ML20069J726

Brief Re Consideration of LOCA in Design Criteria for Pipe Supports.Loca Conditions Must Be Considered in Design Criteria.Certificate of Svc Encl

ML20069J726
Person / Time
Site:	Comanche Peak
Issue date:	04/20/1983
From:	Ellis J Citizens Association for Sound Energy
To:	Atomic Safety and Licensing Board Panel
References
NUDOCS 8304260058
Download: ML20069J726 (40)

Text

9 1

T s

4/20/83

' UNITED STATES OF AfiERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATONIC. SAFETY AND LICENSl'NG B0ARD:

'I In the Matter of APPLICATION OF TEXAS UTILITIES I

Docket Nos. 50-445 GENERATING COMPANY, ET AL. FOR and 50-446 AN OPERATING' LICENSE FOR COMANCHE PEAK STEAM ELECTRIC g

~

e STATION UNITS #1 AND #2 q

(CPSES) g g

p z-e#

d TABLE OF CONTENTS 99.

CASE'S BRIEF 4

E s

REGARDING CONSIDERATION OF L

~

IN DESIGN CRITERIA FOR PIPE SUPP0 O

3

, Page BACKGROUND 1

DISCUSSION.

2 Regulations, Codes, etc., require the considerati~on of LO'6A conditions in the design criteria for pipe supports _

2

~'

10 CFR Part 50, Appendix A - GENERAL DESIGN.C[ITERI[.

FOR NUCLEAR POWER PLANTS 2

~

Regulatory Guide 1.124, " Service Limits and Loading '

Combinations for Class 1 Linear-Type Component Supports,"

Revision 1, January 1978

~_

5 l

Each pipe support must be considered in two ways.:

(1) as though it were supporting the item which was involved in the faulted load of a LOCA; and (2) as though it were supporting the item which was not invol.ved with the faulted load but receives the effects of the LOCA 8

i Thennal stresses vs. thermal expansion stresses "(or~

constraint of free-end displacement) 9 Appendix F of ASME Section III has been misused by both Applicants and the NRC Staff 13 What must be done to assure that all i.tems important to safety will remain operable during a LOCA (and reoperable following a LOCA), and to mitigate the consequences of a LOCA 19 Classes 2 and 3 items must also be considered -

20 8304260058 830420 PDR ADOCK 05000445 6

PDR v

L

p:

s TABLE OF CONTENTS-

~

Page 2 i

]

P_ ate.

The 11/20/81 Denton Memo further supports CASE's

' {

21

- positions

~

~

Applicants have not considered certain'necessary ingredients in their analyses 23 LOCA themal expansion effects should be considered

~

on concrete inserts, the bolts which screw into the inserts, and the steel used for the pipe support i.n the design of pipe supports and associated concrete _.

28 anchors (including Richmond inserts)

IN CONCLUSION 34 y

LOCA conditions must be considered in the design criteria for pipe supports.

Further, each item important to safety-should be analyzed in two ways: (1) as though it were involved in the faulted load of a LOCA; and.(2)~ as though_

it were not involved in the faulted load of 'a LOCA but received the effects of the LOCA, including the c'onstrainT' of free-end displacement (or themal expansion stresses)

,, resulting from the increased temperature due to the.LOCA._ '

35 m

e

_m 6

ah p

S e

9 4

4D e

e

[

s

~

4/20/83 o

~

' UNITED STATES OF AtiERICA

~ '

NUCLEAR REGULATORY COMMISSION '

BEFQRE THE ATOMIC SAFETY AND LICENSING BOARD

~

In the. Matter of APPLICATION OF TEXA$ UTILITIES I

Docket Nos". 50-445-I GENERATING COMPANY, ET AL. FOR AN OPERATING LICENSE FOR

. and.50-446 COMANCHE PEAK STEAM ELECTRIC

~ STATION UNITS #1 AND #2 (CPSES)

I i

CASE'S BRIEF REGARDING CONSIDERATION OF LOCA IN DESIGN CRITERIA FOR PIPE SUPPORTS -

Pursuant to the Licensing Board's April 7,1983, Order during the conference call with all parties, CASE (Citizens Association for -Sound EnergyT, Intervenor herein, hereby files this, its Brief Regarding Consideration of LOCA in the Design Criteria for Pipe Supports.

BACKGROUND

~

~ '

~

On Thursday, April 7,1983, a telephone conference call 'as initiated by the Licensing Board with all parties. During that call, the Board ordered that all parties file briefs "in order to provide a correct in,terpretation of the application of the Comission's regulations to wh' ether or not LOCA"(loss-of-coolant accident)" conditions must be considered 'in the design criteria for pipe supports." The Board Chairman stated that in doing that, he would like to have "a logical discussion of the relationship between the different regulatory materials including the design criteria, the standard review

. plan, the staff g'uidance, the staff practice, and applicable industry codes."

O e

e

• -w

---

w

'-=,-,w w

,e rgw-

-g w-W w

rr-

-v-a-

3 av--

n,-r4

,t

'w-

o y

DISCUSSION

• /

The inclusion of LUCA in the design criteria for pipe supports is one of the underlying issues of the concerns of CASE witnesses Mark Walsh and Jack Doyle.

Indeed, it was because Messrs. Walsh~and Doyle were instructed

~

to discontinue including LOCA conditions in their STRUDL (Structural Design L'anguage computer program) calculations (in addition to their oker concerns) that they r signed their positions at Comanche Peak.

(See Messrs. Walsh and Doyle's testimonies.)

Regulations, Codes, etc., require the consideration of LdCA co'nditions in the design criteria for pipe supports.

10 CFR Part 50, Appendix A - GENERAL DESIGN TRITERIA FOR fiUCLEAR POWER PLANTS

- - There are several portions of Appendix A which are_ pert'inent to the

~

subject of'the consideration of LOCA conditions. For instance:

INTRODUCTION

" Pursuant to the provisions of 50.34, an application for a construction pennit must include the principal design criteria for a proposed facility. The principal design criteria establish the necessary design, fabrication, construction, testing, and perfonnance r'equirements for structures, systems, and components important to ' safety; that,is, structures, systems, and components that provide reasonable assurance that the facility can be operated without undue risk to the health and safety of the public.

"These General Design Criteria establish minimum requirements for the principal design criteria...

"The development of these General Design Criteria is not yet complete...

Their omission does not relieve any applicant from considering these matters in the design of a specific facility and satisfying the neces-sary safety. requirements. These matters include:...

3-

"(3) donsidera~ tion of the type, size, and orientition of possible breaks in components of the reactor coolant pressure boundary in determining design requirements to suitably protect _against postulated lass-of-coolant accidents.

(See Definition:of Loss of Coolant Accidents.)...

"It is expected that the criteria will be augmented and. changed from time to time as important new requirements for these and other features are developed.

"There will be some water-cooled nuclear power plants for 'shich the General Design Criteria are not sufficient and for which additional criteria must be identified and satisfied in the interest of public s afe ty... "

--(Emphases added)

" CRITERIA - I. Overall Requirements

~

" Criterion 1 -- Quality standards and records. Structures, systems, and components important to safety shall be designed, fa6ricated, erected, and tested to quality standards commensurate with the import-~

ance of the safety functions to be perforned. Wherefgenerally recog-nized codes and standards are used, they shall be identtfied and evaluated to determine their applicability, adequacy, and sufficiency and shall be supplemented or modified as necessary to ' assure a quality product in keeping with the required safety function. A.quali ty assurance program shall be established and implemented in order to provide adequate assurance that these structures, systemss and com-ponents will satisfactorily perform their safety functions. Appropriate records of the design, fabrication, erection, and testing of structures, systems, and components important to safety shall be maintained by or under the control of the nuclear power unit licensee throughout the life of the plant."

--(Emphases added.).

" Criterion 2 -- Design bases for protection agains't natural phenomena.

Structures, systems, and components important to safety shall be de-

' signed to withstand the effects of natural phenomena...without loss of capability to perform their safety functions. The design bases for these structures, systems, and components shall reflect:...(2) appropriate combinations of the effects of normal.and accident condi-tions with the effects of the natural phenomena and (3) the importance of the safety functions to be performed."

--(Emphases adde,d.)

m p

a _-

" Criterion h -- Environmental and missile design bases. Str'uctures,

systems, and components important' to safety shall be design 5d to accommodate the effects of and to be compatible with the environ-mental conditions associated with normal operation, maintenance, testing, and postulated accidents, including loss-of-coolant acci-dents. These structures, systems, and components shall besppropriately protected against dynamic effects, including-the effects of missiles, pipe whipping, and discharging fluids, that may result from equipment failures and from events and ccr.ditions outside the nuglear power (Emphasesadded.)

unit."

As can be seen from the preceding, these minimum requiremer1ts mandate that all components important to safety be designed with considdration given to LOCA. Further, the Introduction to Appendix A ' clearly states that applicants must consider in th design of the plant what,ever is,necessary

~

to satisfy the necessary safety requirements. Specifically included are the design requirements to suitably protect against post'ulated'~1o_ss-of-coolant accidents (LOCA).

In addition, Criterion 1 requires that if generally-reg'ognjzed codes and standards are used (such as the ASME Code), those-codes a'nd standards must be evaluated and supplemented or modified as necessary to assure a quality product that will operate the way it is supposed to. _In other words, operability must be assured.

A quality assurance program must be implemented for those-. components to assure that they will satisfactorily perform their safety functions, and appropriate records of the design shall be maintained by or.under the control of the nuclear power unit licensee throughout the life of the plant.

Criterion 2 requires that components important to -safety shall be designed such that their opera'bility is assured, and that the des'ign bases for those components shall reflect appropriate combinations of the effects of 9

-e,wx e

\\

^

a

. f s

nonnal and accident conditions with the, effects of the naturaT p$enomena.

and the importance of the safety functions to be performed.

Criterion 4 requires that components important to safety shall be designed with proper consideration given to enviVonmenta] conditions asso-ciated with postulated accidents, including loss-of-coolant accidents.

'This would also include the environmental temperature following,_a LOCA and the ef'fects from that increased temperature.

-The Criteria found in 10 CFR, Part 50, Appendix A,.are thif primary controlling regulations of the Nuclear Regulatory Comission regarding whether or not LOCA must be included in the design criteria for pipe supports. As demonstrated in the preceding, the, regulations hiearly

~

state that LOCA conditions must be considered. ~Further, the regulations clearly state that al1 structures, systems, and components.im'~ rtant

~ ~

to' safety must remain operable and be able to function 4s they are intended to, even 6nder LOCA conditions.

Regulatory Guide 1.124, " Service Limits and Loading Combinations for Class 1 Linear-Type Component Supports," Revision 1, January 1978 Although they do not carry the force of law or of the regulations as set fo' th in 10 CFR, Part 50, Appendix A, Regulatory Guides published r

by the NRC offer guidance for compliance with regulations. Such guidance regarding whether or not LOCA conditions must be considered in the design

-criteria for pipe supports is contained in Regulatory Guide 1.124, Rev.1, January 1978 (CASE Exhibit 743).

In their discussion regarding applicable Regulatory Guides, Applicants state regarding this Regulatory Guide:

i I

r

... _ - ~

..._._m

r s

. _~

1

" Regulatory Guide 1.124

" Design Limits and Loading Combinations for Class 1 Linear Type Component Supports'

~

" Discussion

~

~ ~

~

"All non-NSSS" (Nuclear Steam Supply System = Westinghouse) " supplied Class I linear-type supports comply with Revision 1 (1/78) of this regulatory guide.

~~

"Also refer to Appendix 1A(N)."

--Applicants ' FSAR, page 1 A(B)-52 (Appendix 1A(N) ~ discusses the CPSES NSSS position on Revision 1 (1/78) of Regulatory Guide 1.124. Since Westinghouse does not, as far as CASE is aware, supply pipe supports at CPSES, this would not be applicable.

In any event, Messrs. Walsh and Doyle's concerns d al primarily with pipe supports supplied by NPSI and ITT Grinnel, and for that reason Appendix 1A(N) would not be applicable.

It. appears.to CASE that the.

only reason for including the discussion in Appendix ~lA(N) is the fact that Applicants relied on Westinghouse to prepare FSAR Section 3.9N.3, ASME Code Class 1, 2, and 3 Components, Component SupporJr, and Core Support Structures, as well as Section 3.9N.1, Special. Topics fo.r

.. Mechanical Components Lif.)

As discussed in Regulatory Guide 1.124, it was formulate (with the requirements of 10 CFR Part 50, Appendix A, General Design Crlterion 2, in mind. The Guide states, in part:

"The failure of members designed to support safety.related components could jeopardize the ability of the supported compon~ent to perform

~

its safety function.

"This guide delineates acceptable levels of service limits and appro-priate combinations of loadings associated with normal operation, l

postulated accidents, and specified seismic events for the design of Class 1 linear-type component supports as defined in Subsection NF of Section III of the American Society of Mechanical Engineers (ASME)

Boiler and Pressure Vessel Code.

This guidd applies to light-water-cooled reactors..."

(A. INTRODUCTION; emphasis added.)

]_1/. See CFUR's First Set of Interrogatories to Applicants dated 2/26/81, and Applicants' Answers to CFUR's First Set dated-4/13/81, Questions 1, 2, 3, 5, 7, 11, 12, 15, 18, 19, and especially 17.

e S

=-

' a

~

1

" Load-bearing members classified as; component supports arb e~ssential to the safety of nuclear power plants since they retain components in place during the loadings associated with nonnal and upstt plant conditions under the stress of specified seismic events, thireby per-mitting system components to function properly. They also prevent excessive component movement during the loadings associat6d with emergency and faulted plant conditions combined withL the specified seismic event, thus helping to mitigate the consequences of system damage. Component supports are deformation sensitive because.large deformations in them may significantly change the stress distribution in the support system and its supported components.

"In order to provide uniform requirements for construction, the com-ponent supports should, as a minimum, have the same ASME Boiler and Pressure Vessel Code classification as that of the.supporfed compo-nents. This guide delineates levels of service limits and loading combinations, in addition to supplementary criteria, for ASME Class i linear-type component supports as defined by NF-1213 o.f.Section III.

...the ( ASME) Code does not specify loading combinations...gaidance is required to provide a consistent basis for the design-of component supports... "

(B. DISCUSSION; emphases add,e'd.)

"The design of component supports is an integral par't of the design

~~ of the system and its components. A-complete and cons.istent design is possible only when system / component / component-stipport-interaction is properly considered. When all three are evaluated on an elastic basis, the interaction is usually valid because individual deforma-tions are small. However, if plastic analysis methods are employed in the design process, large deformations that would result in sub-stantially different stress distributions may occur."

(B. 4. Large Deformation; emphases added.)

"C.

REGULATORY POSITION. ASME Code"..." Class 1-linear-type component supports...should be constructed to the rules of Spbsection NF of Section III as supplemented by the following...

.. 5.

Component supports subjected to the combined loadings of system mechanical loadings associated with (1) either (a) the Code design condition or (b) the nonnal or upset plant conditions and (2) the vibratory motion of the OBE should be designed within the following l imi ts :...

"a.

The stress limits of XVII-2000 of Section II and Regulatory Position 3 of this guide should not be exceeded for component supports designed by the linear elastic analysis method. T41ese stress limits may be increased according to the provisions of NF-3231.l(a) of Section III and Regulatory Position 4 of this guide when effects resulting from constraints of free-end displacements are added to the loading combination.

.. 8.

Component supports in systems whose nomal function is to prevent or mitigate the consequenc'es of events associated with an emergency or faulte'd plant condition should be designed within the limits described in Regulatory Position 5 or other justifiab3e limits provided by the Code..."

(Emphasis added.).

"D.

~ IMPLEMENTATION...If an applicant wishes to use I.his regulatory guide in developing submittals for construction permit. applications docketed on or before January 10, 1978, the pertinent portions of the application will be evaluated on the basis of this guide."

~-

(Emphasis added.)

As stated in the preceding, the proper functioning of component supports is vital to assure the safety of nuclear power plants, not only during I

nomal and upset plant conditions under the stress of specifidd' seismic

~

events, but also during emergency and faulted plant conditions.(which 1

~

would include LOCA) combined with the specified seismic event.

Although it is postulated, there is no accurate method oT~ determining

~~

exa'ctly which item may experience an emergency or faulted condition. For this reasoh, and for the reasons stated in the preceding paragraph, each pipe support must be considered in two ways:

(1) as though it were support-

~

ing the item which was involved in the faulted load of a LOCA; and (2) as though it were supporting the item which was not involved with the faulted load but receives the effects of the LOCA. One of the effects of-a LOCA 0

which must be considered is a probable increase in air temperature to 280 F 3

4 within two minutes.

(It should be noted that even this is not conservative.)

2 legulatory Guide 1.124 defines " Emergency Plant' Condition" as "Those operat-ing conditions that have a low probability of occurre'nce" and " Faulted Plant Condition" as "Those operating conditions associated with postulated events of extremely low probability." Both would include LOCA.

3 See CASE Exhibit 659, page 2, 2nd full paragraph, Mark Walsh 7/28/82 direct testimony as corrected on transcript pages 3127-3128;'and CASE Exhibit 659C, page 2, attachment to Walsh testimony,10/9/81 letter from Gibbs & Hill to TUGCO.

4 See CASE Exhibit 659C, page 1, last paragraph.

~

=

9 e

'b'.

)

Only by considering each pipe support in these two ways can the operability of the supo6rts (and thereby the operability of the 4tems

-which they are supporting) be assured under all conditions which they may. experi ence. This assurance of operability is require,d by 10 CFR Part 50, Appendix A, by Regulatory Guide 1.124, and by simple logic.

All items important to safety must be designed to assure that t'h'ey will retain' their capability to perform their' required safety functiQns.

Considering a pipe support in the second of the two ways (as though it 'were supporting the item which was not involved with the fauited load

~

but receives the effects of the LOCA) is discussed under, C.8. and C.5.a.

of Regulatory Guide 1.124. As stated therein, it is permissible to increase

~~

the stress limits according to the provisions of NF-3231.l(a)'of Section

^

IIT and Regulatory Position 4 of the Regulatory Guide only when effects resulting from constraints of free-end displacements are added to the loading combination.

~

NF-3231.l(a)- of Section III of ASME (CASE. Exhibit 744, page 37) states:

" Design, Normal, and Upset Conditions. The stress limits for Design, Normal, and Upset Conditions are identical and are given in Appendix XVII. The allo-M e stress for the combined mechanical lo. ads and-effec.ts which t it from constraint of free-end displacements (NF-3213.">, %t not thermal or peak stresses, shall be limited to three t s

• , s tress limits of XVII-2000."

(Emphasis added.)

Since there has been much discussion both in prefiled and cross-exami-nation testimony regarding thermal stresses and thermal expansion stresses, ar.d since the ASME Code is not as specific on this point as it might have 5

been, it is necessary at this point to d arify exactly what Messrs. Walsh 5 Applicants' witness Reedy confirmed this difficulty in the ASME Code during his cross-examination testimony.

In response to Mr. Walsh's question "Does ASME have different definitions for different subsections?" Mr. Reedy stated in this regard, however, "I think for-your stress analysis terms that you're considering, the definitions are the same." (Tr. 5222/24-5223.)

3 and Doyle are in fact concerned with in this regard. ~Their. concer'ns relate

~

not to themal stress as, specifically defined in the ASME Code, fiut to what is defined in the Code as thermal expansion stress _(or constraint

~

'~

of free-end displacement).

The Code is specific in neglecting the effects of thermal stress as specifically defined in the Code, but the Code also recognizes.the complexity of themal conditions and has subdivided the effects of tem-perature into two categories:

thermal stresses; and constraint of free-end displacement (or expansion stress).

A thermal stress as defined in NB 3213.13 (CASE Exhibit 699) is "a

~

self-balancing stress produced by a non-unifom di.stributi_on of-tempera-ture or by differing thermal coefficients of expansion." An Example of a non-uniform distribution of temperature is where a pipe is.i contact with the pipe support. At the point of contact, the terriperat6re of the steel is the same as the temperature of the pipe. But it few in.ches away from the point of contact, the temperature has decreased a considerable amount. These changes of temperature will cause a stress distribution within the pipe support and will occur during the life of~the plant and perhaps will not adversely affect the overall structura} capacity'of the pipe supports. An example of a thermal stress of differing thermal co-efficients of expansion would be the case where the weld electrode material has a coefficient of expansion different from the' base, metal. This also will have the same effect as the non-uniform distribution of temperature.

The ASME Code defines free-end displacement in NF-3213.10 (CASE Exhibit 744, page 32):

O

5

. _~

" Free End Displacement consists of.the relative ~ motions that would occur between an attachment and connected structure or equipment

~

if the two members, were separated."

~-

Two examples of this form of motion are differential movenients, and

~

expansion and. contraction of a member. For differential movsmint, con-sider a member attached to a ceiling and attached to a floor. In a seismic eventtheceilingmaywanttomovedownward1/2inchandthefl5ormaywant to move up 1/2 inch.

In that case, the stress in the member in. compression

~

would be comparable to the stress that would occur if the membei were to be applied with a load that would give a displacement of 1 inch, provided the material were elastic and had a large enough cross-ectional area to withstand the displacements.

Expansion and contraction of a member would have a simil~ar philosophy.

.~

If a member were attached from floor to ceiling, and the_ air temperature were to increase a certain amount, the member would wani to iiicrease in

~

length, an'd this length could be, for example, one inch _, The stress in the member would be comparable to the stress that would occur if the member were to be applied with a load that would give a displacement of 1 inch, provided the material were elastic and had a large enoug'h* cross,-sectional area to withstand the displacements. Thisformofstregsisdesi'gnated as an expansion stress in ASME,Section III, Subsection NF.

i The definition of an expansion stress as constraint 'of free-end displacement is exemplified in Article NF-1121(a) (CASE Exhibit 745):

" Rules for Supports.

"(a) The rules of Subsection NF provide requirements for new con-j struction and include consideration of mechanical stresses and effects I

which result from the constraint of free-end displacements, desig-l nated as Po in NF-3222.3 but not thermal or peak stresses." (Emphasis added.)

~

l t

=

^ _

NF-3222.3 (CASE Exhibit 744, Pages.34-36) is entitled. ' Expansion Stress Intensi ty, P." Although this paragraph is under NF-3220 DESIGrr0F PLATE e

_ AND SHELL TYPE SUPPORTS BY ANALYSIS (rather tharr under linear supports),

what is important here is that this is the definI' tion to be Osed according toNF-ll21(a). NF-3222.3 states:

" Expansion Stress Intensity, Pe.

This stress intensity is,the highest value of stress, neglecting local structural discontinuities', produced at any point across the thickness of a section by the loadings which result from restraint of free end displacement and the effect of

~

anchor point motions..." (Emphasis added.)

Further, under NF-3200 DESIGN OF CLASS 1 COMPONENT SUPPORTS, NF-3210 GENERAL REQUIREMENTS, NF-3217 Classification of Stresse (CASE Exhi} bit 744, page 33), it is stated:

~

" Table NF-3217-1 provides assistance in the determinati6n of the category to which a stress should be assigned." -

__ Table NF-3217-1 CLASSIFICATION OF STRESSES FOR SOME TYPIC'AL CASES (CASE Exhibit 744, page 34) (which is for plate and s~ hell btit for which the definitions are the same as for design of linear type s_upports,by analysis regarding stresses) states:

i

" Origin of Stress... Expansion "Ll/ Stresses which result from the constraint of free end displace-ment and the effect of differential support or' restraint motions.

Considers the effects of discontinuities but not loc 11 stress concentrations."

l

... and we are again referred to NF-3213:

...For. definition of types of stress, see NF-3213."

(See l

[

page 11 of this pleading.)

In addition, Article NF-3111 (CASE Exhibit 659B, 'page 29, Attachment to Mark Walsh direct testimony) states that:

"The loadings as specified in the Design Specifications (NA-3250) that shall be taken into account in designing a component support l

I

r include, but are not limited to, the following:..

~

"(e) Restrained tfiemal expansion;...

"(g) Environmental loads such as wind and snow loads."

--(Emphas1s added.) '

This is consistent with Criterion 4 of 10 CFR Part 50[ Appendix A.

i

.One of the-environmental conditions involved with a LOCA is an jncrease in 0

the ambient temperature. This temperature reaches 280 F in two minutes after the postulated pipe break (see Footnote 3, page 8,, of this pleading).

r i

The increase in air temperature results in an expansion stress; i.e.,

stress resulting from the restraint of free-end displac'ement' or themal I

expansion stress.

~

Appendix F of ASME Section III (CASE Exhibit 746)_ as been misused by both Applicants and the NRC Staff.

The March 8,1982. TUSI Memorandum to Pipe Support Engine ~ering " Design Criteria File" (CASE Exhibit 659E, Attachment to Mark Walsh djrect testimony) 4 states, in part:

It was agreed that ASME ' ection III, Subsect-ioh NF does not S

"1.

require that themal expansion be considered _in pipe. support design." (Emphasis' in the original.).

"2.

It was stated that information received from Gabe Bove of Westing-house, who is a member of the NF committee, verifies that themal expansion in supports is not generally considered in the nuclear o

industry.

Exceptions do exist, such as members spanning between walls, floor to-ceiling, critical anchors, etc...

"7.

Specification 2323-MS-46A mentions LOCA environmental conditions,

but references sub-section NF for support design.

4 "8.

I't was agreed that themal expansion in pipe support designs would only be considered.in special cases based on engineering 1

. f T

judgement.

If any stress conditions are encountered, they will be further analyzed in detail before any design changes are proposed."

]

Although the March 8 Memo does not specifically me_ntion AEEendix F as being the basis for Applicants' conclusions regar' ding theisil expansion and LOCA, Applicants' witness Chang, in his 8/21/82 deposition (CASE 5xhibit 677, pages 46, 47 and others) referenced Appendix F.

In the June 25, 1983, TUSI Memorandum from Applicants' witness Finneran

~

to PSE " Design Criteria File" (CASE Exhibit 659G, Attachment to Mark Walsh direct testimony), which was for the purpose of expanding on the 3/8/82 and finalizing PSE's position regarding LOCA temperatur considerations relative to pipe support designs, it is stated, in pa'rt:.,

~

" Item 8 of the previous. letter indicated that thermal. expansion in pipe' support design needs to be considered in 'special c_ases'.

It is PSE's position that the only 'special cases' that need.to be con-sidered are when members span between walls or frorii floor to ceiling.

In all other cases local yielding will relieve theie strGsses to a

-~

point; where they are not a problem.

~

"By copy of this-letter G. Krishnan is requested to not rtin any STRUDL themal analysis unless specifically requested to do so by the PSE Engineer."

In the NRC Staff's prefiled testimony prepared'for the September 1982 hearings (Staff Exhibit 201), Mr. Tapia stated (Answer 8, page.5)':

"' Rules for Evaluation of Faulted Conditions,' ASME Section III, Appendix F, excludes thermal stresses resulting from faulted con-ditions in the design procedures. This exclusion is based on the fact that the thermal stresses are relieved by ductile displacement.

The evaluations of plant faulted conditions 'are intended to demonstrate the structural capability of the system, to ensurd operability of the piping. The evaluation allows the material to be stressed above the yield point provided that sufficient ductility exists in the material to allow relaxation of constrained thermal expansion stresses prior to the material reaching failure strain." (Emphasis added.)

m

r-Similar realoning is contained in the NRC Special'Inspiction: Team's investigation report regarding the Walsh/Doyle allegations, Repact 50-445/82-26, 50-446/82-14, bottom of page 17 continued on page 18, which states, in part:

"The Special Inspection Team determined, from interviews. with cogni-zant design engineers and from calculation reviews, that the Appli-cant had not considered LOCA thermal expansion effects on concrete inserts and bolts in the design of individual pipe supports and associated concrete anchors...This decision was based primfrily on the ASME Code Section III, Appendix F, ' Rules for Evaluation of Faulted Conditions,' which does not require that differential thermal expansion stresses resulting. from faulted condition's be included in the design procedure.

This exclusion is based-on the ASME Code rationale that these stresses occur once in the lifetime of the plant, are self-limiting in nature and are relieved by small deformations and displacements." (Emphasis added.),

~

It is not clear what the Special Inspection Team meant regardiog "the ASME Code rationale that these stresses occur once ~1n the' lifetime of the plant, are self-limiting in nature and are reli.eved by}gmall defor-

' mations and displacements," or what the basis for that statement is.

Further, this is contrary to what is stated in Regulatory Guida 1.124 (see especially page 7 of this pleading: paragraph 1, last sentence; paragraph 4, last sentence; and last paragraph).

Further, Appendix F (CASE Exhibit 746, page 481) clearly states:

"F-1220. LIMITS OF DESIGN PROCEDURES.

~

"(a) The Faulted Condition design procedures. contaihed in F-l300 are provided for limiting the consequences of the specified event.

They are intended (NA-ll30) to assure that violation of the. pressure retaining boundary will not occur in components or supports which are in compliance with these procedures. These procedures are not intended to assure the safe operability or reoperability of the system either during or following the postulated event."

(Emphasis added.)

Section 3.2.2 of Applicants' FSAR (Applicants' Exhibit 3), beginning on page 3.2-7), SYSTEM QUALITY GROUP CLASSIFICATION, dis' cusses the system e

--- a

. _~

quality group classifications and safety class definitions.. T'his section

~

states, in part:

"2.

Safety Class Definitions... Supports that have a nuclear safety

~

function shall be the same safety class as fhe.coinponents"that they

'~

support..."

Table 17A-1 of the FSAR, a LIST OF QUALITY ASSURED STRUCTURES, SYSTEMS hND COMPONENTS, also includes listings by Safety Class.

In addition, Table 3.llB-1 of the FSAR lists EQUIPMENT REQUIRED TO

~~

FUNCTION DURING AND AFTER AN ACCIDENT.

The FSAR also addresses requirements for ASME Code Class components (FSAR page 3.9N-69):

"3.9N.3 ASME CODE CLASS 1, 2 AND 3 COMPONENTS, COMP 0_NENT-SUPPORTS AND CORE SUPPORT STRUCTURES

(

"The ASME Code Class components are built to accepte'd ASME Code,

'Section III requirements. For Code Class 1 compo~nents, very stringent requirements are imposed and are met. ForCodeC1 Tass-2and3compo-nents, the requirements are less stringent but adequate,. in accordance l

with, the lower classification.

"See Section 3.9N.1 for more detailed discussions on ASME' Code Class 1 components."

Included in FSAR Section 3.9N MECHANICAL SYSTEMS AND C0![PONENTS are the following (page 3.9N-26):

"3.9N1.4.2 Analysis of the Reactor Coolant Loop and. Supports "The reactor coolant loop piping is evaluated in accordance with the criteria of ASME III, NB-3650 and Appendix F."

(Emphasis added.)

... and (page 3.9N-50):

"3.9N.l.4.7 Stress Criteria for Class 1 Component's and Component Supports "All Class 1 components and supports are designed a'nd analyzed for the design, nonnal, upset, and emergency conditions to the rules and requirements of the ASME Code,Section III. The design analysis or test methods and associated stress or load allowable limits that

17 g

L

(

j-1

~

will be-used to evaluation of faulted condition"are those' tilat are defined in Appendix F of the ASME code with... supplementary option..."

~

^

-(Emphases added.)

. The manner in which pipe supports are designed and constr_u_cted is also important as they pertain to safety-related active valves. 'Se~ction 3.9N.3.2 of the FSAR, regarding Pump and Valve Operability Assurance (page 3.9N-72) details the extensive methods used to demonstrate the operabili_ty of active pumps and valves, including requirements for operability under the specified plant conditions and defining appropriate.acceptince criteria to ensure that the program requirements are satisfied.

It also states (page3.9N-76):

" Safety-related active valves must perform their mechanical motion in times of an accident.- Assurance is supplied that these-valves will operate during a seismic event. Qualification tests accom-

. panied by analyses are conducted for all active:valy'es.2

__ _ But Section 3.9N.3.4 Compone'nt Supports stMes (paje 3.9N482): '

" Active valves are, in general, supported only By the dipe attached to the valve.

Exterior supports on the valve are_generalJy not used." (Emphasis added.) '

This means that the safety-related active valves are supported by pipe which is in turn supported by pipe supports whose operability must

~

be assured.

Thus, Applicants (with the blessings of the NRC Staff) are designing and constructing vitally important component supports, whose o'perability must be assured, in accordance with Appendix F of the ASME Code, which does not (as stated in Appendix F itself) assure the s'afe operability or reoperability of the system either during or following the postulated event.

O

^-

I 18 -

n.

r

^

While it might be argued that it.is permissible to assume that a certain item will be-affected by a LOCA, and that therefore it Tp per-

_ missible to_ use Appendix F in the design and construction of that item, c

this' ignores the second of the two ways in which'it is necessa'ry to con-

. sider each support in order'to assure operability of the support under all conditions which they may experience.

(Seediscussionon[ ages 8 l

and 9.of this pleading.) When we consider this same item in the second

~

- of the mo ways (as though it were supporting the item which was not f

involved with the faulted load but receives the effects of-the LOCA),

- we get an entirely different picture.

}

If we have a. Class 1 pipe (which we'll call P for convenhence) being -

J

~

supported by a Class l-pipe support (which we'll call PS),whi h,is involved

. in-a LOCA, it may be pemissible to use Appendix F in th_e design and con-F

. s'truction of P and PS -- but only insofar as we have coifsidered ths first

.of the two' ways necessary to assure operability.

In this case, another

- Class 1 pipe (which we'll call P2) being supported by a Class'l pipe support

~ (which we'll call PS2) is required to remain operable and must herefore be designed to Level A or B allowables- (design, nomal and, upsqt. conditions),

rather than to Level D allowables (faulted conditions) pmissible under I

. Appendix F.

But what if it is 'P2 and PS2 which are involved in'the LOCA? In that case, it would be pemissible to' design P2 and PS2 to Level D allowables; l

but P and PS would have to be designed to Level A or B allowables~.

What Applicants and the NRC Staff are in effect saying is that should there be a LOCA,and P and PS have been designed to Level D allowables, 0

'I i

. = = -

,e r

e--*

- - " ' ~ " - - ~ ~ ~ ' - -

- ^ ^ - ^ ^ - -

~,

it is permissib1'e to als'o have designed,P2 and PS2 (and P3.and Pi3, etc.)

to Level D allowables. ' But this approach does not assure operability or reoperabili ty.

In effect, it increases the possibility that, should there be a LOCA, not only P and PS will be involved in~~it, but P2 and~ PS2 will then not be able to function due to the increased temperature resulting from the LOCA, which will in turn increase the possibility thaththe initial LOCA will have a domino effect leading to an additional LOCA from the failure of P2 and PS2 and other Class 1 items which have been designed to Level D allowables.

By using Appendix F, Applicants and the NRC Staff dre in e'ffect saying that if there is a LOCA, all the Class 1 items which 'have beed_ designed to Level D allowables can be allowed to fail. Clearly this is contrary

-~

to the intent of all NRC regulations and regulatory guides. And CASE cannot believe that this was the intent of the ASME Codd either. Finally, this approach flies in the face of all logic and common sense.

In order to assure operability and reoperability, and to mitigate the consequences of a LOCA, the most conservative approach wou14 be to design all Class 1 items with faulted loads and to normal operating allow-ables (Level A or B allowables which would include consideration bf the con-straint of' free-end displacement, or thermal expansion stresses). This would recognize the faulted load condition of a sudden increase in load due to a pipe break (similar to the effect of suddenly turning on a garden hose, for example), while at the same time assuring operability under design, normal, and upset conditions.

9 e

e

~a-e--

v m

e w

- 20'-

Realistically, however, in the real world the co binaiion of faulted loads 'with constraint-6f free-end displacement (thermal expansion stress) and Level A or B allowables may not be practical. A more reasonable approach

-might be to make two analyses, giving consideration to t,he two ways in -

~

which it is necessary to consider.each item in order to assure operability-

'under all conditions which may be experienced.

]

The f'irst would be to consider the load combination of cor}straint of free-end displacement (thennal expansion stress) wit,h faulted loads, using Level D allowables. The second method of analysis would consider constraint of free-end displacement (thermal expansion stress)'with normal upset loads and Level B allowables. Both analyses m0'st be done with each item in order to assure operability and reoperability of the-system during and' following the postulated event.

(It'should be noted thaf this does

~~

not consider the effects placed on the Richmond Insert; which is not covered by ASME b'ut by American Concrete Institute (ACI); this will be addressed briefly in the following and in more detail during the upcoming hearings.)

Class 2 and Class 3 items must also be designed to' assure operability under all conditions which may be experienced. To this point, we have discussed only Class 1 items. However, the same requirements should be

~

met for-Class 2 and Class 3 as for Class 1, as stated in the ASME Code:

"NF-3300 DESIGN OF CLASS 2 AND CLASS MC COMPONENT SUPPORTS

'"NF-3330 DESIGN OF LINEAR TYPE SUPPORTS BY ANALYSIS "The design rules and stress limits which must be satisfied for the Design and Operating Conditions are as given in NE-3230" (See CASE Exhibit 747, especially pages 44 and 37.)

=

"NF-3400 DE5IGN OF CLASS 3 COMPONENT SUPPORTS

~

"The design of Class 3 component supports shall be in accordance with the requiremepts of NF-3300 using one of the design procedures indicated in Table NF-3132.l(b)-1 for Class. 3 construction. The applicable table of allowable stresses for a given-material to be used with a specific design procedure is stipulated in Table NF-2121-1."

(See CASE Exhibits 747, especially page 45; 744, especially page

~

31; and 748.)

It should also be noted that we have not attempted-to address here the use of gang hangers (where more than one pipe is shared-by the same support) at Comanche Peak. This is a common practice throughout the containment, and should also be properly considered.

t The 11/20/81 Memorandum for All NRR Personnel from Harold R. Denton, Director, Office of Nuclear Reactor Regulation, NRC, Washingt6n, regarding Standard Definitions for Commonly-Used Safety Classific tion Terms, further supports CASE's positions in this Brief. This Memorandum (which CASE received 4/18/83 during preparation of this Brief) offers guidance regarding the use of certain coninonly-used safety classification terms. Mr. Denton states,

~

. in part (CASE Exhibit 749, page A-1):

"...I am endorsing and prescribing for use by all NRR personnel the standard definitions set forth in the enclosure to this lei;ter...

.For the time being...the definitions in the enclosure should be considered ' standard' and should be applied consistently by. all NRR personnel in all aspects of our safety review and 1-icensing activities

. and should be appropriately reflected in our regulatory guidance documents..."

The enclosure to Mr. Denton's memorandum, DEFINITION OF TERMS, states in part (CASE Exhibit 749, enclosure page):

e e

22 -

[

^

[

" Importantito Safety

~

C'

' ". Definition Fr.om 10 CFR 50, Appendix A (General Design Criteria)

- see first. par,agraph of ' Introduction. '

i

"'Those: structures, systems, and components that provide reasonable

. assurance that the facility can be operated without.un.due risk to the health'and safety of the public.'

. Encompasses the broad class of plant features, covered (not neces-sarily explicitly)'in the General-Design Criteria, that contribute in important way to safe operation and protection of the public

'in all phases and aspects of facility operation (i.e., nonnal operation and transient control as well as accident mitigation).

. Includes Safety-Grade (or Safety-Related) as a subset.

" Safety Related

". Definition - From 10 CFR 100, Appendix A - see sections II}.(c),

-VI.a.(1),andVI.b.(3).

2 "Those structure, systems, or components designed to_ remain functional for the SEE. (also termed ' safety. features') n,eces.sary_j;o assure required safety functions, i.e. :

"(1) the integrity of the reactor coolant preshure b.06ndary;

"(-2) - the capability to shut down the reactor and main'tain it in

~-

a safe shutdown condition; or

"(3) the capability to prevent or mitigate the consequences of accidents which could result in potential off-stte exposures comparable to the guideline exposures of this pa~t.

j

. Subset of. 'Important to Safety'

. Regulatory Guide 1.29 provides an LWR-generic, function-oriented listing of ' safety-related' structures, systems, and components needed to provide or. perform required safety functions,. Additional information-(e.g., NSSS type, B0P design A-E, etc.) is needed to generate the complete listing of safety-related SSC's for any specific facility..."

l

--(Emph.ases in the original.)

l-Safety Class Definitions are contained in Applicants' FSAR (Applicants' Exhibit 3), Section 3.2.2, beginning on page 3.2-7.

Ba' sed on those definitions aim a

e-w-,->

m,,-~,-.,-,g.,..#vy

'-a--

w ab%

e w'e

-me-wm-e---*='y w

-my**e e-eww-efWm

<y

-w--m-.

w m

• w w e *

• eu-

=+mw a-=W"

r 23 -

and on all of t8e preceding, it is obvious that LOCA..(incliidin thermal I

expansion stresses, or-free-end displacement) must be considered in the design of all equipment inside the containment at a minimum (including pipe supports) and must therefore be included as part of the design criteria for Comanche Peak.

~

Applicants have not considered certain-necessary ingredien in their-analyses. Although there will be more detailed discussion and-cross-exami-nation in this regard in the upcoming hearings, there are a few noteworthy examples of these omissions by Applicants which CASE would call to the attention of the Licensing Board.

In Applicants' supplemental direct testimony in the Septemb~er 1982 hearings (Applicants' Exhibit 142F, Answer 1), Mr. Fihner'an dates:

~~

" General Design Criterion 4 requires that structures important to safety be designed to acconnodate the environmental effeets of LOCA.

This 'is a general requirement which does not specify the methods,

by which the criterion must be satisfied. Specifit implementation j.

of the criterion is further defined in the FSAR and project design specifications which specify ASME Section III, Subsection NF as the basis for the design of pipe supports.

i "ASME Section III, Subsection NF recognizes the sel'f-limiting char-acteristics of thermal stresses and thus does not require-their con-sideration in the design of. linear pipe supports.. Aside from the Code, we have shown that slippage of anchors will ~ relieve thermal stres'ses in worst case situations during a LOCA without loss of function of the supports. As such, the requirements of General Design Criterion 4 have been satisfied."

--(Emphases added.)

CASE does not agree that Applicants "have shown that slippage of anchors will relieve thermal stresses in worst case situations during a LOCA without loss of function of the supports" or that."the requirements i

f i

i

-,,,.44--r

,. ~ - <,.....

m.

,-.. -..m, y-,_--..

,,--_--,,-.,_,.-,,9 m-.- - -

m.,

.,,,n

=

of General Design Criterion 4 have bee.n satisfied." ~We will b'e addressing this in detail in the u}pcoming hearings, but the following shouTd be n.oted as well.

It is implied that Mr. Finneran, in the preieding stateinehts, is speak-i ing of Applicants' Exhibit 142D (Attachment to Applicants Exhibit 142, Rebuttal Testimony regarding Walsh/Doyle allegations). AlthougEthere

~

are three different types of configurations of supports included in 142D,

~

Messrs. Walsh and Doyle were primarily addressing the use _of Richmond inserts (Applicants' Case 2, 9th page, Exhibit 1420) and we will address ourselves to this.

In this example, this reaction is due to constraint of fbe-end dis--

placement (or thermal expansion stress). Howeve'r, this reaction does not

~~

consider: bending stresses in the bolts; mechanical 'oads; different.

l corifigurations of pipe supports; operability of the piping syij;em; or

~

the method'used for the allowable load on the bolts.

Further, the analysis is based on assumed load displacements of a Richmond insert.

Regarding the bending stresses in the bolts, CASE agrees with that portion of the NRC Special Inspection Team's statements in, Investigation Report 82-26/82-14 regarding the Walsh/Doyle allegations which st'ates (Report, bottom of page' 21 continued on top of page 22):

~

" Calculation by the Special Inspection Team of the stresses result-ing from the shear, tension, and bending moments for the five load-ing cases analyzed, indicates that bending s'tresse.s in the bolt for the worst-case conditions are 15 times larger than the stresses resulting from shear. Although bending in the. bolt may result in reducing shear on the insert, it imparts an additional bending stress in the bolt which has not been calculated...While there have been questions about whether the bolting is governed by the ASME Code, the NRC staff believes that the total stress (including the bending stress) in the bolts should be evaluated to assure that the value for allowable stress has not been exceeded. The NRC staff requires that this value shall not exceed the ASME Code allowable stress for bolting."

(Emphases added.)

l

~.

Regarding lhe method used for the allowable load on the borts, the NRC l

C Special Inspection Team stated in Investigation Report 82/26,&2/14(page 18,topofpage):

[

"Although the ASME Code is not directly applicable _to th design-

)

of the concrete anchorages, the Applicant adopted the ASME Code l

philosophy in the design of the concrete inserts. This design approach is documented in Sections 3.8.3.3.3 and 3.8.4.3.3 of the FSAR, where it states, '... thermal loads are neglected when they are secondary and self-limiting in nature and when the material is ductile. '

"With respect to the design of inserts such as Richmond iTiserts, the Special Inspection Team found that these components are not governed by the ASME Code nor by any other standard which the NRC has adopted as a regulatory requirement. Thus, th,e only applicable-regulatory standards are the requirements of 10 CFR Part 50, Appendix A - General Design Criteria For Nuclear Power PDnts, Cr_iteria 1 and 2, which require that such components be capab,le of. performing their intended design function which is to carry ths imposed loads without failure."

-~

But the ASME Code addresses this in ASME NF-ll32.5 (CASE 'Exfi'ibit 705):

^

"NF-1132.5 Nonintegral Support Connected to Build 4:ng Structure.

The jurisdictional boundary between a building structure and a non-integral support shall be the surface of the building structure.

Such means of mechanically attaching the support to the-tiuilding structure shall fall within the jurisdiction of Sub-section NF."

Therefore, the bolt which is screwed.into the Richmond insert (which is embedded in the concrete) is part of the pipe support 1nd thus is governed by the ASME Code. The Richmond insert and the. wall itself are governed by the American Concrete Institute ( ACI) Code (specifically, ACI 349-80 " Code Requirements for Nuclear Safety Related Concrete Structures),

as the document which is more directly applicable than any other presently available at this time. It is CASE's understanding that a Regulatory Guide is in preparation at this time to implement the requirements of Appendix B to ACI 349.

O

In his test'imony.in the June hearings, Applicants wiknes' scheppele

~

s testified that concrete'is a brittle material (Tr. 853):

"If you can visualize something like concrete, which is a brittle material when it's subjected to tension it would tend to track..."

This is also confinned by the book THE TESTING AND INSPECTION OF ENGINEERING MATERIALS (CASE Exhibit 750, page 37). This book i.s part of

~~

the McGraw-Hill Civil Engineering Series.

It states:

" Ductility is that property of a material which enables it to be drawn out to a considerable extent before rupture and at the sanie time to sustain appreciable load. Mild steel is a ductile material. A non-ductile material is said to be brittle, i.e., it fractures with rela-tively little or no elongation. Cast iron and concrete are brittle materials..."

(Bnphases added.)

- s Therefore, concrete is a non-ductile material. 8~irice the-Ricbond

~

insert is embedded in the concrete, in shear loading the insert bears

-~

against the concrete, and in tension loading the insert is h.e171 by the concrete. Thus, what happens to the concrete has a majdr bearing on what happens to'the insert; the concrete is a major factor in the pe,rformance of a Richmond insert.

In Applicants' Exhibit 142E, page 1, it states:

" Prior to failure, four fine cracks emanated from the insert on the top of the block on both specimens.

In Applicants' Exhibit 142E, page '2, it states:

I l

"Six cracks emanated from the insert on the top of the slab and l

extended down on four side surfaces to the reinforcement."

Applicants' Exhibit 142D, pages 25, 26, and 27 state:

"The concrete spalled in a 2" radius..."

"The concrete spalled in a 3" radius..."

"The concrete spalled in a 4" radius..."

As indicated in each of the preceding Exhibits, the concrete in all e

e t.

4 27 -

~

1 of these tests on the Richmond inserts was reinforced concretf6,-

The definition of spalling is " chipping or fragmenting, esgcially ofstone"(Webster'sdictionary).

All of the preceding examples from Applicarits' 142D,and 142E are

~

further. demonstrations that the Richmond insert itself, which is embedded in the concrete, must also be treated as non-ductile. Therefoh, the statement by the Applicants in their FSAR (as discussed by the,Special Inspection Team - see p. 25 of this pleading) that "... thertal loads are neglected when they are secondary and self-limiting in nature and when l

. the material is ductile" (emphasis added) is not applidable to'the concrete or to the Richmond inserts which are embedded in the'concretel Since

~

one of the effects of a LOCA is an increase in iir temper.ature (which J

~~

would induce a thermal load), it should be considered in'th.e' design and usE of Richmond inserts '(along with other thermal loads;).

1 The Special Inspection Team stated (Investigation _ Report,82/26, 82/14, page 20, second full paragraph):

.The Applicant has stated that ACI 349-80,. ' Code Requir$ents for

[

Nuclear Safety Related Concrete Structures,' an industry standard not adopted by the NRC as a regulatory requirement," allows a factor i

of safety of two for concrete inserts. The Special inspdttion Team found that the ACI standard specifies load factors. and capacity reduction factors and requires consideration of the forces caused by thermal effects under accident conditions.

In addition, the ACI standard requires a testing program far broader than that which

[

has been carried out for the Richmond inserts. The Special Inspection Team cannot concur that the ACI standard allows a factor of safety of two.to be used in the manner in which it has been used-by the Applicant."

(Emphases added.)

6.It should be noted that this use of reinforced concrete is contrary to the-testing requirements of the American Concrete Institute (ACI). This is logical since there is no way of being certain that the Richmond inserts will only be used in locations next to reinforcing steel.

e t

n...

.-.._.y,,

-__m,

r As. pointed oht in ASME, Appendix XVII-2271.3 (CASE' Exhibit 707):

~

-T

" XVII-227.1.3 Provision for Expansion. Adequate provision shall be made for expansion and contraction appropriate to the function of -

the support structure."

By considering this Code requirement, the effects of LOCA could be

' lessened. One way in which this could be implemented would be by the us -

~

of slotted connections at the point where the bolt screws into Qe Richmond insert, as s~uggested by Mr. Walsh in his testimony.

There are other aspects of this matter which will be discussed further during the next hearings.

On page 17 (bottom of page) of the Special Inspecti5n Tea'm's Investi-gation Report 82/26,82/14, it.is stated:

"The Special Inspection Team detennined, from interviews with cogni-zant design engineers and from calculation reviews, that the Appli-cant had not considered LOCA thermal expansion effects on~ concrete

-~

inserts and bolts in the design of individual pipe supports. and

- - associated concrete anchors."

(Emphasis added.)

From the preceding discussions, it is now obvious that the NRC Staff should require that Applicants consider LOCA thermal expansion effects on concrete inserts, the bolts which screw into the inserts, and the steel used for the pipe support in the design of pipe supports and associated concrete anchors (including Richmond inserts).

In regard to the Applicants' claim that they have considered the slippage at the intersection of a Richmond insert and the tube steel con-nection in their prefiled supplemental testimony (see page 23 of this Brief), Applicants' Exhibit 142D (page 9) contains an analysis of a 6x6x3/8 inch tube anchored with 1-1/2 inch diameter Richmond inserts with threaded e

e A

e rods. ~ The analysM s considered only 'a temperature diffe'rential of :210 F.

~

The results indicate a reaction of 21 kips bearing on the Richmond insert..

In their analysis, the A'pplicants have made sever,al engineering errors.

As demonstrated in the preceding, the Applicants' presumption that the threaded rod is not covered by ASME Subsection NF ls incorrect; the

- threaded ' rod h covered by ASME. The threaded rods used at Comihche Peak are.SA36 rods.

Since SA36 rods are not listed in Table XVII-246 k -1

'of ASME, Appendix XVII (CASE Exhibit 752), and since they are similar to A307 (CASE Exhibit 753, from AISC Steel Manual), the allowable shear strength of the bolt is.3YS (yield strength). Also in the AlSC Manual of Steel Construction (CASE Exhibit 753), the allowable shear 5trength of an A36 bolt is.3Fy (yield strength). The bolt's-yield streng-th lit 300 F 0

is 31.9 as shown in' Code Case N71-10 (CASE ~ Exhibit 751-). Thi ris.the appl-icable Code Case, as iridicated in CASE Exhibit 754,hRC 'Regu'latory Guide 1.85,< Materials Code Case Acceptability ASME Section III Division

1. - Regulatory Guide 1.85 states that:

"C.

REGULATORY POSITION

~

"1.

The Section III ASME Code Cases... listed below.~..are acceptable to the NRC Staff for application in the construct-ion of components for light-water-cooled nuclear power plants...within the limitations s tated... " -

~ It also indicates that other " Code Cases that are not listed herein are eith.er not endorsed or_ will require supplementary provisions on an individual basis to attain endorsement status."

For a 1-1/2 inch diameter bolt, the allowable. shear on the bolt, under Level B allowables, is @r (1.5)2/4] (.3)(31.9) = 16.9 kips. This is less m

4 6

o-

~ than the allowable of 25 kips used by the Applicants. - Reguihtoty.huide i

_'a 1.124 (CASE Exhibit 743,' Position 4) permits an increase of 50% en bolted

- connections when constraint of free-end displacement is considered in the analysis. This would increase the allowable of 16'.9 kips _to 25-kips. This 25 kips allowable is only applicable when an analysis of thermal expansion it included with the mechanical loads; otherwise, the allowable if 16.9 kips is the' allowable.

.The 'value of the Richmond insert the Applicants currently use is based on the Prestressed Concrete Institute (PCI) handbook, as stated in the NRC Staff's Investigation Report 82-26/82-14 (page 20, first' f'ul'1 paragraph).

This method is commonly called the ultimate strength co'ntept. ]When using this form of analysis, loads are increased with road factors and the material'

-- ' is reduced with unde -capacity factors.

- - Although CASE does not, in the interest of time and4 brevity, wish to' go into dethil at this time regarding how the Applicants _have mi,sused the fonnula from PCI, the loading for the Richmond insert should be pointed

~

out. As shown in CASE Exhibit 755, ANSI /ACI 349-76, the load combinations given in equations 7 and 8 under 9.3--Required Strength include _ accident temperature effects, as well as the faulted load combination and a'15%

increase from the OBE earthquake. Using this load combination does not increase the allowable strength of the Richmond insert. 'As can be seen

- from the two equations, the Applicants considered 'only one item out of eleven possible additional loads.

The Applicants' stated that their example was a worst-case condition.

'The conditio~n they analyzed was for a 6-foot long member.

The NRC Special e

G e

v v

r r

Inspection Team skated th'at the wors't-casc condition which bey'e/aluated

,a was for an ll-foot long inember that exists in the containment and-is part of the feedwater system gang hanger with an overa.ll span of approximately 30 feet-(Investigation Report 82-26/82-14, page 18', last p,aragraph).

~

Since the expansion load exhibits a linear relationship, the, ratio of 1T to 6 will provide the load on the threaded rod and Richmond ilsert.

The load is f(ll/6) 21 = 38.5 kips. This value far exceeds the aJlowable used by the Applicants for the bolt and the Richmond insert.

~

Also, neither of the worst cases analyzed by the Applicants and the NRC Special Inspection Team included the results of bo in'tersecting msnbers, such as can be seen in CASE Exhibit 659B,13di (Attachment to Jack Doyle deposition / testimony). Had the Applicants or the NRC 5taff

~~

analyzed two members intersecting at the Richmond inse'rt, 'therT would be two' shear loads of 21 kips each on the threaded rod and-Richmond insert.

These loads'would have a resultant shear force of Y212 + 212 =,30 kips

-- again above the established.allowables used by the Applicants.

~

Also, in the analysis the displacement used by the Applicants for the' Richmond insert is estimated since this 1-1/2" diameteP bolt, has never been tested in shear. This is confinned by the Special Inspection ~ Team (Investigation Report 82-26/82-14, page 18, last full sentence, continued on top'of page 19):

"...there are no deflection test data for 1 1/2-inch Richmond inserts in shear loading. For the reasons discussed below'the Special In-spection Team concludes that additional test data i's required for 1 1/2-inch Richmond inserts."

The values for displacement and load can decrease with an increase in concrete strength, for example as shown for the 1-1/4" diameter Hilti 4

t

. ^

- ~

Super Kwik-Bolt allowables (CASE Exhibi.t 756) going from 4,000 PSI concrete to 6,000 PSI concrete shows a decrease of 435 lbs. in strength. }his de-crease in strength, although not indicated, is also a decrease in the deflection of the bolt.

~

~~

' And finally, during the deposition of Kenneth Scheppele. (Vice-President of Gibbs & Hill), Applicants witness Scheppele stated (CASE Exhildt 757):

"The structural steel within that building, I would say has considersd the thermal effects of a LOCA by several measures, eithbr by means of

-- and this is where we talk about major structures, I'm not talking about structures which are relatively short and which would be attached to concrete by means of bolts, things of this sort but I would say as far as major structures are concerned, these are.$ elements which would be nonnally reviewed from the viewpoint of temperature ejpansion."

In reviewing the calculations of the wall-to-wall' steam generator upper lateral restraint (referred to in Investigation R'eport 82-26/82-14, page

. 25, middle paragraph)(CASE Exhibit 758), it is clearly indicated that.the

~~

ca1culations in regard to LOCA were done after Mr. Nalsh had testified in the' July hearings, and that LOCA was not previously cons.idered,in those calculations.

In reviewing the calculations, the engineer has made two

(

- errors which should be noted.

First is that he is using the yield strength of the material at room temperature, not at accident temperaturg; a similar mistake was made on'the threaded rod which screws into the Richmond insert.

The second mistake was that the engineer is using the main wall reinforce-ment for shear reinforcement. The ACI Code (318-71, par'agraph 12.13) is explicit in stating that stirrups will be U-shaped or multiple U-shaped to carry the shear forces that 'the concrete cannot withstand. Stirrups are not used within the walls 'at Comanche Peak.

There are many other errors regarding these matters which will be

= discussed further during the next hearings.

7

..t Although we dre still analyzing some of the documehts r$ceived recently I from the Applicants on dtscovery, there are a couple of. specific 4xamples which we will defini~tely. be addressing further in.the next hearings which are-especially noteworthy.

~

One example is in further reference to the steam generator upper lateral restraint mentioned on the preceding page o.f this Brief. Fromogpreliminary review, i.t appears that the Applicants used an incorrect method i,n detennining the forec applied to walls A and B of the steam generator, compartments.

' Using energy methods, it appears that the force exerted on' the walls is three times' larger than the-value the Applicants obtained in their calculations (which the-NRC:Special Inspection Team reviewed in prepar,ing tfleir Investi-

~

gation Report 82-26/82-14).

Another example is in regards to the floor-to-wall momen,t' restraint

-~

also-discussed in the middle paragraph on page 25 of Inv4stigation Report

.82-26/82-14:

It appears that the method used by the Applicants is providing erroneous values for reactions and displacement of -the attached pipe.

~

. 'One major flaw in their analysis of the stiffness of the base plates is that' they did not consider the shear lugs underneath the base plates.

The value they used in their analysis for translational s,tiffness at a

- Richmond insert was 83' kips per inch.

If one were to consider the shear lugs acting with th'e Richmond inserts, the stiffness becomes approximat'ely:

-49,000 kips per inch rather than 83 kips per inch as assumed-by the Appli-cants -(and reviewed by the NRC Special Inspection Team). The Applicants claim that~ slippage at the support points relieves the expansion stresses

~ and thattif one were to model in the stiffness of the support, the thermal

expansion stressel during'a LOCA would be relieved and. that 'the supports C could still perform their required ~ function.

However, it is obvi.ous that with a stiffness of 49,0d0 kips per inch, those stresses aren't jbst going to-disappear. This moment limiting restraint has -displacementsUn excess of 2/10".

These displacements are important since the support is adjacent to-the penetration at the contairrnent wall. Large displacements',,of the support at the point where the pipe is may cause-the pipe to shear in the event of'a LOCA.

As we have indicated, these are merely preliminary calcul_ations and

' analyses which we will be addressing further later.

IN CONCLUSION As demonstrated herein, there is n_c_ justification-to be f6Dnd in o

~

10-CFR, NRC Regulatory Guides, other NRC regulations, indust'ry-codes, or elsewhere for the position of Applicants and the NRC Staff of not includ-ing LOCA in the design criteria for Comanche Peak.

In fact, their approach

~

practically assures that some items not involved in a LOCA will not remain I

operable during a LOCA or be reoperable following a LOCA and may even lead to other LOCA(s) due to failure of other items not invol ed in the-initial LOCA.

It is most disturbing to CASE that not only the Applicants have

.taken thi.s position, but the NRC Staff has also in its testimony and in its investigation report regarding the Walsh/Doyle allegations.

From the discussions contained herein, it is obvious that both Applicants and the NRC Staff are not complying with the NRC's own regulations.. CASE should m

e I.

n i.,

^

• not even have to be writing this'Brief. The' NRC.Special Inspection

7. Team should-have uncovered these problems (and the many other problems

~ hich will now have to be litigated in the upcoming hearings) during w

their investigation ot--the Walsh/Doyle allegations-(just as the3RC Region IV inspectors :and investigators should have uncovere,d Ehe many probl_ ems recently identified by the Construction Appraisal Team (CAT) in its recently released report). The NRC Special Inspection Team made no effor to con-tact. Messrs. Walsh or Doyle regarding their concerns, but rather-contacted only the Applicants and representatives of their contractors, sub-contractors, and suppliers (see page 8 of Investigation Report 82-26/82-14,' i' tem ;1.

PersonsContacted).

Had the NRC SpecialLInspection Tear 6 made an effort to contact Messrs.

,Walsh or Doyle regarding their concerns, much of the information contained in-this Brief would have already been known and acted upon, 'and 'there is

a. good possibility that the upcoming heaiings would not even have been necessary. LOCA is:only one of several major concerns and problems which

~

will be addressed by. CASE further in the upcoming hearings. When all the facts are pfesented, it will become obvious that the concerns of Messrs.:Walsh and Doyle were well-founded, just as their concerns regard-ing the. inclusion of LOCA in the design criteria for pipe supports were well-founded.

In conclusion, LOCA conditions must be considered in the design criteria for, pipe supports.

Further, each item important to safety should be analyzed in two ways:

(1) as though it were involved in the faulted load ofla LOCA; and'(2) as' th~ough it were not involved in the faulted load of a LOCA but received the effects of the LOCA, including the constraint

e 36 -

o T

of free-end displacement'(or thermal expansion stresses) resulting from

~

~

the increased temperature due to the LOCA. Only by cons.idering each item

_in these two ways can the operability and reoperdbility_of the Osystems impcrtant to safety be assured under all conditions which-the'y may. experience.

Only by the i_nclusion of LOCA conditions in the design criteria for pipe supports and other items important to safety can the public health. and safety be protected (assuming the many other problems at Comanch'e Peak are also addressed.and corrected prior to fuel-loading).-

Further, and more importantly, these problems must be addressed and corrected not only on paper, but also in actual practice.

1 Respectfully submitted,'

<d _. db ptf s.) Juanita El-lis,' P_ resident CASE (Citizens. Association for Sound ~ Energy) 1426 S. Polk Dallas, Texas

-75224 -

214/946-9446 W

W e

et O

4 e

S e

i e

e l

t

UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION 91f BEFORE THE ATOMIC SAFETY At1D LICENSING. BOARD'

.N

%\\

s In the Matter of APPLICATION OF TEXAS UTILI, TIES Q

_ v f-C; GENERATING COMPANY, ET' AL. FOR Q

Do N and,,,50-44S.y,5Q-446'd.

h//

, AN_0PERATING LICENSE FOR Q

COMANCHE PEAK STEAM ELECTRIC Q

O

/

STATION UNITS #1 AND #2 (CPSES) 1

6,

CERTIFICATE OF SERVICE

]

By my signature below, I hereby certify that true and correct copi,es of CASE'S BRIEF REGARDING CONSIDERATION OF LOCA IN DESIGN CRITEillA FOR PIPE SUPPORTS have been sent to the names listed below this 20th day of Apri1 198 3,

by: Express Mail where indicated by

• and First Class Mail elsewhere_.

• Administrative Judge Peter B.-Bloch Alan S..Rosenthal, Esq., Chairman-U. S. Nuclear Regulatory Comission Atomic Safety and Licensing Appeal Board

-_4350 East / West Highway, 4th Floor U. S. Nuclea? Regulatory Commission Bethesda, Maryland 20014 Washington, D._.C.,

20555

• Dr. Ke'nneth A. McCollom, Dean Dr.W. Reed,Joinsoq,'_ Member

~

Division of Engineering, Atomic Safety %d Licensing Appeal Board Architecture and Technology U. S. Nuclear wegulatory Commission Oklahoma State University Washingtor., D. C.

20555 Stillwater, Oklahoma 74074 Thomas S. Moore, Esqb Member

• Or. Walter H. Jordan Atomic Safety and Licensing Appeal Board 881 W. Outer Drive U. S. Nuclear Regulatory Comission Oak Ridge, Tennessee 37830 Washington,.D. C.

40555

• Nicholas S. Reynolds, Esq.

Atomic. Safety and Licensing Appeal Panel Debevoise & Libermari U. S. Nuclear Regulatory Commission 1200 - 17th St., N. W.

Washington, D. C.

20555 Washington, D. C.

20036 Docketing and Service Section

• Marjorie Ulman Rothschild, Esq.

Office of the Secretary Office of Executive Legal Directo,r U. S. Nuclear. Regulatory Comission Maryland National Bank Building Washington, D. C.

20555 7735.0ld Georgetown Road - Room 10105 Bethesda, Maryland 20814

• Ms. Lucinda Minton, Law Clerk Atomic Safety.and Licensing Board Panel

. Atomic Safety and Licensing Board 4350 East /Wes t Highway,' 4th Floor Panel Bethesda, Maryland 20014 U. S; Nuclear Regulatory Commission Washington, D. C.

20555

1 Certificate of Service -

Page 2 l

s

.* David-J. Preister, Esq.

-T. Assistant Attorney General

~ Environmental Protection Division

~

Supreme Court Building

, Austin, Texas 78711 John Collins Regional Administrator, Region IV U. S.' Nuclear Regulatory Comission -

611 Ryan Plaza Dr., Suite 1000

~

Arlington, Texas 76011 Mr. R. J. Gary Executive Vice President and General Manager l

Texas Utilities Generating Company 2001 Bryan Tower Dallas, Texas 75201 Lanny Alan Sinkin

~

~~

838 East Magnolia Avenue San Antonio, Texas 78212

~

~

ddmb fN:J

'ps.) Juanita Ellis, President LASE (Citizens Association for Sound Energy) 1426 S Polk Dallas, Texas-75224 214/946-9446 e

O e

.