, similar to austenitic materials, no transi tion from ductile to brittle behavior is observe'd as the service temperature is reduced.

The measured elongations are lower than those I

similar to the elongations that have been observed in the heat-affected that would be measured for austenitic stainless steel base metal but zone (HAZ) of welded austeni tic stainl ess steel componen ts.

Conservative acceptance criteria have been developed for this situation I

in the ASME Boiler / Pressure Yessel Code,Section III, Code Case N-47, based upon strain limits in both the base metal and weld metal.

These strain limits have been adapted consistently to apply to the borated I

stainless steel basket.

The TSAR evaluations show a safety maroin of greater than 10 relative to these conservative strain limit cri~teria.

Therefore, the l ower ductility does not compromise the structural integrity of the basket.

I I

9 JANUARY 1987 I

(0068W)

I 5(e) NRC Comment (TSAR Material Report Addendum)

A report will be prepared defining the results of tests perfonned to I

include relevant details on such factors as how specimens were prepared, date including Cv, yield tests and elongations.

The report will extensometer readings, etc.

Infonnation will be on both base metal and weld metal.

I 5(e) GNSI Response I

The Material Evaluation Report provides all the pertinent in formation relevant to both base material and weld material.

E 5(f) NRC Comment (TSAR Material Specification) temperature range (20-400,roperties will The following material p be defined as function of the C).

o Young's Modulus I

o Expansion Coeff o

Yield Strength o

Ultimate Strength I

o Poisson Ratio (only at room temp.)

5(f)

GNSI Response l

The in formation reques ted has been declared Proprietary by General Nuclear Systems, Inc.

This response has been included in the attached appendix which includes additional Proprietary information.

Point 6 Questions in this area were covered in Points 1-5 preceding.

l lI

,I I

I I

10 J ANUARY 1987 (0068W).

E I

I I

I I

I APPENDIX TO NRC COM4ENTS I

ON CASTOR V BASKET ANALYSIS I

I I

I

• • • NOTE ***

The information provided in this document has been declared I

Proprietary by General Nuclear Systems, Inc., and is therefore not included in this Non-Proprietary version of the CASTOR V/21 Borated Stainless Steel Basket Evaluation Peport.

I I

I I

I i

11 JANUARY 1987 (0068W)

I

.I I

/

MATERIAL EVALUATION REPORT FOR CASTOR V/21 BASKET MATERIAL JANUARY 1987 REVISION 1 t

I

• • • NOT E**

• The information provided in this document has been declared Proprietary by General Nuclear Systems, Inc., and is therefore not included in this Non-Proprietary version of the CASTOR V/21 Borated Stainless Steel Basket Evaluation Report.

I (0027W)

I

I REVISION FOR CASTOR V/21 TSAR I

4. 2.1. 4 FUEL BASKET ANALYSIS The fuel basket has been analyzed under maximum normal handling accelerations and under maximum thermal loading conditions.

These I

analyses have been performed to demonstrate that no component or weld in the basket structure fails under these limiting conditions.

The fuel basket structural analysis is based on a maximum handling load of 3 g as identi fied in Section 3.2.5.4 The analysis considers both lateral and vertical loadings on the basket struc ture.

It is noted that seismic loadings will be much less than

+ 3 g; therefore, the handling load acceleration will envelope the seismic load.

I The stresses in the fuel basket were calculated using the reference PWR fuel assenbly for which the CASTOR V/21 is designed to accommodate.

From Table 3.1 -1, it can be seen that this fuel assenbly will be a 15 x 15 PWR fuel assembly, having a weight of 654 kg (1442 lb) and an overall assembly length of 4057 mm (159.7 inches).

The total fuel weight for a complement of 21 assemblies is then 21 x 1442 = 30,282 lbs.

An acceleration of 3 g will be applied to the results.

The basket is constructed of a combination of stainless steel I

(Material No. 1.4541) and borated stainless steel (Radionox A18).

The properties of these materials are provided in Section 3.2.5.3.

The fuel basket geometry and dimensional information is provided on the proprietary drawing, A500.11-15, Rev. "e", included in Appendix 1.

An optional fuel basket constructed entirely of stainless steel (Material No.1.d541) is also available.

The following structural and thermal analyses are applicable for both fuel basket options.

I I

I The thermal loading for the thermal s.ress analysis of the basket is based on the maximum fuel heat load and environmental conditions.

The basket peak temperature profile for 21 KW peak fuel, 54*C anbient, and solar insolation is shown on Figure 5.1-9.

Detail s of the basket structural (normal handling) and thermal stress analyses are provided in the following sections.

The structural analysis of the fuel basket under accident conditions is presented in Section 8. 2.1. 2. 3. 3.

4.2.1.4.1 STRUCTURAL ANALYSIS - NORMAL HANDLING I

( A) Lateral Loading I

1. Fuel cavity plates at location X.

The following analysis was based upon the dimensions provided the proprietary drawing, A500.11-15, Rev.

"e" (Appendix on 1).

Reference is also made to the dimensioned sketch in Figure 4.2-19.

For a typical fuel cavity X, the distributed load, w, is:

I weight of fuel assembly W*

tuel cnannel area

_ 1442 lb.

= 1.03 lb/in?

=

tibi In) t u. / In) _

Checking for bending stresses with a 3 g acceleration on a unit beam (10 nun thick plate) assume simply supported at both ends (again see Figure 4.2-19).

Note that this assumption does not take credit for the pusset plates which do provide additional support.

I C

"m. (

ax

)x3g l

2

I I

where:

WI

(

T = _1.03)(8.7)

M

=

9.8 in-lb

=

max 8

= Tl bh3 (1)(.4)3 I

I

.0053 in4

=

12 C=

0. 4 in 0.2 in

=

(9.8 in-1b)(0.2 in) x 3g

(.0053 in")

1100 psi o =

Also evaluate the bending stress caused by the acceleration resisted by the sideplate (0.2 in or 5 mm).

This analysis assumes that the plate would resist the entire loading.

Subsequent analyses (calculation A.5 of this subsection) will be presented which shows that the gusset plates will also restrain the lateral loading independently of the plate.

Again,

,MC o -

x3 M

9.8 in-lb

=

0. ? in C

I 0.1 in

=

=

h (1 x 0.2 )

I

=

.00067 in I

=

(9.8 in-1b)(0.1 in) g,

x 3g

(.00067 in')

4410 psi

=

o I

S, for the basket material is 12,600 psi at 3 50 *C, The stress limit for primary bending is 1.5 Sm = 18,900 psi.

Thus, the above-calculated bending stresses are well below the limit.

I

I FIGURE 4.2 - 19 k

L f

l 10 mm P 3

s I

P 103 lbslin 10 mm t

n o e

e s e

l b

l_~

87 in

- l 10 mm P I^

A

/

(Bosket Cell Edge)

I e

220 mm(8.7 in)

Loading i

I g

r, 10 mm or.4 in _\\

VMMA E

h (Basket Celt cross-se: tion)

I 2

h El 10 mm E'

P s

I d*

I I

I A-A Fig. 4.2 - 1.

CASTOR V/21 - Sketch of Tuel GNS

=**o....

I I

4

I I

Next check the cell beam for shear:

I a) Plate Thickness 10 m (.4 in)

=

2 Unit Width Area 1 x.4 4 in

=

=

Maximum Shear Load 1.03 lbs/in x 8.7 in 4.48 lbs

=

=

SUPPORT REACTION Shear Stress:

3 V

3 4.48 S'i "T A "T

x x3g MAX ogg,,x 50.4 psi

=

b) Plate Thickness 5 m (.2 in)

=

3 4.48 gg 33.6 psi x 3 g I

o

=

=

2

0. 2 3g 100.8 psi 0

=

The allowable stress in shear is 0.6 S, = 7,560 psi at I

350*C.

Therefore, it can be seen that the calculated shear stresses in the cavity plates at location X are well below the limit.

2 Shear Blocks The plates separating fuel positions I and II (typical of I

four places in the basket) are held in position by bolts which are inserted through the adjoining 20 m pistes and threaded into a small rectangular plate which serves as a shear block.

These bolts are M20 x 45 and have a thread root diameter of 20 m (.787 inches).

The bolt material is A2-70 stainless steel (equivalent to Material No. 1.4541).

At 300*C, the highest temperature at the local position in the basket, the bolt has a yield strength of 48,900 psi and ultimate strength of 76,100 psi (Ref. DIN 267 Part 11).

A I

5

I I

typical plate is shown in Figure 4.?-20 Assume a uniform loading of 1.03 lb./in.2 over the cross-hatched area in Figure 4.?-20.

The load per bolt will be:

2 (1.03) lb/in )(8.7 in)(25.2 in)/2 W1

=

113 lbs

=

I For a 3 g load the shear stress in the bolt will be I

(3)(113)/(.25)(, )(.787)?

=

t 696 psi

=

The allowable shear stress for the bol ts at 300 degrees I

Celsius is.6 S,

(. 6)(1/3)( 76,100) 19,220 psi.

The

=

=

factor of safety is therefore 21.9 I

I I

I I

,I I

I

I

I FIGURE 4.2-20 CASTOR V/21 FASTENER DETAIL FOR FUEL BASKET I

I 8.7

.n 220 l

l L

680MM l

l 640MM Typical I

l 159.45 in 4050MM p

I

/

f M20 x 45 a

x_u 25.2 ir

~

i 1

640 M

/

E N

/g i

i g

exSTtNER oEr m LOAD AREA ACTING ON FASTENERS I

I I

I I

7

I

3. Chevron Spacers Next, consider the spacer between fuel positions II and III.

This geometry is shown schematically in Figure 4.2-21.

The lateral load from fuel element II will be reacted by the 10 mm plate which forms the side of fuel position III.

If a I

simple beam analysis is used, then a beam of 150 mm in width (see Figure

4. 2-21 )

can be assumed to carry the load.

Assuming that each of the 16 spacers carries an equal share of the load, the spacer load can be approximated as two line loa.is acting on the plate.

Analyzing a width cf the plate equal to the width of the spacer, the applied load F for a 39 lateral load can be calculated as:

I (1442)(3)f 5.9) = 160.07 lbs y

I 059.45)

The shear and moment diagrams are shown in Figure 4.?-?1 By I

taking a

moment summa tion, the reaction loads can be determined as:

R 123.27 lbs

=

j R

36.80 lbs

=

p The maximum moment which will occur at the point of application of the load will be.

(2)(123.27) = P46.54 in-lbs M

=

max The bending stress may then be calculated as,

'I MC (246.54H.197) o =

q=

,g3

= 1,619 psi where:

C

.3937/?

.197

=

=

3 1/12 (5.9)(.3P7)3 4

I 1/12 bh

=

=

.03 in

=

I The allowabic stress is 1.6 S, a 18,000 psi at 350'C.

This results in a safety factor of 11.67 I

a

I FIGURE 4.2-21 I

I 2

1.03 lbs/in yy i,,,i,,,,, 9 i, i,

.,i, l

l

}

l l

SPACERS 50MM I

t(2")

b 2"

b 8.7" l

N f4

-- 8. 7 i n I

b SN21 ALTERNATING

.g SPACERS f(16 PLACES)

Nk i

0 t, *h\\N\\\\W 1

4050 MM 159.45 in.

(Y i

g i

m i_

I

_T

-E, mm -,No I

I 9

E 4

Weld SN-21 Next, check the shear stress in weld SN?1.

This weld is shown on the weld detail drawing to have a continuous bead of I

525 m near the upper and lower ends of the plate.

In the central portion, the weld is intermittent with 100 m of weld, and a space of 257 m between each weld.

Thus, out of 357 m only 100 m is welded, resulting in an efficiency of 0.28.

The weld is a full penetration weld which will develop the full strength of the plate for the welded portion.

The shear stress in the 10 m fuel cavity plate was previously calculated to be 50.4 psi for 3 g loading.

The shear stress in the weld can be estimated as, 50.4/.28 180 psi T =

=

This is well below the allowable shear stress for the basket material of 7560 psi at 350 degrees Celsius.

The factor of s a fe ty is 42. 0.

5.

Gusset Plates I

The basket design incorporates gusset plates at several positions near the outer periphery of the basket next to the barrel.

A typical detail is shown in Figure 4.2-22 The gusset plates are welded to the adjoining fuel cavity plates by wolds on both sides (weld details SN7 and SN11).

There are seven gusset plates along the axial direction of the basket.

Consider first the smaller gusset plate (! tem 27) at a

typical cavity location X.

Loading in the 0*-180' direction (V in Figure 4. 2-22) will be resisted by the I

shortest leg of this gusset plate which has a welded length of 60 m.

This plate will react the load in shear.

The shear area of the weld is, 2

(3)(.707)(60)(2)(7)

A1

=

1781 mm

=

?.76 in

=

10

E The fuel cavity plate is also welded (SN8).

This wel.d has a shear area of 2

(6)(100)(5)(.707) = 2121 mm A2

=

2 3.98 in

=

The total shear area 2

A1 + A2 6.04 in,

=

=

I The loading for 3 g's is (1442)(3) 4326 pounds.

=

I The shear stress is:

I P/A 4326/6.04 T =

716 psi

=

=

This stress is well below the allowable stress in shear of 7,560 The factor of safety is 10.6.

For loading in the lateral direction (indicated as L in Figure 4.2-22) the fuel cavi ty plate is supported by seven axially located gusset l

plates (Item 22), each with a weld length of 135 mm above and below the plate (SN-11).

This corresponds to a total weld 2

I area of 6.2 in resulting in a shear stress of 697 psi.

The sa fety factor in this case is 10.8.

Note that this I

calculation does not take credit for the fuel cavity plate weld (SN-18),

which would further reduce' the resulting calculated stresses.

I

'I I

I

,I 11

I FIGURE 4.?-22 BASKET DETAIL NEAR POSITION X I

I l

t 135MM-s

,,,,,,,,,,,,,x TYPICAL FUEL CAVITY I GUSSET PLATE I

L I

3N 3/

y SN11 I

SN8 SMMX100(320 )

I N

w N

l GUSSET PLATE d

60MM b

y

~

fhSN7 I

I I

I I

i2 1

c

I 6.

20 mm Plates The compressive load in the cruci form members due to a lateral acceleration of 3 g will be checked next.

First I

consider the 20 mm plates on each side of fuel position V.

Assume that the weight of elements I,

II, III and IV is supported by the 20 m plates when the cask is horizontal.

An acceleration of 3 g's will produce a force in the plates (based upon a one-inch axial length) of I

I F

4(1.03 lb/in )(8.7 in)(3 g's)

=

107.53 lb/in

=

2 For the 20 m (.787 in ) pletes, the compression area based upon a one inch axial strip is.787 in.

The direct or membrane stress is thus, P/A 107.53/(.787)(2)

=

=

o 68.28 psi

=

This is a very small axial stress and is almost negligible compared to the other loadings, f.e., hending loads.

Even if I

the total compressive loading were assumed to be carried on the 5 mm plate, the stress is less than 300 psi.

Thus, it is concluded that the membrane stresses due to lateral loading of 3 g are small and will not lead to overstress in the fuel plates.

The buckling stress for the 5 m plate was calculated to be 10,561 psi.

This assumed a one inch axial strip and the classical Euler equation was used with a pin-pin end condition.

The gusset plates along the 5 mm plate will actually increase the critical buckling stress.

I Based upon the foregoing analyses, it is concluded that the basket meets the design requirements for the lateral 3g loading condition.

!I 13 l

t

I (B) Vertical Loading The fuel is supported in the vertical direction by 10 mm plates (Items 17 and 18) which are bolted to the bottom of the basket with I

M12 x 25 bolts.

This bolt material is also AP-70 as previously described in Section A(2).

There are 10 bolts in each plate 17, and 8 bolts in each plate 18.

I Consider first Plate 17 at one typical location.

In a vertical orien ta ti on, this plate must support fuel elements VII and VIII totally, one-half of elements on either side of it (a total of 4) and one-quarter of the central element (III).

Thus, the total weight supported by the plate will be, I

1442 (2(1) + h (4) + h (1))

WT 6178.5 lbs

=

=

I The ten bolts will be assumed to share the load equally.

Each bolt has a root diameter of 12 mm (.472 inches).

The bolt area I

is 10 ( v/4)(.472)2 1.750 in?

A

=

=

The tensile stress developed in the bolts is P/A 6128.5/1.750 3502 pst o =

=

=

for weight only loading.

For 3 g's the stress will be g

E 10,506 psi.

The allowable bolt stress is 25,367 psi which l

results in a safety factor of 2.41.

' I Plate 18 contains eight bolts which must support the weicht of l

four elements (elements I and II plus one-half of elements VI, VII, XIV, and XV).

Thus, the total weight would be (4)(144?)

5768 lbs.

I 14

E The bolt area is 8 ( n/4)(.472)

A 1.4 in

=

=

This results in a tensile stress in the bolts of P/A 5768/1.4 4120 psi

=

=

=

I o

For a loading of 3 g's, the tensile stress will be 12,360 psi.

The allowat,le stress for the bolts is 25,367 psi resulting in a safety factor of 2.05.

I In addition to the bolted plates, there are two types of 10 rnm E

thick support plates that are welded to the bottom of the basket cavity plates that provide vertical support to the fuel.

These plates are shown in Figure 4.2-23.

The fuel will rest upon I

these plates as shown by the cross-hatched fuel bearing area.

In the vertical position, the fuel rests upon the bottom nozzle of the fuel assembly.

The four support posts on the botton nozzle distribute the load to the fuel basket near the corners of each fuel cavity.

The fuel botton nozzle is very rigid compared to the flexibility of the support plates (in hending)

I and therefore the load will tend to be concentrated near the weld rather than spread over the fuel bearing ' area as shown in Figure 4.2-23.

I First consider the "L"

shaped plate (Item 16).

This plate supports one-half of one fuel element and one-quarter of two others for a total of one fuel element on average.

If we assume that the weld is in pure shear then the shear area can be calculated as follows:

I 2

(235 + 50)(3)(.707) 604 mm A

=

=

2

.9369 in

=

E I

is

I The shear stress for a 3 g vertical load is (1442)(3)/.9369 P/A 4617 psi T =

=

=

The allowable shear stress for the plate material at this location is (.6)(12,600) 7,560 psi.

This resul ts in a

=

safety factor of 1.64 For the corner support plate, Item 19, only one-quarter of a fuel element will be supported.

The shear stress may be calculated similarly, 2

.3287 in.2 (100)(3)(.707)

A 212mm

=

=

=

(3)(1442)/(4)(.3287) 3290 psi T =

=

I The allowable shear stress for the basket weld material at 350*C is 7560 psi, resulting in a factor of safety of 2.30.

E E

I 1 I E

I I

l I

16

I FIGURE 4.2-23 E

I

-*-- 10 0 I

I\\3

~

I 5

i t, EEL.E_

235 i

285 SUPPORT PLATE - ITEM 16

'I I

N3

/

l A

/

l d' / /

' I

! /

50 l

v i

/

/

%,A 50 +

j SUPPORT PLATE - ITEM 19 17

I Next check bending stress in Plate 16.

The vertical 3 g load will be assumed to act one inch from the weld along the short side of the plate (see Figure 4.2-24).

The moment of inertia of the plate is hbh I

.023 in

=

=

where:

b 110 nn 4.33 in

=

=

h 10 mm 4 in

=

=

= h = (720 3.?)

o 18,782 psi

=

I The allowable stress, S,, for the plate is 12,600 psi.

The allowable for primary bending is 1.5 S I

18,900 psi,

=

m resulting in a safety factor of 1.01.

Thus, it is concluded that the fuel support plates are satisfactory to resist the assumed 3g vertical lead.

The stresses and resulting safety factors determined in the foregoing structural analysis calculations are summarized on Table 4.2-5(a).

I

' I I

g w

s

.e..

FIGURE 4.2-24 I

I I

/

i iP I4 I

lI/5

~

~

l

~ /;t I

/

Y 1.0' +

1 N 3MM s

s

,F

,,,, s u mn s

_j,110MN b

I I

I I

l 19

4.2.1.4.2 THERMAL STRESS ANALYSIS I

A thermal stress evaluation of the fuel basket was performed to determine the thermal expansion stresses in the basket due to the maximum expected fuel heat load and ambient temperature.

A thermal analysis of the basket was performed resulting in the temperature distribution shown in Figure 5.1 - 9.

This bounding temperature distribution was used as input to the thermal stress analysis.

The ANSYS finite element model is shown in Figures

4. 2-?5(a ),

4. 2-2 5(b ), 4. ?-26(a ) and 4. 2-26(b ).

The fuel separator plates were modeled using 2-D beam elements having a unit depth in the axial direction.

The gusset plates were modeled using ?-D plane stress elemen ts.

The beam elements and plate elements are connected by using comon node nunbers at the interface.

This ensures that in-plane displacement degrees of freedom are coupled on both the plates and beams.

The additional rotational degree of freedom associated with the beam element provides bending stiffness for the fuel separator plates.

The finite element model is thus capable of accurately modeling the interface between the gusset plates and the fuel separator plates.

I Since the gusset plates only exist at seven axial locations along the fuel basket, the finite element model was run both with and without the gusset plates to bracket the thermal stresses.

Because of the symmetry of the basket geometry and loading, only one octant was modeled.

Symmetry boundary conditions were applied to the beam and plate nodes at the symmetry planes.

Figure 4.2-27(a) shows the temperature distribution in the basket gusset plates.

These temperature values, in addition to the temperatures of Figure 5.1-9, were used in establishing the nodal temperatures that were used in

g the analysis, see Figure 4.2-27(b).

In the case where the gusset 5

plates were omitted, the same nodal temperatures were used for the beam elements (representing the fuel separator plates).

20

Table 4.2-5(b) lists the thermal stresses calculated at each joint in the fuel basket for axial l ocations wi th gusset plates and without gusset plates.

These stresses were obtained from the beam element bending stresses at the corresponding nodes.

I Figures 4.2-28 and 4.?-29 show these stresses pictorially at each joint in the fuel basket.

The differential expansion at the gaps is also noted.

As can be seen from Figure 4.2-?8, the thermal stresses are quite low with the highest stress calculated to be only 10,046 psi.

This stress occurs in the barrel portion of the basket which is fabricated from stainl ess steel.

The stresses shown are the largest absolute value of membrane plus bending stress.

The limit on the combination of primary plus secondary stress intensity range is 3 S,.

Assuming a temperature of 350 degrees Celsius, find from Table 3. ?-4(b ) S 12,600 psi, 3S 37,800 psi.

It has

=

=

m m

been shown that the normal operating stresses on the basket are quite low due to mechanical loads such as the 3 g handling load.

The highest thermal stress in the fuel basket itself is 7793 psi.

This stress occurs in the joint identified as Detail Z (SN 17/18) on the fuel basket drawing ( A500.11-15), and is for the case which includes the gusset plates.

An acceptance criterion based on strain Ifmits is discussed in the Proprie tary Material Evaluation Report

/Ref.

'1 /.

Since all i E calculated stresses are below the material yield limit, the strain i E criterion is also satisfied.

In the case of the highest thermal stress (7793 psi) the corresponding elastic strain is.0314%.

This results in a factor of safety 60, based on the 2% strain

=

limit for weld metal bending.

l l

l It is noted that the basket design is able to accomodate thermal expansion without developing high thermal stresses because of the gaps left for differential expansion and the overall design concept l

21

i I

which allows for free expansion between the center module, the four corner modules, and the cruciform section.

This can be seen from a I

comparison of Figures 4.2-28 and 4.2-29 The 3.5 mm gap provided to accommodate thermal expansion is shown to be adequate as witnessed by the 3.42 nun relative deflection for the gusset plate case and 3.25 mm deflection for the case without gusset plates.

Because of the importance of these thermal expansion gaps to the thermal stress analysis results, fabrication OA controls ensure that the specified gaps are maintained.

Specific measures instituted to verify the presence of these gaps are described in Appendix 8 I

This analysis further assumes that the basket is centered within the cask cavity.

The acceptable conclusions of this analysis are also valid as discussed below in the event that the basket is not centered.

In the vertical position the basket rests upon the bottom of the cask cavity.

If the basket is off-center with respect to the cask it will move toward the center as the basket heats up.

The only forces resisting this centering motion are friction forces between the basket and the bottom of the cask cavity; while the driving I

force is the thermal expansion of the basket once the gap at one side of the basket / cavity interface has cl osed.

Even if a coefficient of friction of 1.0 is assumed to exist between the basket support pads and the cavity bottom, the centering force will i

be equal to a maximum of one g in the lateral direction.

It has been shown in previous structural calculations (Section 4.2.1.4.1) that this loading is acceptable and when combined with the thermal stresses no material limits prevent the free expansion of the basket.

L

If the cask were assumed to be in the horizontal orientation, then the basket would experience a side load of 1.0 g which also has been shown to be acceptable and the basket would be free to expand vertically through the full diametrical clearance.

I A series of calculations were made to determine the stresses caused by the self-centering forces that would be developed when the basket heats up after being loaded.

It was assumed that the basket was biased toward one side of the cask and the center of the basket (measured vertically) expanded out against the cask at the nearest point of contact.

This would develop a moment which tends to tip the basket up at one bottom edge with the other edge taking the full weight of the basket.

It was shown that the welds which connect the basket plates to the outer cylindrical shell are ca pable of I

wi thstandaing this condition with a large factor of safety.

The stresses on the outer barrel were also evaluated and found to be very low.

The 20MM plates were evaluated for shear under this condition and the stresses were negligibly small.

The local radial loads necessary to move the basket against friction were also evaluated and found to be negligible.

Additionally, thermal load tests / Pef. 2/ were performed (under both steady state and cyclic thermal loadings at rated conditions) on a CASTOR V/21 cask to verify the structural integrity of the basket.

No permanent deformation of the basket nor any other indications of

,I excessive stress occurred during the tests.

A summary of the fuel basket stresses, including normal structural loading from the previous Section and thermal stress determined in this Section is contained in Table 4.2-5(a).

I 23

M M

M M

M

--Ib r-f9 N

8 CALCULATED LIMIT Um COMPONENT LOAD' STRESS (PSI)

CATEGORY PSI COMMENT 17.2 Di Center of 10 mm plate L

1,800 PB 18900 S.F.

=

Center of 5 mm plate L

4,400 PB 18900 S.F.

4.29

=

ISO Shear on 10 mm plate L

50.4 S

7560 S.F.

=

Shear on 5 mm plate L

100.8 S

7560 S.F.

75.0 C3

=

71 21.9 Bolts on Shear Blocks L

696 S

15220 S.F.

=

y C

II.67 rq Spacer on 10 mm plate L

1,619 P3 18900 S.F.

=

r-42.0 no Wold SN21 L

180 7560 S.F.

=

Gusset Plate (item 27)

L 716 S

7560 S.F.

10.6

=

g 178 - Buckling OK g

20 mm Plates L

68 PM 12100 S.F.

=

2.41 M

Vertical Support Plate Bolts (item 17)

V 10,506 PM 25367 S.F.

=

N 2.05 rn Vertical Support Plate Bolts (item 18)

V 12,360 PM 25367 S.F.

=

M 1.64 Vertical Support Plate Wold (Item 16)

V 9,872 S

7560 S.F.

=

t.n 1.01 Vertical Support Plate Bending (Item 16)

V 18,782 P8 18900 S.F.

=

3m Gusset Plate (i.. )

T 6,380 Q

36300 Note that PM + Q or PB + Q Is less than limit g

Barrel T

10,046 Q

36300

><E 4

--(

THERMAL

• L = LATERAL LOADING OF Jg, V VERTICAL LOADING OF Jg, T *

=

PRIMART MEMBRANE, Q = SECONDARY

• • PB = PRIMARY BEN 0 LNG, S = SHEAR, PM

=

l l

UI TABLE 4.2-5(b)

THERMAL STRESSES AT VARIOUS LOCATIONS IN THE BASKET UNDER DESIGN LOADING CONDITIONS AT AXIAL LOCATIONS HAVING AT AXIAL LOCATIONS NOT HAVING NUDE NO.

GUSSET PLATES GUSSET PLATED STRESS STRAIN SAFETY FACTOR STRESS STRAIN SAFETY FACTOR (psi)

($)

(psi)

($)

I 76 188

.0008 2,500 188

.0008 2,500 4

1,4ll

.0057 350 1,481

.0057 350 75 639

.0026 770 640

.0026 770 9

681

.0036 555 881

.0036 555 74 810

.0033 6 05 810

.0033 60s 12 1,322

.0053 3 75

'i,322

.0053 375 16 454

.0018 1,110 454

.0018 1,110 Id 1,491

.0060 330 1,498

.0060 330 25 4,54I

.0183 105 1,902

.0077 255 29 2,458

.0099 200 686

.0028 710 I

37 7,793

.0314 60 I,I55

.0047 425 33 7,6 34

.0308 65 1,432

.0058 340 Gusset Plate 6,380

.0257 75 Barrel 10,046

.0405 45 10,046

.0405 45 I

NOTES: III See Figure 4.2-26(a) for node no. location.

I-(2) Strain is equal to stress divided by 24.7E6 (at 400'C) x 100.

I3IAllowabie strain is taken to be 25.

1 l

t l

l I 25 t

1

I FIGURE 4.2-25(a) N0DE NUMBERS FOR FINITE ELEMENT MODEL I

I g

l aV ls s

1"

',,4;, p.w? Cl,+,,

i,

,4:=== q

,; w u,,.

I g

/,r }.,

U W

T e

if fs

/

lv l

i

rs l

o e

fi is at

-9 y-

.c o.y A

I,

,I l

7

.. T*

4. n s

I

l

• e/

I L:

J G'G W21 BASKET l

I l

26 g

I FIGURE 4.2-25(b) NODE NUMBERS FOR FINITE ELEMENT MODEL I

I t

I

$ ='",9 42 m

f ef..'

s I

/',s ls l e

}

• 11 j,/

• JE J."

l' jo 3-d

• /(

I Q

/

g)/

d d

C I?

,J

~

I

y., v lJa f

to j,

s I

7 Tf 7

i I

i

.N I E J

O'6 Ve'21 BASET I

I 27 m--

I FIGURE 4.2-26(a) ELEMENT NLNBERS FOR FINITE ELEMENT MODEL I

., 2. _m 1r r-

@$f..~

g1

=

27 p

jaa a,,

s-n1i er i f as

.r ;... h y.

I

( J:

Ji jo Y.Tf s'

4 i?

3o.

I

?,\\ s\\f w

a

% $ )\\ ':

\\

y 6,

e i;

w l

• f.',b h'\\'h'Y

, yc n

si 4

m

\\ 'r

.r-p J*,

I C

l l 37

• )

Jf gr k

i=

I 1

I VF M

,W d

/

6f th V 21 BASI:ET i

I 28

I i

FIGURE 4.2-26(b) ELEMENT NLNBERS FOR FINITE ELEMENT MODEL R

.Cy 1r gg.

y; 27 f.'

l21 Y E l% r.;

I p - l 31

/.

3i

' So a

45 I

qq d

.h l31 c.

I?

I l#

\\ f' y

3<

VS at l

4

/0 _

' I C

l*'

97

_f Jf W

k In 7

I t

W l

T*

,W -

r, I

3 To" I

L l

Gr!5 u 21 Basi:ET 1

I I

29

I FIGURE 4.2-27(a) TEMPERATURE DISTRIBUTION IN THE GUSSET PLATES I

I SN 7

,37 l

\\

146 l14 1

155

~SN 7 I149 152 155 15s 16 163 I

\\

152 155 158 168 162 165 167 169 171 1158 154\\158 162 165 168 178 172 174 176 178 188 113914615315816316717117517B189182184 185 F 188 129 138 146 153 168 166 172 176 181 184 188 198 192 194 195 196 197 287 I

A)

Gusset Plate Item No.

27 in GNS Basket Drawing A500.11-15, Revision e (see Appendix 1)

I 183 192 fiI2 288 I

1 SN 11 1 02 2e8 289 192 200 259 217

\\ l192 288 209 21 I

{

192 39 217 225 t

233 11 280 289 217 225 233 241 g

192 28 289 217 225 233 241 249 1192 28B 28 217 225 233 241 249 257 li?2 28B 289 21 225 233 241 249 I 265 l

183 192 298 289 217 225'233 241 249 257 265 274 278 278 B)

Gusset Plate Item No.

2?

in GNS Basket Drawing A500.11-15, Revision e (see Appendix 1)

NOTE: The temperature distribution shown is only for the gusset plate area which is welded to the basket plate.

30

v I

g FIGURE 4.2-27(b) NODAL TEMPERATURES USED IN THERMAL STRESS ANALYSIS I

l d

. - tio

~su i.y ar 4L,h

,,I IS' l

gp=::11'".

'*.fj.g pa..j, 1

i

,,r,;r '"

ae

.u

?

s.-

l

'.. -.. \\\\

. n:

O.. / l a

J'w)$w./,.,

\\.

.a n

.zz _ l __

l

. :j a

,=

E

,,dfInc..t1s *

.ttf__.w so

..soo 2t3 a.f sos

,C/

.SJ /

I E

33' M

.s#

3 47 i

l JV'

!I JY2 350 357

' a' !:. a l E:ASI:ET 15AR TEMP bl5T. J UllE, Id86 I

I ll 31

I FIGURE 4.2-28 STRESS INTENSITY FOR BASKET WITHOUT GUSSET PLATES I

.327 MM l

10046*

2491 14 3 2,/,,./

I 686

/

7 454 I

3.25f0,/-

~A

\\

I

,1m 810 881 E

1155 N

640 g

1411 MAX STRESS IN RING t

o I

I I

lI 32

FIGURE 4.2-29 STRESS INTENSITY FOR BASKET WITH GUSSET PLATES I

{,10046* -

.480 MM l

1491

/p I

/j.#

N 454 7634 p

3.423 MM /

,,/

2451 O

• r A

6380 I

s 1322

'g,e

\\

e' I

10

\\

881 4541 639 lI

• MAX STRESS IN RING o MAX STRESS INTENSITY IN PLATES

/ g 188 9

6 I

G145 BASI'ET

.lI I

I 33

8. 2.1. 2. 3. 3 ANALYSIS OF FUEL BASKET (SIX-FOOT DROP ACCIDENT)

The fuel basket was evaluated to determine the stresses due to a deceleration of 52 g's in the horizontal direction.

An analysis of the six-foot drop accident condition was presented in the I

preceding section of this report.

This analysis assumes that the CASTOR V/?1 cask is dropped from a height of six feet onto the storage pad.

The worst case situation was found to be the side drop orientation in which a

deceleration of 52 g's was cal culated.

As noted in Section 8. 2.1. 2. 3. 2, the analysis was considered to be very conservative.

For the purpose of this I

analysis the horizontal deceleration is, therefore, assumed to be 52 g's.

Stress limits er the accident condition loads are based upon the design criteria of Table 3.2-1.

The limiting stress values are calculated in accordance with the ASME Code (Ref. 4.2-9) Appendix F and Reg. Guide 7.6.

I The allowable stresses used for this analysis are determined from l

the material properties of Radionox as follows.

The maximum fuel j

basket temperature is 350*C.

At a temperature of 350*C:

l 2

S = 29,100 psi (201 N/mm )

y 2

S = 67,000 psi (462 N/mm )

The value of S, taken to be the lesser of ?/3 S or 1/3 S '

y u

which in this case would be the former.

Thus, S, = 19,400 psi (133.8 N/mb For primary membrane stress intensity, the limit is the lesser of

.7 S or 2.4 S,.

Thus, the allowable stress is:

u l

2 2.4 S, = 46,560 psi (321 N/m )

lI 34

The stress intensity resul ting from the sum of the primary membrane stresses and the primary bending stresses is limited to the lesser of 3.6 S, and S.

Thus, the allowable stress is:

u S = 67,000 psi (462 N/m )

u The two ANSYS finite element models used in the evaluation are I

shown in Figures 8.2-22 and -23.

(Figure 8.2-22 is identical to Figure 8.2-23, except elements are added to represent the gusset plates ).

An enlargement of the indicated gusset plate is shown in the lower right corner.

Note that quadrilateral elements are used in areas of high stress since they are more accurate than triangular (constant strain) elements.

I The models are composed of ANSYS two dimensional plane strain elements.

In places where adjacent plates are not wel ded, I

coincident nodes are used in the separate pieces to allow relative motion between the plates.

The coincident nodes are coupled in the direction of contact when appropriate.

Since the plane strain elements are assumed to have unit thickness, the material properties of the gusset plates were altered by the ratio of actual thickness to total l ength, i.e.,

7(10)/4080.

This factor is applied to both the density and modulus of elasticity.

I Figure 8.2-24 is a node pl ot on which arrowheads indicate locations and directions of imposed displacements, and the solid lines indicate where pressure (which simulates the fuel weight) is applied.

Symmetry boundary conditions are applied to the nodes on the basket vertical centerline.

In addition, nodes at the lower edges of the 20 mm and 10 mm plates are restrained in t' e vertical r

(UY) direction to represent contact wi th the barrel.

It was determined that the gap between the lower edge of the 5 mm plate I

and the barrel will decrease to approximately 1 m due to thermal expansion of the plates.

A displacement of 1 m down is imposed on the modes at this point, in effect, closing the gap.

35

I An acceleration of 52 g's was applied to the structure.

The effect of this loading on the fuel cells, which are not modeled, is applied as a pressure along the top surface of the plates supporting the fuel cells.

Since each fuel element weighs 65a Kg, and the surface area of the supporting plate is 220 m by 4080 m, the pressure for 52 g's is:

52 x 9.807 x 654/(220 x 4080) = 0.3716 N/mm2 Figures 8.2-25 and 8.2-26 show the deflected shapes of the two model s.

In each Figure, the displacements are plotted 15 times the actual size as indicated by *DSCA = 15 on the right side of the plot.

The stress results of the two-dimensional analyses are shown in Figures 8.2-28 and 8.2-29.

Figure 8.2-27 shows the nodes in the model without gusset plates, where the stresses are the highest.

Based upon these results, it was evident that the two-dimensional model was nol: capable of accurately accounting for the effect of the gusset plates on the 5 m fuel element separator plates.

I Since the analyses indicated that the effect of the gusset plates is very important, it was decided to perform a more detailed analysis of the 5 m plate which is partially supported by gusset plates spaced at approximately 650 m (25.6 in. ) along its length.

Figure 8.2-29 shows the details of how the gusset plates are used to support the 5 m plate.

The finite element model is shown in Figures 8.2-30 and 8.2-31.

Loadings for the three-dimensional I

model were obtained from the two-dimensional analysis models and from the applied loading.

The line 1 cad shown in Figure 8.7-29 was obtained from the two-dimensional model s.

Since the structure is highly redundant, the vertical load which is applied to the 5 m plate is greatly dependent upon the relative stiffness of the fuel basket components.

This load was calculated in both two-dimensional model s (wi th and wi thout I

36

I gusset plates).

The worst case was found to be the model without gusset pl ates.

This load was applied to the three-dimensional plate model along with the equivalent force due to 52 g's acting on the fuel element in the adjacent fuel cell.

Only half of this fuel load was applied to the model since the other half would be taken by the 20 m vertical plate which also supports the fuel.

The loading due to the fuel element directly above the 5 m plate was modeled as an equivalent uniform pressure plus a line load at the gusset plate.

The relative flexibility of the 5 m plate in bendir.g compared to the vertical support provided by the gusset I

plate will result in a larger portion of the load being carried by the gusset plate.

Since the actual distribution of the load is a complex function of plate stiffness and flexibility of the fuel el ement, a distribution of 25 percent uniform pressure and 75 percent line load was assumed.

The actual distribution may be different from this but the approach taken in this analysis is to show that there exists an equilibrium loading condition that meets the applicable stress limits.

In an actual accident, the 5 mm plate will probably yield and allow the loads to redistribute in a I

way which will result in the lowest energy mode for the structure.

Thus, showing that the structure can meet the acceptance criteria under an assumed load distribution that meets equilibrium is tantamount to establishing a lower bound analysis.

The finite element model is composed of ANSYS STIF63 elements I

which are quadrilateral flat shells.

Along the free edge of the gusset, the elements become triangular flat shells.

The thickness of all the elements is 5 m.

This is the actual thickness of the horizontal and vertical plates and is the thickness to midplane of the gusset due to symetry about this plane.

Symetry boundary conditions are applied to the node points on the axial ends of the model.

The lower edge is restrained in the i

vertical (UY) direction due to contact with the barrel as is the horizontal edge.

All boundary conditions are shown in Figure 8.2-29.

37

At 52 g's, the uniform pressure due to the weight of the fuel element is determined as follows:

Wt. = 654 kg x 9.807 = 6413.78 N Area = wl = 220 x 4057 = 892,540 mm2 g 's = 52 P essure = 52 x 6413.78 = 0.3737 N/mm2 892,540 The portion of the fuel assembly weight supported by the plate segment in the model is found to be:

2 Model area = 220 x 325 = 71,500 mm Force = pa = 0.3737 x 71,500 = 26,717 N This load will cause the sapport plate to bend after which most of the load will be concentrated over the gusset plate.

As stated earlier, the loading on the model was divided so that 3/4 was applied as force to the top surface of,the gusset plate and 1/4 was applied as unifom pressure to the horizontal plate, i.e.

1 i

2 Applied pressure = 0.3737/4 = 0.0934 N/mm l

Applied force = 26,717 x 3/4 = 20,038 N The applied forces are divided so that each of the 10 inside nodes receives 1/11 of the load or 1822 N and each end node receives half of that, or 911 N.

l 38

The results of the analysis are shown in Figures 8.2-32 through 8.2-35.

The deformed shape plot in Figure 8.2-32 indicates that the 5 m plate deflects approximately 1 mm at the point farthest from the gusset plate.

The stress intensity at the top, middle and bottom surface of the plate elements is shown in Figures 8.2-33 through 8.2-35 respectively.

The nodal point stresses do not exceed the limits on primary membrane for that portion of the plate which is subjected to a uni form membrane load.

The membrane stresses in the vertical part of the plate were less 2

than 100 N/m everywhere.

The gusset plate is subjected to a complex distribution of stresses.

These stress are categorized as primary membrane plus primary bending due to the eccentric load on the gusset which will create a bending moment.

The stresses in the gusset plate do not exceed the allowable stress 2

2 of 462 N/m.

The maximum stress of 205 N/m shown in Figure 8.2-34 occurs at the very bottom node.

The triangular element used at this point is a constant strain triangle which is known to be much stiffer than a quadrilateral element.

The nodal stresses at the two nodes at the base of this element were averaged to obtain a better estimate of the stress.

This 2

resulted in a stress of 145 N/m which is well below the limit 2

of 462 N/m.

Based upon this analysis it is therefore concluded that the fuel basket will meet the stress limits for an assumed deceleration of 52 g's.

In addition to the stresses in the plates, it is necessary to check the welds and fasteners which connect the plates together.

These items may be evaluated by the hand calculation methods used in Section 4 of this report.

Weld SN21 Referring to Section 4.2.1.4.1(A) item (4) it was shown that the shear stress in the weld SN21 was 180 psi for a uni form deceleration of 3 g's.

For 52 g's this stress would be:

(

) (180) = 3120 psi (21.52 N/m2) a =

39

The allowable stress is taken to be 60% of the limit. on primary 2

membrane stress for accident conditions, or f.6) (321 N/m )

193 N/m2 (27,985 psi).

Weld SN7 and SN8 Referring to Section 4.2.1.4.1 ( A) Item (5) it was shown that the 2

shear stress in welds SN7 and SN8 was 716 psi (4.93 N/mm ),

This analysis very conservatively assumed that the total load was taken by the welds on one end of the 10 m plate which supports the fuel element (SN7) and the gusset plate weld (S N8).

The

" free" end of the plate was assumed to carry no load since there is a 3 m gap left between the fuel basket and the outer barrel at assenbly.

For the accident conditions, however, it will be assumed that the 3 m gap closes due to the elastic deformation of the gusset plate.

This assumption is justified on the basis of the small gap and the plate width of 220 m resulting in a flexible cantilever.

Thermal expansion due to the heat load will also tend to close the gap.

Thus, the 10 m fuel plate will be supported on both ends.

This reduces the 3 g shear stress in the 2

weld to approximately 357 psi (2.46 N/m ).

For 52 g loading the shear stress would therefore be 42.6 N/m (6182 psi).

This stress is well below the allowable stress value for pure shear, accident condition, of 193 N/m2 (27,985 p)i).

Bolts on Shear Blocks Referring to Section 4.2.1.4.1( A) Item (2) it was determined that for a 3 g loading the shear stress in the bolts which are used to-attach the shear blocks to the fuel plates was 696 psi.

This stress was calculated considering only the bolt cross sectional area.

There is an additional load path from the shear block to the plate via a 30 m long weld (SN7).

The shear area of this weld is 30(3)(.707) = 63.63 rn2 (.099 in ).

This reduces the 2

40

shear stress from 696 psi to 587 psi (4.05 N/m ).

Thus, for a 52 g

deceleration the stress would be 10,175 psi (70.17 2

N/m ).

This is well below the allowable stress in pure shear of 31,962 psi (220.4 N/m ) which is based upon the bol t material properties at 300*C, The results presented in this Section are sumarized in Table 8.2-2.

I I

l l

I TABLE 8.2-2 SIM1ARY OF FUEL BASKET STRESSES AND LIMITS FOR ACCIDENT CONDITIONS - 52 g LATERAL LOAD I

CALCULATED CATEGORT LINIT SAFETT COMPONE NT STRE55 PSI F ACTOR CONNENT 2,

(N/se )

S em Plate 13,521 P,

46,S14 3.44 See Figure 8.2 34 (93.25)

(321)

S se Piste 20,909 P, + P, 67,034 3.21 See F i gures 8.2-M (et

'" ~

(144.23 (462) model I

10 se Plate 8,494 P

46,514 S.48 From 2-D Analysis e/o (58.58)

(3213 gusset PL-nodes 1620 & 1689 10 se Plate 19,855 P

+P 67,034 3.50 From 2-D Analysis e/o g

(132.13 (462) gusset plate - node 1782

)

20 se Plate S,938 P,

46,584 7.84 From 2-D Analysis e/o (40.90)

(321) gusset plate. nodes 1426 and 1508 20 se Plate 18,995 P, + P, 6 7,0 H 3.53 kom 2 0 Apelysis e/o (134.03 (462) gusset plate - mode 1329 Gesset Plate 10,759 P,

46,514 4.32 See Figure 8.2-34 (Stress (74.23 (321)

Cont. DI Gu sse t Plate 26, 762 P, + P, 6 7, W 3.H At boPos CMap of pleH (150.08)

(462) told SN 24 3,120 Sheer 27,985 8.97 Sheer in sold (21.52)

(193)

I tolds SN7 + $NS 6,877 Sheer 27,985 4.53.

Sheer in sold (42.63 (193)

I I

Sheer Stock 10,179

$neer 38,962 3.14 Sheer in toit (70.17)

(220) ep. PRINARY MEN 8RANE, P, = PRINART SENDING

E L

i FIGURE 8.2-22 MODEL OF FUEL BASKET NOT INCLUDING GUSSET PLATES I

!ii i

-u rs a.e

[n!g 7$7 o

an m

I I

I

I Ii 1

I gag gg l

E R

-l e

g i

fT I

see;;

i I

s l

)

.i f

.l d

l l I 43

iI FIGURE 8.2-23 MODEL OF FUEL BASKET INCLUDING GUSSET PLATES

..i j s

I

-se m

lI s";r TR

'I god 3?

5 us m

I I

l msssmmas 1I

~

~~~~'lll:::x

,g::;! l t

3*

g F

h l

l i

l i

I 3

J 5

I i

2 4.

o 3

d I t l

!E 44 lI

I FIGURE 8.2-24 FUEL BASKET MODEL SHOWING RESTRAINTS AND LOADS I

A N.

@W T

A

.450 th TZ GJ

$ (9 =

.4 d v4 CD th (U F e s.44 e eF I

etOOs so

.4 F m O a lai m m o O m a V) e WTEQQmmE

".) M it A v41 F E LL E G.

NQX t

I

.-E:::::::::I p:1:....,.. s...

.. i.s

, e.

. s.

........... ! :1 pE t

I

,e is

. r::::::::::::::

5 *; '

=

r

.: j....**

g-...

.s..t.i,

.&:.: *M,

.E:::::::::::

.:3 33 j:ji::::::::::: w....

l I

h ::::.......... ::::

1

.... : ::::::::::::::: :::::::::::::. ::::4

n.:.::::.:.:.:.::::

Jiiiiiiiiiiil*iii*diliiiiiiiiili*ili!!iE::::g:::::.iiiiii.iiiiiliiiiiiiifliiiiiiiiiiiiilli y e 9

~

e u

JL A sk A

sk A

Jha JL 8

E B

M I L 3

m.

m I

, I l

e4 4 :

m 45

I FIGURE 8.2-25 DEFLECTED SHAPE PLOT OF FUEL BASKET NOT INCLUDING GUSSET PLATES I

sI 4

dbw m

Cr)

SQ ru bdYY Y

h B

T F IL K d F (9 X 4 gC=WWW smego NQXbo N

M I

I I

. I I

I I

1 P

I s

l l

lll"I

'g h

lii i.M iii iii %l 3

x m

sms sms B

IF A I

I

,I

>4 l

T I

1 I

we 46

i n

,v,ia s. -- -

ANSYS 4.38 k

7g DEC 3 1988 co

)

13:49:29 y

1 POST 1 DISPL*

STEP =1 ITER =1

{

o 9

j 20=1 E

l

~

DIST=825 3

l DMAX='i.24 y

XF=375 l

1

• Dscas i

?,

?'^

s p;;;

I, N uii Ases=si S

iU ii I

a

~

ug E

j EEE' P

j 1

j

!===

{

yagrys m i

a e Iil E

98 E

O WIII Ca rn

-g ansecar utm s

M M

M M

M M

M M

M M

M M

M M

M M

i 4

r

..a a.

l ANSYS 4 83 DEC la 1988 k

a i.

16:23:38 m

POST 1 DISPL.

'y i

STEP =1

("#

ITER =1 l

w m =

t gy.g g

m 8,

DIST-413 gr

'-"J XF=375 B

YF=-375

[h

~~#

^

DMAX.2.3 g0 r 2/3

• DSCA=f5 G"

r227

~

~

.I 2 52 hh r

m m r @2 2 7 fgj l

z z

$f

$h

~

34 G 't 225

-116

..........T.,,

isa m

- < ~

g"

$2

/67 i

i g-i 38 Gf g3

~ ~ ~

318

- z 32

=3

+

1 L zz7 d

< z G zi4 Y y:

N ffff M

,)

' 77 i

% /S9 i

1 qn3

=

( 22,

'H3 2

z 3

o 3

t./wir = n 2 w/ a m "

g i

y w

1 m o M b!THOUT GUSSETS

[-

g NOTE:

The displaced shape and stresses indica ted in this figure I

correspond to a 2-dimensional model without gusset plates.

More

)

realistic stresses are calculated usino a 3-dimensional model.

I FIGURE 8.2-28 NODAL STRESS INTENSITIES (N/m )

WHERE PRIMARY PLUS SECONDARY STRESS INTENSITY ARE COMPARATIVELY HIGH - GUSSETS INCLUDED

.I v"u n

I go S

7C;?

USE SEE I

u l

$L& wmu%% )

,r h

l 6N g

j ::;f j j;:::%

1 r

I u,.-..

y 6-I 5

l h

I

=

=

z 5

5-J B

B e-N 3

E I

L m

I a

' I s

49 I

I FIGURE 8.2-29 A SCHEMATIC OF Smm PLATE AND GUSSETED AREA 0F FUEL BASKET I

I I

egg?

l w LINE LOAD FROM UPPER STRUCTURE AND ADJOINING CELL

_y g'

IFORM PRESSURE I

/

di -VERTICAL SUPPORT LATERAL

% -LATERAL SUPPORT SUPPORTS I

i s

MODEL

~ 6/"

~

/

) // 's #

g I

95

,W' I

50

i e

e e

e m

e e

e m

e e

e M

m M

m m

m e

r a

1 I

9 ANSYS 4.3 5

8 NOU es 193g y

g 12:29:13 y

j iJ 7

PLOT NO.

1 8

Q 6

PREP 7 ELEMEN 15 NNUM-1 4

q o

g ORIG SCALING o

o XU=-1 9

8 8

_,7 vv.i 2

ZU-1 fc[6 7

~

pI97 225 6

5 XF=477 5

j 4

Q

' 4 YF=-477 g

6 s

Jr 1

ZF=-125 g

W 7

m 2

j L8 6

g 9

g b

3 w

i

% d j R

1 DETAIL MODEL OF QUSSET AREA 52 0's

I j

m 4,gg g

-r i

DEC 15 1988 m

i 13:37:40 PREP 7 NODES L,

m FBC=1 PRBC=1 r-o l

4 XV=-1 8

w l

YV=.5 i f

%,W' ZU-1 U

u DIST=193 I P Q

i XF=489 o

o YF=-502 8

t ZF=-137 8

ut

~

N F

\\

g 1

1 O

O A

I 4

i e

l 8

e l

• 8

];

1 i

\\

i i

DETAIL OF QUSSET AREA ADD LOADS FROM UPPER STRUCTURE I

l

M M

M M

M M

M M

M M

M M

M M

M M

M M

M 1

rvaia - 1. e -

ANSYS 4.ES 4

DEC 3 1998 13:18:36

• m POST 1 DISPL.

b STEP =1 ITER =1 mx 9

XV=-1 NNNW y

vva.s ZU=1 N

g I

N N

Na 5

i N

N

/

YF=-502

?

N N

f l

\\

N

\\

/

/

N f

ZF=-137 0

N N

/

DMAX=1 02 N

N N

s s

DSCA=ELOS m

j N

/

/

g s

N N

N

/

?:

NN N

s s

p N

N s

/

\\

%N N

N

/

/

s

~

g

/

j NN N

N o

s

/

m N

N s

N N

w

/

N j

N N

N

% /

/

N

% Q N

s4

}

s /

/

(N 1

N N

s%

% Y Nr l

l DETAIL OF QuSSET AREA COMB. LINE AND PRESSURE LOADS g

l l

I

i cv.,s

s. w -

E gg 4.85 9

JAtt 81 t

I ts:45:35 pogyg STRESS L

w STEP *9988 ITER =1 SI P$

C TOP

$ p, 1

i

• 5

• p m n XV=-1 3

L '

4a YV=.5 5$

n B

f i

ZU-1 3g pisT=214 a

na XF=489 y

4 ec E f 0

VF=-502 p,,_ 37 5; f EDGE

\\

Mx=293 3

@n 3

n

(

MN=5.23 r-9 A=23.1 3

g B=41.1 0

I B

C=59.1 P

0 p=77 1 i

3 EF F

(,

E.95.1

%a F.113 Q.131 g

H*1 m

D 1 167 Ja188 h

c,"

a g DETAIL OF QUSSET AREA ADD LOADS FROM UPPER STRUCTURI t I

i

ll 2

ev-..

..m-ANSYS 4 33 i

JAN S 1937 m

16:48:53 POST 1 STRESS L,

STEP =9999 ITER =1 SI

= m MIDDLE EN i

M m

i XV=-1 E

o YU=.5 kU a

zv=1 a E!

DIST=214 E

J B

P XF=489

/

YF=-502 3*

a m

i ZF=-137

g p

EDGE B

o nx=205 m

I )

MN=.124 E

g

)

A=17.2 j

B g

B=36.2 3

g C=55.2 8

i D=74.2 P

l B

E=93.2 1

F=112 x

Q=131 5

i H=159 8

I=169 J-1SS i

M, m

DETAIL OF QUSSET AREA ADD LOADS FROM UPPER STRUCTURI t

?

g 4

)

i

)m W

m m

m m

m m

m m

W m

M W

m m

m m

M i

m,,,

s.& -

59 ANSYS 4.88 JAN 8 1887 b

16:48844 0

POST 1 STRESS STEP =9999

\\

ITER =1 SI

?3 30 1

BOTTOM m

E F n c aG XV=-1 gS

,, D 8

• C YV=.5 23

{

d-H zu=1 dg H H_

g5 DIST=214 R

gg g

Mc D

F= 592 i

a E

k ZF=-137

'p EDGE g

u MX=298 g

3 3

I Cr" k0 9

E 3

MN=S.23 p

)

A=21 1

E h is k

B=49 x

%~'s 7

i 8

E=.2 F=116 s

G=135 d

H-154 9

jel I.173 m

W J"

9 i

k i DETAIL OF QUssET W A W Lonos FRom UPPER ST E M t

l

References

/Ref. 1/ "Ma ter,ial Evaluation Report for CASTOR V/21 Basket Material ",

General Nuclear Systems, Inc., January 1987.

/Ref. 2/ " Thermal Load Test with Modified Fuel Basket for the CASTOR V/21 Spent Fuel Cask", Virginia Power Corporation, December 1986 I

I I

I I

I I

I I

. I I

I I

I l

s,

APPENDIX 3 Material Specification for Radionox A-18 (X 2 Cr Ni B 1911)

I

- Fuel Basket (BS 05)

I l

• • • NOTE ***

The information provided in this document has been declared f

Proprietary by General Nuclear Systems, Inc., and is therefore not included in this Non-proprietary version of the i

CASTOR V/21 Borated Stainless Steel Basket Evaluation Peport.

I i

,-..,n-

I APPENDIX 8 Summary of Basket Dimension I

Peasurement Technique I

l I

• • • NOTE ***

The information provided in this document has been declared l

Proprietary by General Nuclear Systems, Inc., and is therefore not included in this Non-proprietary version of the CASTOR V/?1 Borated Stainless Steel Basket Evaluation Report.

I i

I

'I I

EVALUATION OF A CASTOR V/21 FUEL BASKET SUBJECTED TO A DRY SPENT FUEL STORAGE DEMONSTRATION TEST AT IDAHO NATIONAL ENGINEERING LABORATORY I

REVISION 2 January 1987 I

I I

General Nuclear Systems, Inc.

220 Stoneridge Drive I

Colunbia, South Carolina 29210 I

I

1.0 INTRODUCTION

1.1 REPORT OBJECTIVE A dry spent fuel storage cask, designated CASTOR V/21, was supplied by General Nuclear

Systems, Inc.

(GNSI) for an unlicensed demonstration test at Idaho National Engineering Laboratory (INEL).

This demonstration program is a cooperative effort between Virginia Power Company and the Department of Energy in accordance with the Nuclear Waste Policy Act of 1982.

The objective of this testing, in I

part, was to collect data for comparison with computer models and to expand the data base related to dry storage cask performance in I

support of NRC licensing activities for "at-reactor" dry storage cask installations.

The testing of the CASTOR V/21 was successfully performed during the summer of 1985.

During the post-test examination of the spent fuel, personnel at INEL observed that the CASTOR V/21 fuel basket showed slight indications of separation at certain joints in the basket structure.

The purpose of this report is to provide the results of the investigation conducted by GNSI I

concerning the cause of these observed indications.

1.2 CASTOR V/21 TEST CASK HISTORY GNSI, under contract to Virginia Power, supplied the CASTOR V/21 for testing in compliance with the Virginia Power specifications stated in NUS 2001 and the GNS quality assurance program in effect at the time of fabrication (IAEA QA-Standard).

The cask body was cast and machined in 1983 and the fuel basket was fabricated in June 1984.

The CASTOR V/21 test cask had certain unique features to support the collection of temperature data at INEL.

These special features were unique to the INEL test and are not relevant to the cask design described in the TSAR (Rev.1) which was submitted for NRC approval in early 1985.

A special primary lid with ten instrument l

penetrations was manufactured to accommodate thermocouple lances l

(0063W)

I used to measure temperatures at speci fic locations in the cask I

cavity.

This method of instrumentation required close attention to fabrication tolerances to ensure proper alignment between the thermocouple lances and the installed fuel assemblies, especially the fuel assembly guide tubes.

One important aspect of this l

tolerance consideration is the position of the fuel basket in relation to the primary lid.

This relative position can be controlled by the magnitude of the gap between the cask cavity wall and the outer diameter of the fuel basket.

The magnitude of this basket-to-cask wall gap for the test cask ranged from 0.8 mm to 1.5 I

m.

This gap size is determined from as-buil t cask and basket dimensions, since direct measurement of the gap was not required by the quality control program in effect at the time of fabrication of the test cask.

I 1.3 INEL TEST PROGRAM During July / August 1985, the cask was loaded with fuel assembifes from Surry Power Station.

The decay heat load of the assemblies I

used in this test ranged from 1.0 KW to 1.83 KW for a total loading of 28.4 KW which is far above the design heat load.

Figure 1-1 shows the configuration of the loaded cask.

The design heat load specified for the CASTOR V/21 in the TSAR is 0.9 KW average per fuel assembly with a power peaking factor of 1.1 yielding 1.0 KW maximum.

This corresponds to a total cask loading of 19.2 KW I

average and 21 KW maximum.

During September, temperature data was collected with the cask in five different conditions.

The cask was tested in the vertical and horizontal configurations with cavity I

atmospheres of helium, nitrogen, and vacuum.

The vacuum condition was tested with the cask only in the vertical orientation.

Heliun and ni trogen atmospheres were tested in both the vertical and horizontal orientations.

The cask was allowed to reach themal equilibrium in each case to permit steady state data collection.

Cask and fuel temperature data was collected from the thermocouple lances inserted into the cask cavity and from surface-mounted I

2 (0063W)

I thermocouples on the cask outer surface.

Details of the resulting temperature distributions are provided in Ref.1.

On September 30, the cask primary lid was removed to perform an inspection of selected fuel assemblies.

This process involved the removal and reinsertion of fuel assemblies in the basket fuel cavities while the fuel was visually inspected.

During this fuel inspection, indications of separation were observed between certain metal plates in the upper portion of the basket structure.

However, this situation had no effect on the ability to remove and insert fuel assent 311es during the fuel inspection.

Extensive photographic documentation of the visible portions of the basket I

structure was collected and studied for this report.

1.4 SUMARY OF INVESTIGATION Upon notification of the observations made at INEL, GNSI initiated an investigation to determine the cause and implement corrective action as appropriate.

The post-test condition of the basket was also reviewed with respect to basket function.

Based on various considerations including test temperature measurements, the I

demonstrated ability to remove / insert fuel assemblies, and the redundancy of the basket structure; it was concluded that the observed indications had no adverse effect on criticality sa fety, heat transfer capability, or basket structural integrity.

I Investigation of the test cask and test conditions concluded that constrained thermal expansion of the basket was the cause of the observed indications.

Analyses performed to support this conclusion are provided in more detail in this report.

I The conclusions reached in this evaluation were taken into consideration with the CASTOR V/21 cask design intended for licensed storage of spent fuel at connercial Independent Spent Fuel Storage Installations (ISFSI).

The Topical Sa fety Analysis Report (TSAR) which describes the CASTOR V/21 design for use under licensed (0063W)

I conditions identifies certain basket dimensions (and gaps) which will accomodate thermal expansion of the basket under limiting TSAR condi tion s.

The fabrication quality control program will verify the implementation of these gaps in the as-built condition.

The basket described in the TSAR has an improved design compared to the basket tested at INEL.

The structural basket analysis contained in Rev. 2A of the TSAR shows that the indications in the test basket will not occur under licensed operating conditions.

I I

l l

t i

I I

I (0063W)

I 0*

I

/M$03 T N g

'///////

1.11 10

'///////

"" / /

V27 A8

, V25 08 :

I

'1.83 21 1.83 11

'//////

?lllll/

'//////

/

l////h

!lll/l' llllll/

l

.l

///////

T11 g4 V15 D6

/ Y24 D5 l 1.11 2e V04 A5 V14 A6 1.79 20 g1,00 g;

1.00 3

,1.79 12 f

" ////// '///////

/////// '//////,_

~

~

I f

I II i

I I

i I

i ll l//////,

///////

///////

'///////,

T T09 D3 l

' T16 D2 s V05 Al l T12 A2 '

T07 A3 -

270* - [1.11 19 ', *h1.11 8 ',

1.02 1e

, 1.11 4 $

90, 1.11 13i J,l

~ ~ ~

s

/

~

1m 9//////

'//////

'/ /

///////

'///////

A I

I I I I

I N

I i

1

' //////

/////// ///////

/

///////

/

WM///

$V08 85 V11 86 V13 C6

'V12 C5 /

'1.79 18 l1.00 7/

B4 l

' 1.00 5l 1.79 14

'/////// //////}

'T13 I.

H '

7/i'//// '/>'i'/i'A

?///j///

"///////

'////H/

l

, V01 88 V09 C8 1.83 17-

~ " " " ;

, 1.83 15, l

'//////A

-T08 8k ? llllllA Fuel Assembly ID Fuel Tube ID Decay Heat (KW) 180*

X = loading sequence I

FIGURE 1-1 CASK LOADING MAP FOR INEL TEST (0063W)

2.0 BASKET GE0 METRY AT INEL Figure 2-1 provides detailed dimensions of the CASTOR V/21 test cask / basket which were used in the various post-test analyses of the observed basket condition.

Figure 2-2 is a quadrant of the basket which identifies terminology used in this report.

The outer diameter of the fuel basket barrel is nominally 1,524 m while the nominal diameter of the cask cavity of 1,529 m.

This implies a nominal gap (labeled as G1 on Figure 2-2) of 2.5 m.

Manufacturing documentation for the test cask at INEL indicates an as-built gap ranging between 0.8 and 1.5 m at room temperature.

The presence of this small gap size is needed to accomodate alignment and installation of thermocouple lances into the fuel assemblies.

It is determined by calculation that the unrestrained radial thermal expansion of the basket under the INEL test conditions (28.4 kW) is as high as 3.3 m.

A thermal stress analysis using a finite element model demonstrates that stresses in the affected basket joints increase rapidly when gap closure occurs and further thermal expansion is constrained.

I 6

(0063W)

O I

YMM$

t (18 P4 AMS)

}l ITEM 1 I

D l

,/

1 tvan s I

e w

u kf f

,Uf 1

- ~~

~

=l' l

% 41 w

/

==

r=

n

(,,,,

I am R

so -

I

._:_____..?

T j

ll p

1 1 u

Ty

. ____g e _

M a

M

\\rm..

k i

eseo j

zy I

=

{80 l

me.mman 6-=

kEb...

d' d I

910 m

199 I

FIGURE 2-1 FUEL BASKET ASSEMBLY GEOMETRY I

l (006 31)

I I

I CASK BODY GAP G1

_L l

^

p OUTER GRID I

_ /,

x

[

JOINT J2 I

I 3

.v

~

INNER GRID s

/

/

T lI

[

./

~

-f - - -l l

1 I

k

/_

l l

I i

l l

JOINT J1 l

1 BARREL l

I l_.______.__'

FIGURE 2-2 8ASKET GAP AND JOINT TERMIN0 LOGY (0063W)

3.0 MEASURED AND PREDICTED TEMPERATURE DISTRIBUTIONS Following completion of the thermal performance test at INEL and in response to the observed basket indications, GNSI cal culated cask temperature profiles based on INEL test conditions using the GNS thermal calculation computer code.

These calculations were performed to provide the basket ligament temperature data required as input to the basket thermal stress analysis.

Figure 3-1 shows the temperature profile calculated using the GNS thermal model based on the INEL test conditions for the case in which the cask cavity was evacuated.

Figure 3-2 provides a comparison of INEL-measured temperatures with calculated temperatures at corresponding locations.

This close agreement between measured and calculated temperatures justifies the use of the GNS thermal model to provide basket ligament temperature input for the thermal stress analysis.

I 9

(0063W)

0 ). 03. 46. 8). 04 Se le to Os. Se to te Se. 86 Os te le et 96 86 86 06

86. 86 St. We

• s. **. "

et. ee.

86. 83. St. 86 4)_ Se 46 4.s 6e ts. Ge to to. to ga to se le et 86

96. es. 86 SS 86 86 96._9 %

86 to et

86. 86. 96. 96 46 ' to. 69. 96. 93. ta. 97. 99.ata.it2.lta.ata. 681.666 ist.ast see see s ea.m. 8 62.16 7. 6 71. t 7 3. 4 75. 6 7 3. 3 7 3.-

36. 86. 86. 83 86

86. 91. 96. 884.184. 4 88.18 8.164. 4 t7. t 6 9.12 8, t22.123123.122 L19 123 163.181.198.196.201.244.261.286.266.

96. 16 86. 86 06

06. 93.888.186.812.'167.822.426.138 133.t36.138.139 139.133 135 133 481.198.247.214.219.223.223.223.225.

86. B6. 86. 86 06

06. 96. 4 0 4.112.124. 826. 4 32.118. t 42.14 6. tSt. tS2. t3 4 L5 4. l33 13e 432 693.212.222.238.235-139.246.241.24L.

86. 86. 86. 86 86 to. 94. 8 89.18 0.127. 8 35. 4 4 3.149.154.109. 6 6 3.166.168169. 4 68 164 167 288.223.233.243.248.252.234.255.2:4.

So. Be te se 86 06.Itt its 174 135.t44 1:2 16e 166 ISt. tit tie. net is3.its tai tse 220.237.247.254.264.263.265.266.265.

86 se es. se se. e.ie: si) 13t.ta:.it3.te?.tii.its its ice it; ist g, p, g,3 g94 232.247.257.264.269.273.275.276.273.

36q U.4Lt. L; lit 412. 6 4 3. 2 t;. 2 6 4 7 3 t. 2 3 9. 2 e 7. 213. 7 6 t. 2 6 4. 2e 2. 26 2. 26 i 23 2. 2 6 4 285 287 242.237.266.273.278.282.744.284.204.

86 168.182.122 138 137. 28 9. 2 36. 2 4. 2 6 2. 2 71. 214. 2 0 4. 7 6 7. 24 t. 2 9 4.19 6. 2 4 7 2 76. 26 3 238 226 232.266.274.281.286.289.291.292.298.

86 it1.124.139 LSt 8 64.238. 213.27 L.282.29 8.298.3L3.146.188.349.347.314 292.27s 244 232 263.274.202.20s.292.293.297.

297.

86 114.836.134'178 183.248. 273.28 7.299.344.3 LS. 3tt.323.324.324.323.119 386.292 238 144 278.283.208.294.19s.343.382 e 342.

86 121.847.569 187 78 3. 264. 28 7. 3t 2.3 6 3.322.37 0. 333. 336.337.317.135.132 318. 384 271 255 278.288.294.299.343.345.387.

.367.

86 128.839.884 284 222. 27s. 3t s.3 t e.325.334.34 t.14:. 34 7.349.348.346.343 329.316 284 267 286.294.299.383.386.369.368.31s.318.

84 43;.178.197 224 239.291.382.325.134.144.358.354.317.358.138.356.112 339.326 296 277 192.299.303.366.389.348.312.382.312.

It. i43.Il2.211 234 234. 38 2. 32 3. 331. 343. 36 3.159. 36 3. 36t.36 6.36 6.364.36t 3 4 s. 33 337 288 7" 38 3. 38 6. 38'. 3 t l 117. 313. 313 313 06 632.194.22 249 769. 3 t). 312. 344.133.361.16 6.31 t.372.37 3.37 3 17 5.367 335. 34 4 367 34 3 n! 3e' 3es.sie sii 3nL 317.317 2;?

, 86.65.2t9.239 263 213.323.348.3 1.368.367.372.376.378.379.379.

'.373 362.333 327 386 li' 37L 374. 316. 32s 37 e 3 3 6 334 3361

6. 312. 3 4 4. 35 4. 36 6. 37 2. 37 7. 3s t. 3 8 3.3 8 3.3 8 3 b"3 0 '

A.84.22e.235 177 29 3?S 365.338 336 338 :37 337 34e.347.344 3st.34e.34e 3se 484.239.248.278 29L Sig. 34t. 354. 36 3.37t.376.3 t l.184.3st.3s7.3s7.M 182 373.364 344 348 332.337.36e. 3 3. 3es. 3e). 368. 368.3 s.

226.248.269.298 3t* 119.347.3 9.367.374.379.333.516.118 349.389.388.14 377.369 3:2 347 348.367.378.373.376.377.378.378.37s.

133.274.168.363 3 t t 3 31. 353. 363.17t.37 6.3t l.185.18 8.349.19 t.19 t.189.387 34 8. 3 73 359 354 347.374.378.381.383.305.385.386.383.

g fit 7'i 1ta it 13t_ta

!!p Y 31, 33, Sti.3st 3p 3pe 3er 3eg 3ge 3t' 301.376 366 366 374.300.384.387.394.391.392.392.392.

a 29e ;te tit

?

342

, t

!.e 1e; 3't 3,e :19

• s.

Na is. !s. 3s. It* 14 37' 37e 372 349 388.387.398.393.396.397.398.398.398.

332.34C.347.333 333

t9,37 2. 318,3 3 ;,38 7,34 g, g e s,3t e,;9 7.19 4.19 6.34 7. 34 e 191. 38 7 379 376 387.392.196.198.481.482.483.483.403.

348.354.339.363 36: 365. 373.3se. 39t. 39 4. 397,4tc.4t 2.4t3.4te.4t4.4t3.4t? 397.392 st2 3se 391.397.488.463.455.467.484.48s.40s.

3 8.343.367.37t 37 374.363.198.394.348.4tt.4tt.486.487.ett.4ts.4t7.4t6 481.397 387 3e4.396.481.484.407.449.416.411.442.486.

346.378.374.376 377 3 7 4. 3 s t. 39 4.19 8. st2.4tt. 4ta. 4 t e. 418.4 t2. 412. 4 61.4 t 8 4 8 3. 481 39 3ss 399.464.487.414.482.484.4l4.485.414 372.37e.379.30; 1st 378. 3 9 6.197. 4t n.4 t t.4 tt. 418. 4 3. 4 t 4.4 63. 413. 4 6 4.4 61488. 40 4 394 392 462.467.418.483.413.486.417.

.487.

377.381.383 38*

18 5 3 81.19 4. 4 8 8. 4 0 4. 4 8 8. 411. 413. 41). 416. 417. 4 t ?.4 L 7.416 414.487 398 393 483.418.413*413.4II.438 43'-

Ol'.

18t.384.30s.348 3t?.344.196.4L2.4tt 4tt.413.4tS.4t7.4t8.4t9.459.4t9.41 413.469 481 398,467.411.444.416.418.469.424.

.425.

383.386.318.369 384 3tt. 39 7.483.4t ?. 4 s t.4 8 4.4 L 6. 4 s t. 428.421 421.428.4 t9 413. 412 453 468 449.483.413.487.419.428.421.421.421, 343.387.309.39t 19t 3 s e. 3 e s. 4 t t. 4 t s. 412. 4 t t. 4 t ?. 4 t 9. 4 21. 42 6. 4 22. 4 2 5. 4 7 t. 4 8 7. 413 4tt 483 418.414.416 418.469.428.421.428.421.

303.387.304.399 398 336.398.444.4t8.412.4L3.467.419.421 422.422

.421 418.413 467 465 eis 4 3.413.417.4 s eig 47s.47s 47s 30s.301.318.389 109 1st.197.et2.st7.4Lt.4 4.st?,4tt.47t.421.422 421 488.415 448 468

• is '87 4'3 4 4:1 *!6 ' * *

4'4-379.342.345.357 347 334.196.4t?.sts.4tt.413.416.4Lt.419.42t.421.. 421 419.414 418 418 ell.433.464 483.446.sia 4 7.467.467 174.370 352.384 384 33 2,3 94,4 gg,4g e,4 g g. 4 t t. 4 6 4. 4 8 6. 4 6 9. 4 4 9.428. 421.47t 418. 4 8 6 448 til 412.464.453.466.417.488.*18 418.

I 344.37;.377.38C 38; 379.396.197.4tt.4tg.4tg.4gt.463,43.437,4 g,413.4t9 417.415 ell 412 414.413.416.487.418.419.419. L 489.

1*l I64 364 191 31e 19e !t' to? 3et sei efe sti ett 4); et) 43g 41* ale 41*.41t til 413 413.416.410.419.426.420.425.

.425.

3?r 3ae it; 3:e 3e; 3 g 3,g 7 s, - :g. ;e3 3.,

geG e: ar3 ett.at?.4g3 ait all 411 481 414 416.418.419 471.427 472.423.423 473.

'I 3 l let 3ei !?e 1st tee 3.

g; 34a 3 e r, 14; 3e.. Mv att set art art a 9 417 4L2.414 41a 413.4;e.e'i.477 47e 47e 875 4 *- eM 333 3* 2. 3at. 367. 374. 38 4,34 7,39 6. 39e. 39 9.4(2.4(3 og 7,4(g,a,t. 4 t g ett ett 434.450.436 418 434.427.429.436.433*434 434.434.434 33' 337.367.374.364.387.392,397.4t6.4te.4L7.4tt.4tg.4L1,4ge.413 412 451 433.416.480 428 427.436, 435.437.438.439.439.439 341 36 t. 378,37 8. 3g e. 39t,39 3,4 g t. 4 ge,4 87. 4 g g,4 g 3 o g e 4 LS.4 L 6. 4 61 4!) 414 416.418.419 428 429.433, 438.448.441.441.442.441.

342 3 3. 373. 381 - 38 7.19 3. 39 s. 4 t 3. 4 t ?. 41t. 413.4 65. 4 a. 4 5 7. 4 :3.4L7 454 4:5 417.419.421 422 438.433.

.44s.

.443.444.444.444, 344 363.376.383.394.396.4tt.4t!.4ft.462.413.417.4L8.469.419.410 4: 3 4L4 418.428.422 424 433.437.49e.442 4 443.444.446.446.

9 341 3e' 377,3gt.3g g,397,4g3,4gt.4 L L 4 64.4 L6.4 L9 4 L9 42t.47L.4L9 4 66 4 e e.419.42s. 432 424 434.430.441 443.

.446.447.447.447.

346 360.378.383.392.398.4(3.4t8.468.464.417.419.42f 421.424.47t 4e6, 447.489.421.423 42$ 434.439.441.444.446.447.448.448.448.

les 368. 378. 3se.192. 398. et 3. ett 4 6 2.4 45.4 6 7.4 tt. 42t. 42 8. 42 8. 47t 46 4,7.419.421.423 423 434.439.442 444.446.447.448.448.448.

34e, 36 8. 3 78. 3st,39 2. 39 4. 4tl. 48 0. 4 L t. 4 4 4.417.419. 428. 421.42 6.424 tit. 4 ~. 7.,419. 42 6. 4 2 3,,4 2 5 4 3 4. 4 3 9. 4 41 4 4 4. 4 4 4. 4 4 7. 4 4 8. 4 4 8. 4 4 8.

o 28.4 KW (See Fig.1-1) l

[

//7 i

dC h/

o 23*C MBIENT

/

_j_

i f

L e.

/

o YAtllU'4

/

/

o NO SOLAR HEATING

/

~

~~"

.m o

CASK VERTICAL

/

1.-

==

FIGURE 3-1

/

O BASKET TEMPERATURE PROFILE DURING VACUUM TEST CONDITIONS I

10 (0063W)

h.

J

/f ni.c THERM 0 COUPLE

.3 V

v LOCATIONS i

EN i

/EM =

17 I

I I I I I

!615 %.::' M VA G

iii i [y/jy/}

\\yg, I

si 4

~

~

I VACUUM HELIUM I

LOCATION MEASURED CALCULATED

• MEASURED CALCULATED

• Al-l 41 4 436 349 373 Al-3 420 444 352 379 E

399 41 9

,330 361 l

A4-5 410 41 9 337 355 l

AS-3 406 422 340 356 l

A7-5 321 303 262 244 A8-3 386 381 31 7 309 B7-4 323 303 263 244 C5-2 406 422 338 356 I

• Calculated using the G'15 thernal computer code FIGURE 3-2 COMPARISON OF INEL MEASURED AND CALCULATED MAXIMUM TEMPERATURES (C*)

I (0063W)

I

4. 0 FINITE ELEMENT ANALYSIS OF FUEL BASKET THERMAL STRESSES 4.1 PHYSICAL ARRANGEMENT The fuel basket of the CASTOR V/21 test cask (Figure 2-1 ) is composed of flat plates of different thickness ranging from 5 mm to 20 un.

These plates are welded to each other by stitch welds to form cell s for storage of the fuel assenblies.

The basket is encompassed by a 10-mm-thick cylindrical shell (referred to in this report as the "b arrel ").

The basket-barrel assenbly is placed inside of the cask cavity.

I The arrangement of the plates in the fuel basket results in joining I

of 10 mm plates to 20 mm plates and 5 mm plates to 10 mm plates.

Under thermal expansion l oadin g, such joints provide stress discontinuities.

Al so, stress concentration will result at these joints since the plates are eccentrically connected.

The test cask did show dislocatiens of the joint and weld cracking at precisely these loca tions.

Hen ce, detail ed analysis is performed of two representative joints where indications are observed, referred to in this report as "J1" and "J2" (Figure 2-2).

4.2 COMPUTER MODELING A two-dimensional ANSYS finite element model was created to evaluate the thermal loading of the as-built fuel basket under the INEL test condi tions.

The thermal loading on the ba sk et-barrel assembly arose due to non-uniform heating of the basket.

When fuel assenblies are placed in the cells, the heat dissipation to the cask wall causes a temperature gradient in the basket.

This, in turn, causes uneven

!I thermal growth in various components, giving rise to stresses.

l l

l I

1?

(0063W)

I Another major source of stress occurs when interaction between basket co'9ponents results in restraining thermal expansion.

The basket geometry is as described in Section

2. 0.

The model ccisisted of a

1/8 section, taking advantage of the basket symetry.

Detailed models of joints J1 and J2 were also evaluated.

The n.adel also includes the effect of temperature on material properties important to thermal expansion, such as modulus of elasticity and coefficient of thermal expansion.

The temperature distribution for the test conditions (28.4 KW, I

anbient temperature 23*C, no solar heating) was determined using the GNS proprietary heat transfer code which was shown in Section 3.0 to give good agreement with measured temperatures.

4.3 FINITE ELEMENf ANALYSIS RESULTS I

The calculated thermal stresses for the INEL cask and tes t conditions are summarized below.

A cap of 2.5 mm is conservatively used to determine stresses which led to the observed indications at I

INEL.

Manufacturing records indicate that this gap is more likely less than 1.5 m.

Calculations performed for various gap siz?s showed that joint I

stresses increased rapidly for reducing gap sizes when the initial gap size is too small to permit unrestrained expansion.

The resulting thermal stresses calculated for a 2.5 m gap size are as follows:

JOINT J1 163,900 JOINT J2 114,200 (near gusset)

I JOINT J2 138,100 (away from gusset) l l

l I

13 (0063W) i

These calculated stresses substantially exceed the yield and ultimate strengths of the basket materials.

Material testing has demonstrated that the yield strength of the basket materials is nominally 30,000 psi to 38,000 psi for the basket temperature of interes t.

Similarly, the ultimate strength is nominally 67,000 psi to 72,000 psi.

Even though the thermal stresses are considered secondary, the finite element analysis demonstrates the importance of sizing gaps G1 and G2 to provide for unrestrained thermal expansion of the basket members.

This finite element analysis methodology can also be used to determine recomended gap sizes to accomodate thermal expansion under licensed TSAR conditions.

The following section provides further analyses supporting the temperature dependent failure mechanism which led to the observed basket indications at INEL.

I I

l I

i i

14 (006W)

5.0 FIILURE MODE EVALUATION OF INEL TEST BASKET 5.1 EVALUATION OF JOINT J1 The terminology for joint J1 and gap G1 is as previously defined in Figure 2-2.

Joint J1 connects four plates, two 20-mm-thick plates and two 10-mm-thick plates.

The plate configuration is shown in Figure 5-1.

The test basket had a small gap at location G1 which closed during heat-up.

Inspection records indicate that this gap for the INEL test cask ranged from approximately 0.8 m (.031 inches) to 1.5 m (.059 inches) at the cold condition.

The larger gap will be assumed in this failure mode analysis for conservatism.

The radial thermal expansion of the plates can be approximated using a weighted average summation of plate length and temperature (see Section 5.2).

The free thermal expansion of the plates indicated in Figure 5-1 was calculated to be 0.159 inches (4 m) based upon a calculated average temperature of 634*F.

This temperature is based upon calculated temperatures for the vacuum test condition.

Because of the symetry of the basket, the plates shown can be assumed to behave as a one-dimensional spar which expands radially outward from the center line.

It will be shown that the radial expansion of the plate will be large enough to cause the gap between the barrel and the cask wall to close and thereby restrain the thermal expansion of the plates.

The barrel temperature will be considerably lower than I

the average plate temperature and thus the plate will be restrained from expanding by the barrel.

However, the barrel stiffness is much less than the axial stiffness of the plates in compression.

The barrel sti f fness was determined from a two-dimensional finite element model of the barrel with point loads app 1ted at each point where the 20 mm plates weld to the barrel. The barrel stiffness, KD, was found to be 4.654 Ed Ib/ inch (per unit axial length along the barrel).

The conbined stiffness of the 10 m plate in series with l

15 i

(0063W)

L

the 20 m plate was found to be 6.95ES lb/ inch (per unit axial length).

Figure 5-? depicts the relationship between the increase in temperature of the fuel basket and the stress and the deflection in the plates.

The compressive axial forca in the plates increases as the basket heats up.

The rate of inct case, K, is the combined I

spring stiffness of the barrel and the plates acting as spring in series.

The gap between the barrel and cask wall closes at approximately 312*F (Point ?).

Once this gap closes, the force in the plates increases much more rapidly since the barrel cannot move radially outward and the thermal strains in the plate must he accomodated by compressing the plates.

(Note that these values are all plotted on the basis of a unit length in the axial direction of the fuel bask et. )

The load in t plate increases to Point 3 at which point the 10 mm plate will yield.

It is noted that the PO m plate is still well bel ow yiel d since it has twice the cr'os s-sec tional area of the 10 mm plate.

Once the 10 m plate 6

yields (at approximately 355'F), plastic strain will be incurred as the temperature increases, thus, from Point 3 to Point 4,

the thermal expansion of the plate is accommodated by inelastic compression of the 10 mm plate or the weld which joins the plates.

Upon cool-down, the basket will contract and the high com;*essive loads in the plates will decrease (Point 4 to 5) and the 10 nn plate will experience tensile loading when the contraction of the fuel I

basket due to decreasing temperature reaches the point that the permanent deformation of the 10 m plate causes a stress reversal.

As the basket cools down to room temperature, the tensile load in the 10 m plate will exceed the load to fail the weld.

A finite l

element model of the basket was developed which was used to determine the residual stresses due to the inelastic deformation of the plate.

It was determined that the tensile load developed in the 10 m plate at 70*F was 4?P4 pounds.

This assumes that the full strength of the 10 mm plate can he carried in tension by the weld I

area.

The weld joint is shown schematically in Figure 53.

Tbc weld designated as $N8 is an intermittent weld, 5 m on the side, and is g

welded for 100 m with a ?S7 emspace between welds, thus, the weld I

16 (0063W)

only covers 28 percent of the plate along the length.

The throat of the weld is (.707) (5 mm) = 3.53 m =.139 inches thick at the minimum point.

The force required to cause tensile yielding at the weld is:

I r, =

.,. s (29,100) (.139) (.98) = 1133 lbs/ inch

=

The maximum tensile load that the weld could sustain if it did not fail during cool-down is:

I F

"ub max (67,000) (.139)(.28) = 2,608 lbs/ inch

=

Thus, the welds will break as the basket cools down from the test temperature.

The strain in the weld can be estimated by assuming that the plastic strain incurred during heat-up will result in a gap between the plates at room temperature.

The plastic deformation of the 10 mm I

plate can be found from Figure 5-2 as.159

.070 =.089 inches.

The strain in the wold would thus be:

.0 454 in/in t =

=

=

I This strain (45.4%) is well above the acceptance criteria of it on membrane strain which has been defined in the Material Evaluation Rerport.

Actual failure strains for welded Radionox are typically 14 I

to IA percent as determined by material testing which also is documen ted in the Material Evaluation Report.

Mditionally, the strain would not be uniform over the leg of the weld but rather be l

concentrated at the throat area.

I 17 (006 N)

h I

Figure 5-2 can also be used to draw two additional conclusions.

First, it can he seen that if the gap between the barrel and the cask wall is not large enough to accomodate the expansion of the fuel basket and the gap closes as shown at Point 2,

then the compressive load in the plates increases very rapidly.

Even if the thickness of the 10 mm plate is increased, the load would quickly exceed the yield point of the plate material (i.e., Point 3').

Clearly, increasing the plate size will not resolve the problem of weld cracking.

Secondly, if the gap is large enough to preclude '

closing during the test condition, the load in the plate increases along the Ifne 2-2' in Figure 5-2.

As can be seen from the figure, the load in the plate remains well below the yield load and no inelastic strain would be incurred and the basket plates would cycle elastica 11y, well below the yield stress for the material.

+

5.2 EVALUATION OF JOINT J2 i

The configuration of joint J2 and surrounding plates is shown in l

Figure 5-4 This joint connects four plates, two of which are 10 mm-thick, and the others 5 m thick.

Since the fuel basket is symetric, only one quarter of the total plan view is shown.

The joint detail is shown in Figure 5-5.

A good approximation of the thermal expansion of the basket may be obtained as follows.

Tne most severe test condition was the vacuum test, therefore, it will be assumed that the failure occurred as a result of this test and the temperature distribution used in this evaluation will be the vacuum condi tion.

The averace temperature of a length of plate which extends from the horizontal center line to the barrel (see Figure 5-4) may be determined from the following equation:

I N

(ERJ 5.1)

I T L Where: N = Number of plates I

j j

"I T,

L = Length of plate

=

j E

Lj i =1 Tj = Plate average temperatures I

I 18 (0063W)

6

/,-.

Using the lengths and temperatures shown on Figure 5-4, the average plate temperature was found to be 590*F.

The thermal expansion of the equivalent plate will then be:

(9.6E-6) (24.65) (590-70) al AT at

=

=

I p

~

0.123 inches (3.1 mm)

=

As-built records for the INEL test basket indicate that there was no gap present at the location defined as G 2.

Thus, the thermal expansion of the plates will be restrained.

The thermal strains I

must be accomodated either by compression of the plates, bending the 20 mm plate which forms a part of the basket (see Figure 5-6) or

~

,by buckling the plates.

Compression of the plate will develop an axial stress.

This can be calculated as:

(25E6) (9.6E-6) (590-70)

E oA T a =

=

124,800 psi I

=

This is much greater than the yield stress of the material.

The maximum load that can be developed by the 5 m (.196 inch) plate is F

y.A (29,100) (.196)

=

max 5,704 lbs

=

The strength of the 20 m plate may be assessed using a limit

. analysis.

Figure 5-6 shows the geometry and assumed fail ure mechanism.

It can be shown that the limit load, P, is:

! 2 o

P (EON 5.2)

=

tp 19 (006'M)

2 t

Limit moment for a beam of thickness t Where:

M=

y

=

o 4

and yield stress, oy 29,100 psi o

=

y Evaluating the above equation resul ts in a limit load of 5,541 lbs/ inch.

This load is very close to the load to cause yielding in the 5 m pl ate.

This indicates that plate buckling should be checked as a mode of failure.

The plate buckling model is shown in Figure 5-7.

Roark (Re f.

?)

gives the formula for plate buckling for this geometry as:

E

( )2 (EON 5.3) c' = K Where:

t = Plate thickness = 5 m =.196 in.

b = Plate length = 52" (the distance between gusset plates)

K = 22.? (Ref. 2, Table 35)

E = Youngs Modulus Thus, evaluating, find o' = 9,704 psi This assumes that the edges of the plate are simply supported.

Since the welds are intermittent fillet welds, the assumption of simply supported edges is justi fi ed.

Further, any initial imperfection in the plate as well as eccentrici ty in the line-of-action of the loading will reduce the theoretical buckling load.

It is therefore concluded that the plate will buckle well before it yields in compression or fails the 20 m support plate.

(0063W)

/

The 5 m plate will buckle due to the thermal expansion.

If it is assumed that the deformed shape of the plate may be approximated by a sine curve, it can be shown that the amplitude of the sine curve will be approximately.68 inches for an expansion of.123 inches in length of the plate (see Figure 5-8).

This will induce a rotation of approximately 13* at the ends of the plate.

The plastic strain that will be induced in the weld by a rotation of the plate was determined from a finite element analysis of this joint.

The finite element model is shown in Figure 5-9.

The ANSYS computer program was used to evaluate the plastic strain in the weld due to an imposed deflection of the plate as shown in Figure 5-9.

The weld area was modeled using two-dimensional plane strain el emen ts.

The stress-strain properties used in the analysis were those of 1.4541 material (U.S. equivalent AISI 321), which is the basket material used in the test articl e.

The results of the analysis indicate that very small rotation angles produce considerable plastic strains in the wel d.

The highest strains occurred in the fillet area of the weld as indicated in Figure 5-10.

The maximum equivalent plastic strain is plotted in Figure 5-11 as a function of rotation angle.

As can be seen from this figure, the maximum or peak strain is approximately 12 percent for a rotation angle of 5*.

The buckling evaluation indicates that the rotation angle will be considerably larger than 5* and may be as large as 13*.

Failure strains for Radionox weld material are typically 14 to 18 percent.

For a rotation of 13*, the peak strain will be on the order of 31 percent, almost twice the failure strain.

A single cycle of strain of this magnitude would be sufficient to cause cracking in the weld.

It is noted that if the Gap G2 is large enough to prevent contact, then the axial load on the 5 mm plate is greatly reduced and no buckling of the plate would be expected.

21 (0063W)

I I

I y -'

I I

I I

[

P CASK WALL GAP KB h

v I

\\

24.87"

\\

I 20MM +

i

+

10MM + +

4.52" 77D7 i

\\__

.1M sz,,..

.....,M.

MODEL I

~

FIGURE 5-1 LOCATION OF PLATES IN BASKET AND SIMPLIFIED MODEL 22 I

(0063W)

M

-oo i

m M

FREE THERMAL EXPANSION (INCHES) r-o B

0.0

.060.070

.159 am I

.l l

g n

i g

p.. :

.i

.j.

q. _- [ l

_4 1l i

i i

.i i

a

+

N t.

.;. t..

g

.3'.....

m z

...g l i_,

,/

t*

z 20,000

_i_.-

_1 4

i

/

w z.

L. s,

$l m

q /.

i 5

E I

i Ul g

g

, _.f 9

a

,2 I

N w

i i

/-

..eg l

e s

i l

d A

i o m

.j l./;. i. ;

2._.

m S

30 m,

,/

10MM PLATE' YIELDS

'l i

g

[

10,000 m

ym r

3 m

{

. __. j

i. Lg

._j.

l o

4 20 a

,,p gm 6E l

I j _.t j

.! a*_.

i

.,i a

z o,

4 x

.l l..!,.

_.__.4 t

1 E 5,000

- j

. l..-....

i 2_' to m

q z

o j

'.y

__._.4 i

.. ul.

i U

..y...

t I

i m

m

. :..p. L.

g m

i __..

.2_..

1.

1.

g M

l I

-g 2

i i

1 e

O m

m o

1 l

l

[

j i _i._

n.

o

..j i

g c

,. __.i j

l

.l l

, _f...' -

g

.--.l l

l e

. __ {.._+.__., 5j

_{..i..-.

I _.

?

w g

z

_. _ 4 _

q_.

l m

S 5,000 o

.. [

j 6

l m

m I

l j

i j_ j m

z i

o, w

_ _ t._ u. _ __

t

.___.L.p

.__4 t

t _.

i._...

g 7

E-*

l j

l l

l l

{

i j.

j l

i i

in,

I t

o s'

i i

i c

i 1

u o

0 70 100 200 300 400 500 600 634 700 U

AVERAGE TEMPERATURE OF PLATES F

I I

I I

A

.089 *

+

t M

Y

!l 20MM 10MM

> F

+

I i

5 N 12 x 100 (257) (Typical)

SNB lI i

'I I

I I

FIGURE 5-3 FAILURE MECHANISM FOR JOINT J1

'I 24 j

l I

GAP G1 I

GAP G2 86 C

/

h JOINT J2

+

+

SMM I

f 9.06" n

204 C I

v v

l w

s.

+

+-10MM s

9.06" e 387 C 365 C f

p

[_. _,

2,83"

/

4 j

I i

(

l h

l

/

h 20MM k

4'52" l

344 c

' 415oc V

O L JOINT J1 LIMIT ANALYSIS I

l FIGURE 5-4 TEMPERATURES (C*) AND DIMENSIONS RELATING TO GAP G2 FOR VACUUM TEST CONDITIONS 25

.I (0063W) l

I I

/'

I a

I l

j I

I I

?

I

/

I.

n' HORIZONTAL PLATE (20MM)

I

~

+

+- SMM I

SMM l

)

I

?

~-

10MM I

• 10MM I

I FIGURE 5-5 JOINT J2 DETAIL 26 I

(0063W)

E i

I I

I I

U P = Thermal Expansion Load I

20MM

)

t I

I s

Horizontal (

54MM %

I 230MM I

+- ty :::

1 2 4

/

OPlastic Hinge 0

6 0 +0 1

2 1

I FIGURE 5-6 ASSUMED FAILURE MECHANISM FOR 20 MM PLATE IN FUEL BASKET E

.?7 E

(0063W)

I I

I h

9"

-+

a

=

/b 172

=

-+

<-- o use a/b =

.2 (closest tabulated value g %

.196 I

t

=

SMM

=

b b/t 52/.196 = 265 >>10

=

j

. meets limits of formula

}r

+a+

FIGURE 5-7 PLATE BUCKLING MODEL 9.05" V

.6 8 "-

i 3

+

6.

6 =.123" FIGURE 5-8 PLATE BUCKLED SHAPE as I

28 (0063W)

I m>E NW W

• e8 E

I Tm W

GB

.* v.J m e5

• NW

v. M m m6

.e LA

> 4 m we e ee I

W-ek

.* F I EE m

ome e CEmO 3 M la. La.

C E

.NOX>

I E

I 5

U<

% \\\\\\ \\\\ \\S \\\\\\ \\ \\\\\\\\ \\\\ \\ \\\\\\\\\\\\\\

I

{

S O

I

/ / / /

2 x

x n

9 N

p I

\\

x x

\\

/

I x

/

\\

/

\\

/

N

/

N I

k N

/

7

/

/

/

~

?

N I

/

s E

I e.e O

7 l

\\\\ \\\\\\ \\\\\\\\ \\\\\\

um C

sm e

FIGURE 5-9 FINITE ELEMENT MODEL WITH B0UNDARY CONDITIONS I

29 (0063W)

I N ta

.J e

I ee L

m TS m

M M FU wMM SWwe e MA T M ER S M EA F **

2m o ** m I

> we T -e e e e

e ee e 4A e 9-L E

• • D-t

XC E IK ** m W W e me eCO C L 9 O D- >

D M h. k. E m C

LMM NOX>QQ I

I I

ZO I

M*

/ / / /

O D-I N

Z 9

M N

j l

l I

L g

s x

~

K M

l i

I N

i O

E

[

N /

s e

O i

i E

~

2",

g i

J H

e D-M E

M N

O A

• 3 P-tal M

m C

en e4 I

FIGURE 5-10 LOCATION OF HIGHEST STRAIN (0063W) 1.. -.

I

.12 I

i I

l I /

I f

t

.10 E',=

(i)

O

)

K

,ll g

l l

e l

t

/ i l

1 xj i

I%

i l

i i

h

.08 3

I l

/,

l e

Im I

U 5

}

I Im5

.06 l

l l

6 1

l I

l I

i e

1 z

4 I

w l

Ay 4

i I

w O

i o

.04 l

W i

I i

l 7_

Ii i,,

i I

.02

.- 7 4

I

=

I

_._ 7 i

i i

O I

O 1

2 3

4 5

o 0,

ROTATION ANGLE (DEGREES)

I FIGURE 5-11 MAXIMlN EQUIVALENT PLASTIC STRAIN (0063W)

REFERENCES 1.

Letter October 11, 1985, G. H. Beeman (Battelle), to M. L. Smith, (Virginia Power),

Subject:

Test Data for GNS V/21 Cask.

2.

Roark, R. J., and Young, Warren C., " Formulas for Stress and Strain",

Fif th Edition, McGraw-Hill Book Company, New York, NY.

I l l I

4 (0063W)

.