ML20070M080

Reactor Vessel Core Support Cone Structural Integrity

ML20070M080
Person / Time
Site:	Clinch River
Issue date:	12/21/1982
From:	WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML20070M075	List: ML20070M074 ML20070M080
References
NUDOCS 8301120262
Download: ML20070M080 (18)

Text

c:

t REACTOR VESSEL CORE SUPPORT CONE STRUCTURAL INTEGRITY Presented in Response to Questions Raised at ACRS Meeting of August 18, 1982 Also s ea 1% 72. ) Lew aq s: st: io, y, R. Leay au ter

5. Ct.e. k, hie <4sa) A.n.,3: p be. ts-ss, e 412., wie <M.I

-k T:

$taari3 ha. C. /cli.L tivec s rWe r R.~.4.< N.+

mee4io),"

w h.-w is,ign h ADVANCED REACTORS DIVISION Madison, PA December 21, 1982 l

l B301120262 830111 PDR ADOCK 05000537 O

PDR

TABLE OF CONTENTS Introduction Discussions 1.

General Description i

2.

Core Support Cone Thennal Environment 3.

Core Support Cone Materials Considerations 4.

Core Support Cone Stress Assessment Summary i

Conclusion i

I l

6

- - - - - - -, -. - - - -, -. ~ ~

-n

-,,,.-n

e INTRODUCTION A neeting was held on August 18, 1982 in Washington, D.C. to present the design of system components for elevated temperature service as well as several other topics to the ACRS. The design of the reactor vessel was in-cluded in the agenda since it is exposed to elevated temperature service.

During the presentation, several questions were asked concerning the core support cone to core support and vessel attachments and the conse-quences of attachment failure. The concern, in particular, was for gross failure of the weld joint of the cone to core support plate re-sulting in the core support structure and core dropping away from the control rod into the inlet plenum.

The example cited was BWR experience with inter-granular corrosion of the support attachments.

This paper addresses the ACRS concerns and shows that the CRBRP Reactor Vessel core support cone'is structurally adequate with-out additional backup supports or routine in-service inspection. The paper addresses the design of the weld joints, stress levels and conservatisms in the design loadings and stress analysis.

I l

4

~

2.

Core Support Cone Thermal Environment The core support cone is in a benign thermal region of the reactor vessel located between the inlet plenum and the reactor vessel / core barrel annulus.

In.this region, the design temperature during full power steady state operation is 775'F and the design thermal transients are relatively benign. There is conservatism in these design values because this steady state temperature exceeds expected temperatures by as much as 75'F and the design transients are correspondingly more severe.

Transient temperatures in the inlet plenum were calculated using results from an empirical equation for inlet plenum mixing.

The mixing constants at-different locaticns within the plenum were determined from two loop and three loop mixing tests (pqrformed at HEDL in the one quarter scale Inlet Plenum Feature Model. 31 The plant duty cycle dvent transients were analyzed and three events were selected as umbrella transients for the core support cone design.

These three events are as follows:

1 1.

N-4A (up) - Plant loading from 405 to 100% power.

l 2.

U-2e - Plant loading at maximum rod withdrawal rate - from 40% to 100% power.

3.

F-4A - Saturated steam line rupture.

l The first two of these events are symmetric.and the tn1ra is asymmetric. Tne l

transient boundary temperat'ures at the botto9 of ' he support cone are t

illustrated in Figures 3 through 5 for the three umbrella events. The location on the support cone where these temperatures apply is shown on Figures 6 and 7.

The transients that occur in the reactor vessel core barrel annulus are similar to the inlet plenum transients but they are mitigated as a result of flowing through the core support plate. Thus, the transients above the support cone are even less severe than those below and the temperature difference across tne support cone is small.

Reference 1) HEDL-TME 76-33, PM McConnell, et. al., " Inlet Plenum Feature Model Flow Tests of the Clinch River Breeder Reactor: Addendum V Results," March 1976.

3.

Core Support Cone Materials Considerations The operational environment of the weld between the core sup' port structure (CSS) and the reactor vessel cone (RVC) is benign and thus minimizes concern for.in-service materials related problems.

The design temperature of 775'F is sufficient!y low that metal loss due to sodium corrosion will be virtually non-existent.

Furthermore, at the very low flow rates involved, no allowances are necessary for erosion losses. The only sodium-related change in the material will be some slight surface carburization, possibly to a carbon. level of about 2000 ppm. However, due to the low carbon diffusion coeffi-cient at 775 F, the carburized layer is unlikely to exceed 3 mils in thickness and thus constitutes a very small percentage of one total cross-section thickness.

In eddition to effects due to sodium exposure, consideration was also given to the possibility of neutron-radiation-induced embrittlement.

The low temperature involved'(775'F) assures that no helium embrittle-ment will be experienced and radiation effects will be restricted to displacement damage. The support cone displacements per atom (dpa) are s

0.0001 maximum. Generally, the displacement threshold at which radiation effects such as ductility loss and fracture toughness begin to be observed in the austenitic stainless steels and their weldments is considered to be 0.5-1.0, dpa.

Below this value, radiation effects may be ignored. Since the predicted displacement level is nearly four crders of magnitude below this threshold value, no radiation damage is foreseen for the CSS /RVC weldment.

Consideration has also been given to the possibility of material property changes arising from prolonged operation at 775'F. Thermal aging is known to be detrimental to the austenitic stainless steels and their weldments, particularly when a sigma phase is produced and severe embrittlement results. However, the operating temperature of the CSS /RVC weldment area (775*F) is much lower than that required for sigma formation (s1000'F mininum) and no embrittlement is expected to result from this reaction.

i The other effect of thermal aging is to induce carbide precipitation, leading to increases,in tensiile and yield strengths, and associated losses of ductility. Again, although some carbide formation is likely at the CSS /RVC weldment, the low operating temperature will ensure that this reaction will remain very localized at the surface and thus will not produce any measurable changes in mechanical properties over the plant lifetime.

k i

t

4.

Reactor Vessel / Core Support Cone Stress Assessment This section presents a sunsnary of the stress levels and fatigue damage at the upper and lower weld joint areas of the core support cone (Fig. 8. Sections B-B and A-A, respectively).

The summary of stress analysis results on the core support cone are shown in Tables 1 and 2.

Section A-A has the minimum factor of safety [ allowable stress / calculated stress] of 1.186 for an upset load combination of OBE plus thermal. This factor is a comparison of the primary plus secondary stress intensity range with the appro-priate criteria for this stress intensity. Section B-B has a minimum factor of safety of 1.238 for a faulted load combination of SSE plus pressure. This factor is a comparison of membrane plus bending stress intensity with appropriate stress intensity limits from Code Case 1592. Section A-A is the limiting section for primary plus secondary stress intensity limits and fatigue damage, while Section B-B has the lower margins for primary stress intensity limits.

Analyses of the core support cone were performed for both mechanical and thermal loading conditions. The conservatisms in the analyses for each type of loading will be addressed in the following paragraphs.

The components were' analyzed to the ASME Code for Nuclear Components which contains margins of safety, such as using minimum properties for the physical properties of materials, combined with a factor of 3 on tensile strength, and a lim,it of 2/3 on yield strength for primary stress intensity, and for fatigue limits, a factor of 20 on cycles and 2 on stress. Additional checks are required which include insuring

" shake down" and preventing " thermal ratcheting". The applicable ASME Code Edition, Code Cases and RDT Standards are listed in the PSAR.

PRIMARY STRESSES The mechanical loadings consist of differential pressure, dead weight and seismic. The following is a list of the conservative approaches used in the evaluation of the primary stresses due to the mechanical loadings. These approaches apply to both Sections A-A and B-B unless noted otherwise. Other areas were analyzed using standard techniques defined in the structural criteria documents.

1.

The stresses due to differential pressure (ap) were calculated using 170 psi (design ap), while the maximum operating pressure is 139 psi.

2.

The dead weight stresses are in the opposite direction to the ap stresses.

In the evaluation they were either ignored or the absolute value was added to the pressure stresses. However, this is a small ettect because the dead weight (DW) stresses are small compared to the op stresses. The largest ratios of DW/ap stresses are obtained for the membrane stresses. Tables 1 and 2 show the i

l l

relative magnitude of the stresses produced by these two loading conditions.

Ip^

^ :p m s pq

."O

./

j, N x s

T s

i j

g q.

3 f

f; y

1 a

r.,

y yf'z y

/

v Q

e'

.y i

g,"g_

q

m.

3.~

The total reismic stresses wLre cal'eulated for Sestion A-A by N'*

direct sunenation of the stressOntensities due~ to the North-

,i Souph (N-S) East-Ves"t (E-W);and' Vertical seismic. loadings.

A There,an two conservative aspects to this approach. On an

, ~

' axisymtric structure, the~ combination of the NsS.and E-W s

directions by the appmpriate square root of the sum of the

.x squarw(SRSS) method'tsplies' that the combined stress cannot exceed the maximum stres;; from either of the. two seismic directions.

4 In additioni the c:irect hmmation of the vertical component L ' with eitherfof the horizontal components is conservative because the combita! tion should beidone by SRSS method. Additional con-servatism arise's f om workin' with stress intensities instead of g

sti'ess components",'

'a. 'The total primary stress intensities were obtained by adding the stress

, intensities of the' differ'ent mechanical loadings'in absolute.

fashion. A less conservative, but acceptable approach would be

'to perfonn the combination (with appropriate signs) on. the stress component basis.

1

~'i 5.

The stress allowables,at' Section A-A and B-B are based on the High Temperature Coda, Case 1592 at 85,0*F, while the ma~ximum operating

'~

temperature is'7E0*F and the design. temperature is 775'F. The.

~

reasons for this c;onservatism is that the' structure will be sub'-

jected to temperatures ab6ve 800 F (: 835 F for.A-A'and 810 Fcfor B-)) for a limited time (about 10' hours in the reactor life) during sone thermal trancients.

6.

The Reachor Vessel was-designed and ' analyzed to a conservative'. seismic spectra. The seismic spectra which are present in the RV' specification were later superceded by spectra which reduced'g-levels by approxf-mately 25%. This reduction wo~uld result in a proportionate redetion

(

in seisroic strasses at the core. support cone Section B B.

7.

The structural evaluation of Sectian B-B was based on a 41/2", thich cora support cone. The actual-minimum thickness of plate used was 4.97".

Additional analyses showed that, in general, the. primary stresses arn less than those used in the nasults for the 41/2"

~ ~

thickness and the stresses due to thermi transients remained approximately the same in both. cases.

SECONDARY M*sBRANE.PLUS BENDING AND, FATIGUE STRESSES i

These ciiteria are generally dominant for cyclic thennal loadings.

In the analyst /s, the evaluated primary plus secondary stress' intensity

~ panges (pL'+' P '+ Q) satisfy the 3S,., stress limit-at Sections b

A-A and F-B in the Core Support Cone'. The _ minimum.cargin due to this N '

stress category is 1.'186 at Section A-A.

Fatkue' dar. age on the core support cone has been considered and is maxim a at Section, A-A.

The total fatigue damage is 0.06 based on the AS".E Code criteria.

Fatigue

.--' dqmag'e at Section B-B is <0.1 and is not controlling on the cene.

~

~.

y n

r

)

5-In summary, the demonstrated factors of safety on primary and primary plus secondary stresses are satisfactory and meet the ASME high temperature Code criteria but could be increased significantly by reducing the conservatisms included in the analysis, such as using the SRSS combination for seismic loads, using actual cone thickness, using stress limits for 750*F, using the operating instead of design differential pressure, using new seismic spectra and adding stresses component by component before calculating total stress-intensities.

.l l

I l

l.

TABLE 1 g

SUMMARY

OF CSS / CONE PRIMARY STRESS RESULTS (SectionA-A)

Loadings Load or Primary Stress Intensities (psi)

Primary plus Secondary Stress or Stress Intensities (psi)

Cri teria Category.

Membrane Membrane + Bendino Pressure 3054 5602 5602 ti)

Dead Weicht 390 351 351 OBE(2) 3189 3757 3757 Loadings SSE (2) 4502 5293 N/A Thermal Stress Range 37731 (Linear)

Pm (3) 6243 P1+Pb()

9359 d.ormal (P1+P +Q)r

'41488 (6)

+

b Upset I4)

K St = 21230 3 % = 49185 Allowable Smt = 14800 t

FS (5) 2.37 2.27 1.186 Pm 7558 Faulted 10895 (SSE+

P1 + Pb AP)

Allowable 1.2St = 23160 1.2KtSt = 25500 FS 3.06 2.3 t

Fatigue I n/n 0.06

< 0.9 allowable per T-1435 of CC1592 (1) Dead weight has opposite sign to other stresses. It is conservatively excluded from the evaluations.

(2) The seismic loads are the summation of N-S, E-W and Vertical.

(3) Nonnal + upset (0BE + AP) are evaluated to Code Case 1592 allowable.

(4) Allowables at 850'F. This is conservative since the temperatures are below 800*F at this region.

o (5)

Factor of safety =,a a c

t'd (6)

(P1 + Pb + Q) range = OBE + Qrange thennal (pressure decays before transient is severe)

@ARD TABLE 2

SUMMARY

OF RV/ CONE PRIMARY STRESS RESULTS (Section B-B)

Loading Load or Primary Stress Intensities'(psi)

Primary plus Secondary Stress or Stress Criteria Category Membrane Membrane + Bending Intensities (osil 11100 Pressure 3800 11100 1100 Dead Weight 1100 1100 OBE(2) 2600 4200 4200 Loadings SSE (2) 5100 8400 N/A Themal (6)

Stress Range (Linear)

Pm 6400 16400

~

Normal +

P1+Pb

<3S (Pl+P +Q)r Upset b

m (0BE + AP)

I3 Smt = 14800 K+S+ = 21230 3Tm = 49185 Allowable

>l.0 W FS (4) 2. 31 1.295

~~

~~

Pm 8900 20600 Faulted p), p (SSE + AP)

Allowable 1.2 St = 23160 1.2K St t = 25500 FS 2.6 1.238 i

I n/n

<.l(6) < 0.9 allowable per T 1435 of CC1592 Fatigue y

(1) Dead weight has opposite sign to other stresses. It is conservatively excluded from the evaluations.

(2) Seismic loads are the combination of N-S, E-W and Vertical.

Allowables at 850*F per CC1592 as in Table 1 (for conservatism).

(3)

(4) Factor of safety = -[

a ed (5) (P) + Pb + Q) = OBE + Qrange thermal (pressure decays before transient is severe)

This result is based on a scoping evaluation using current design transients, (6) which are less severe than those used at the time the component design The scoping study established the magnitude l

analysis was performed.

of the factor of safety and fatigue damage summation but precise values are not available.

SUMMARY

s 1.

The reactor internals are supported by the reactor vessel core support forging which is attached to a 5 inch thick cone which is in turn welded to the core support structure. Welding was performed in accordance with ASME Code requirements as supplemented by RDT standards. The welds were accepted based on meeting radiographic and liquid penetrant examina-tion requirements contained in the ASME Code and applicable RDT Standards.

2.

The pressure loading (normal operating condition) is upward resulting in com-pressive loads on the cone which tends to avoid creation of potential flaws.

3.

The weld joint designs are such as to position the weld joints away from the transition region between the cone and the 24 inch thick core plate and the vessel core support forging and the cone.

4.

The thermal environment in the vicinity of the reactor vessel core support cone is benign.

5.

The material of construction is austenitic stainless steel which has excellent retention of ductility throughout the 30 year life of the reactor.

6.

The neutron radiation induced embrittlement is neglegible as are other environmental effects on the material.

7.

The stresses in the vicinity of the welds are low and substantial margins are available based on conservative analyses. The primary plus secondary stresses for normal plus upset loads have been shown to be less than 3S. The fatigue damage summations are < 0.1.

m 8.

Six locations are provided through the Horizontal Baffle Assembly baseplate which provide access to the annulus between the reactor vessel and the core barrel including the upper surface of the cone, if a future need for such access is identified.

i l

l i

t

CONCLUSION Based on the information presented, it is concluded that the core support -

cone has been designed and fabricated in a manner that precludes the need for routine in-service. inspection. However, capability for future inspection of the core support cone is provided and can be utilized to the extent practicable by the development of under sodium in-service inspection techniques.

J v

i

-)

e e

g i

l

~

4 8" DIA ACCESS PORTS AROUND CIRCUMFERENCE jF

,l Q

')

7 2

N k

HORIZONTAL BAFFLE f

s ASSENBLY s)

(/

A>

Q f CORE SUPPC RT STRUCTURE E

f M

I 9 s; -

i m

N 70 r

o 9

8

\\

j

),

u i

l s

v 4

v REACTOR VESSEL If CORE SUPPORT CONE INLET PLENUM

(

Figure 1 REACTOR VESSEL / CORE SUPPORT STRUCTURE ASSEMBLY

~

8 KETCH SHEET n

r. se o WESTINGHOU E ELECTRIC CORPORATION O

l l

REACTOR VESSEL CORE CORE SUPPORT SUPPORT C0 iE STRUCTURE i

l 45*

161.00 Dia.

175.00 Dia.

t I

Figure 2 REACTOR VESSEL CONE CORE SUPPORT STRUCTURE WELD PREPARATION DETAIL i

1

.------=,,------.--n.--nn-r,----

1 I

l l

i r

l

.e n.*

i l

l INLET PLENUM SODIUM TEMPERATURE RESPONSE DURING A N-4A PLANT LOADING BETWEEN 40% AND 100% POWER

,..p,

l s

l i

l i

j

,,0.

i i

t t

f l

i 4

l i

l l

77.0.

l l

u.

e

.... l s

1, i

700.

a:

h i

I 680.

w l

l

..L.

660.

/

i 640.

1600.

1900.

2300.

2700.

3100.

3500.

3900.

4300.

TIME - SECONDS LEGO O

?

7 7

LCl l

'D O-LC2

=D---G-LC3 l

-b- - d -

LC4 Figure 3

O.

INLET PLENUM SODIUM TEMPERATURE RESPONSE

.DURING A U-2E PLANT LOADING AT MAXIMUM R00 WITHDRAWAL RATE BETWEEN 40% AND 100% POWER 700.

l l

l i

_ __ J _.

.... !.... J..

'l i

i

.,4 0.

e i

i i

.. k......-

I 4

.... -/

i i

i 720.

..l

..i i

i i

i I

',i i

j l

u.

l i

i i

o 700.

g i

l l

._. 3

..L..{..__ _ __. !..

.. _4. _.

14 i

+-

680.

l l

m g

__.. {...j I

t w

l l

8 p--

660.

3

/.

I

/

S40.

(

)

620.

O.

250.

500.

750.

10 TIE - SECONDS t

l FIGURE 4

v INLET PLENUM S0DIUM TEMPERATURE RESPONSE DURING A F-4A SATURATED SIEAMLINE RUPTURE I

I i

i i

i I

y,.,

I j

i

=

.s?.,

g

'/.

\\

l I

000.

i I

l

\\

g 760.

l

.. - / h

.\\

\\

l C

l

</

my

=- -...

j p

p,

} :/

i 720.

/ [ ~.*

i h

t,

\\

I %

yNN

^

a:

I T

s N

[

4 680.

v s

3\\-

x %

h

_ fj

-2%

W 640.

600.

O.

500.

1000.

1500.

20(

TIE - SECROS LEGD O D

b-tci )>

See Figures 7 & 8 LC2 for identification I

.D - G-LC3,,i1 of locations te

.y g.

i i

Figure 5

bn

'l i

k N

h3i on 30 l

J

)

O oo N

)

O oo

=

@ oo

)

a;

. O oo

)

5 c

g

}

F 8

QC l

0 oo g

G:

y-

)

5 D

o v

o

)

l l

)

O oo MC:

)

O oo l

l K

E i

FIGURE 6 Measurement Locations in the CRBR i

Inlet Plenum - Elevation View l

l t

l

.. --