ML20078F967

Testimony of Jl Riley Re Contention 18 Concerning Unanticipated Rapid Embrittlement.Affirmation of Svc Encl. Related Correspondence

ML20078F967
Person / Time
Site:	Catawba
Issue date:	10/03/1983
From:	Jeffrey Riley CAROLINA ENVIRONMENTAL STUDY GROUP
To:
References
NUDOCS 8310110270
Download: ML20078F967 (14)

Text

'

TCi UNITED STATES OF -

NUCLEAR HEGULATORY Cup.M1SStu BEFORE THE ATOMIC SAFETY AND LICE 5 SING BOARD SYR'N In the Matter of ) .b DUKE POWER COMPANY, ET AL.

dJU 6 M

) Docket nos. 50-413 (Catawba Nuclear Station,

)

)

50-414 gS$$yGid co i

ERt.ny"W ~.

Units 1 and 2) )

' t

. l DIRECT TESTIMONY OF JESSE L. RILEY FOR DESG  !

( ';;, RE CONTENTION 18 Please state your name and place of residence l

My name is Jesse L. Riley. I live at 854 Henley Place in Charlotte, North Carolina, ZIP 28207 l

Are you a member of the Carolina Environmental Study Group?

Yes,. I have been a member since 1970.

What is your purpose in testifying in this proceeding?

~

I am testifying in support of CESG contention 18. It states, as clarified and admitted, that the Catawba reactors have not been shown to be immune to the unanticipated rapid embrittlement with ,

use found for a number of other. reactors, including Applicant's l Oconee units. 4"The Staff has interpreted this contention as a -

claim, principally, that the NRC's projection of the amcunt of j increase in reference temperature RT which results from neutron l irradiation damage, isnonconservatik,thattheamountofreactor material degradation for the Catawba reactor vessels cannot be accurately measured, and, as a result, that there is not reasonable assurance that the Catawba reactor vessels can and will be operated within acceptable safetytmargins for material degradation." I have no quarrel with this explicit reading of the contention.

Is your testimony to be an attack on the regulations */

l No. The inte$t of Appendix G--Fracture Toughness Requirements,  !

and Appendix H--Reactor Vessel Material Surveillance Program  :

Requirements, 10 CPR 550, is laudable. These appendices seek to -

assure that reactor vessels do not fail because embrittlement, in interaction with other factors involved in a potential failure, has increased to the point that the reactor is in jeepardy. The means,. "

of seeking this desirable end are technical. The relevant technology is not mature as evidenced by anwndments to these appendices in July, .

1973, February,.1976, April,1976, and in 1983 My concern is with the inadequacy of the technical provisions to reach the intent of the regulation, the avoidance of reactor breach. .

The subject is metallurgy. Are you a metallurgist?

No. .?

8310110270 831003 J*

.1?'?

PDR ADOCK 05000413 -

3)N.Ch[fg .

. i What, then, are your qualifications to testify in this matter?

I have been employed in chemical and physical research for I 45 years. As my r.esume4ahows, I have worked on a great variety of problems. As a staff ^ member of a large research organization, I have heard discussed and read reports on an even larger number. i Not only as the conductor of personal research projects, but as an internal consultant I have had many opportunities to perform as a ,

scientific generalist: to see whether the factors under consideration bore a reasonable relationship to the problem;; to judge whether all the key variables appeared to have been identified; to see if the relationships between factors appeared to be such as to S i ve the desired outcome. I have been interested in the question of reactor integrity since 1971. I believe it will be useful to the Board to [

consider my views as to deficiencies in the current Staff attempt to provide a reasonable assurance of reactor vessel safety as it is affected,_ inter alia, by a more rapid embrittlement of the welded i reactor base metal than is presently assumed, and a proposal for the ,

nondestructive testing of reactor vessels which will provide a higher degreee of assurance than the nondestructive procedures employed in Section XI of the ASME Boiler and Pressure Vessel Code, Rules For Inservice Inspection of Nuclear Reactor Coolant Systems.

What specifically in your research experience bears on the present matter? 1 I have prepared fibers with a wide range of physical character-istics from several polymers. The differences in properties relate both to chemical and to morpholoE ical differences. Although not '

identical, there are marked similarities in the properties of polymeric strucures and metal structures. I have tested fibers under a variety of conditions: ambient conditions; wet; hot wet; over a spectrum of strain rates; over a spectrum of cyclic load and relaxation patterns; and in cyclic loading fatigue. Fibers show brittleness phenomena.

They exhibit transistions in state, including the commonly referred to glass / rubber transition. A test similar to the Charpy V-notch test is performed on polymer specimens to characterize brittle / ductile behavior. I have*been' involved in high precision measurements and am well aware of the considerations in characterizing accuracy. j I routinely make statistical characterizations of data, noting such factors as variance, the significance of differences, and' confidence limits. Finally I-have studied at length the kinds of flaw causing the tensile failure of fibers. f,  :

Where would you like to begin? ..

First by demonstrating that the* Staff realizes the limitations  !

of the regulations strictly to define reactor safety 're  ;

For example, .in the SER, Section 5.3.1.1, pp'. 5-15 to 15uirements.

, compliance - i with Appendix G,.10 CFR, six exemptions are stated an'd justified. l The Staff standard in regard to co=pliance, pp. 5-13 and 14, is one  !

of " reasonable assurance." This is hardly the language of complete  !

certainty. ,

.i '

Will you describe the subject of your remarks? --

w The subject is the reactor vessel. The primary structural elementa.

- ~escc .- .

.m

of both the vessel and the vessel cover, or head, ar.e a group of ferritic steel forgings joined together by welds. The head and the vessel are joined together by flanges which are bolted so as to effect a seal with 2 metal 0-ring s.

What is a ferritic steel?

The kinds of ferritic. steel which may be used in a reactor vessel forging are described in the ASME Boiler and Pressure Vessel Code, Material Specifications, Part.-A--Ferrous. For example the forgings in Oconee reactors correspond to SA-508. It applies to quenched and tempered vacuum treated carbon and alloy steel forgings for pressure vessels. Four classes are' described, differing in carbon-contents and alloying elements such as manganese, phosphorus, sulfur, silicon, nickel, chromium, molybdenum, and vanadium. In view of the important role now attributed to coppsr in accelerated RT NDT incrasee with neutron fluence, it is interesting to note that copper #1evel is not a requirement for SA-508. The vacuum treatment reduces the content of dissolved gases in the melt and makes for a less porous metal. Tnere are also tensile requirements for the-four classes with tensile strengths ranging from 70,000 psi to 105,000 and yield strength from 35,000 to 85,000 psi. A ferriric steel is magnetic, corrodible, and requires protection from the reactor coolant.

But the most critical property of a ferritic steel in this context is that it has two modes of failure under stress: brittle and ductile.

What is the difference betwe'en brittle and ductile failure?

In brittle failure separation is at the grain bouncaries. The extension before failure is small and the load at break, the strength of the piece, is relatively small. In ductile failure the applied forces, tensile or shear, propagate through the grain boundaries.

the crystals extend, deform. When break occurs the failure surface is not crystalline and bright. It is dull. The break is called fibrous, a characterization of the appearance of the stretched out crystals.

Mhat determines'whether a break is ductile or brittle? .

A number of factors. The com history and mechanical processing. ' position, Chromium chemical, thermal at some elevated temperatures, precipitates out in' the *graih boundafy . region as the carbide. This increases brittleness. At another elevated temperature precipitated chromium carbide particles redissolve and move from the grain boundary into the crystals, reducing brittleness. Test conditions are a factor. The more rapid the application of stress, the more likely a brittle failure. Test tempersture is a key factor. The transistion from brittle to ductile failure in a test like the Charpy V-Notch test occurs over a broad and variable rgnge of temperature.

For example, a break may be 100% brittig at -40 F.. It may not become 100% ductilg until a temperature of 140 F is reached, a transition span of 180 F. Othergpecimensofnominaglythesamematerialmay span as little as 160 or as much as 240

-g-What factors in brittle / ductile transition are of particular concern to you?

There are many factors. I am greatly ccncerned that they are not addressed by staff. First there is the relevance of the samples to the most probable point of reactor failure. In all likelihood the point of most likely incipient failure (PIF) is in the region of a weld. The Charpy samples are cut from coupons which are both unirradiated and irradiated. The base metal is tested. Heat affected zone (HAZ) base metal is tested. The weld metal is tested. But welds are not tested. Welds are subject to porosity defects. And pores are accepted under the code, see ASME B&PV Section III, Appendix VI, VI-1000 Porosity Charts, particularly those for four inch weld thickness,1971 Edition, p. 501. Not only are welds porous, there is the possibility of slag inclusions. Indeed slag inclusions were found in Duke purchased reactors using an improved ultrasonic test method, were cut out, and rewelded. There is also the question of the intimacy of the weld metal / base metal bond, a juncture which can be seriously weakened by the presence of the thinnest of layers of an oxide, of a flux residual, of slag. And then there is the matter of the changed properties of the base metal in the HAZ. Such changes are not uniform. There is no assurance that the sampling of coupons indicates in the slightest the worst case HAZ's in the reactor vessel.

An indication of the variance in HAZ Charpy data is given by a comparison of Figures 5-3 and 5-4 from BAw-1697, a Babecek and wil 18 report on the testing of Oconee Unit 3 coupons irradiated to 3 12x10 nyt. Attachment A.

Is there any further consideration to the matter of Charpy sample relevance?

Yes. Cracks are the bugbear of~ welds. Cracks can form because the weld metal is intensely hot, a melt, in relation to the base metal.

Even if.the coefficients of thernal expansion tre alike, the much greater cooling of the melt rerults in much greater contraction on reaching the same temperature as the base metal. It is part of proper

( welding practice to stress relieve welds at an elevated temperature I

over an appreciable period of time. The result is a great reduction in cracks, but not total elimination. The welding crack problem is very much a subject of awareness on the part o'f ASME experts and of Staff.Section XI of the ASME Code provides Rules for Ins ~ervice Inspection of Nuclear Reactor Coolant Systems. 'From the Forewcrd, "The areas most predominantly selected for examination are those associated with welds in the pressure containbypomponents." (p. (e))

However even here inspection is not complete. The inspection ". . .

shall cover at least 10 percent of the length of each longitudinal weld, and 5 percent of each circumferential weld." (Table IS-251)

The method is to be volumetric, i.e. sample throu6h the material l rather than be confined to near the surface, a measure of the weight l given the inspection. (Table IS-261) But inspection may not be l adequate for another type of deficiency, cracks in the austenitic l

(stainless steel) corrosion resistant cladding applied by welding to l the interior of the ferritic shell. Assurances as to the integrity of the cladding go back to the procedures presumably followed in manufacture. Inspection of the cladding is only visuti (Item 1.14, Table 13-261) and confined to 36 inch square patches, 6 in the closure head and "6 patches, evenly distributed in accessible sections of the vessel shell." (Table IS-251, I-1) The SER expresses concern that l

~$~

cracking of ferritic components "will not occur durine fabrication and [will] minimize the potential for subsequent cracking." Emphasis supplied. (SER 5 3 1.1 (5)(a), p. 5-14) Similarly the SER relies on the process to provide " reasonable assurance that underclad cracking will not occur during the weld cladding process." There is no reference to what may happen later. (ibid. (b), p. 5-14)

Are there some limitations to the Charpy test for nil ductility reference temperature specifically related to irradiated specimens?

Yes. Before the reactor has been used it- has neither been irradiated nor stressed. After service, it has been both. Although before stress the Charpy specimens- bear a cle'se relationship to the base metal, except in heat affected zones, this is no longer true after stress. A reactor is expected to go through 200 cycles of heating and pressurization,coolin6 and depressurization, in its service life. (FSAR) These cycles alter the physical properties of the ferritic base material. The absolute tensile stress of the ferritic component or the vessel is reduced to 20% of its initial value after 200 cycle's of design loading (ASME Code,Section III, Fig. 109-1, Design Fatigue Curves for Carbon, Low All.oy, and High Tensile Steels);

the absolute tensile stress of the austenitic cladding is reduced to 25% of its initial value (ibid. Fig. 1J9-2, Design Fatigue Curve for Austenitic Steels . . .) It is the properties of the _ stressed, irradiated vessel materials that are relevant, not the " material property" of the unstressed but irradiated coupons. Add to this the realization that the weakest parts of the reactor are not represented in the test program, that corrosion and stress corrosion can occur at cracks not detected in the inservice inspection program, and the conservatisms in the program, fourfold higher flux for the coupons than for the reactor, and various arbitrary add-ons to the RT , and there is very.little reasonable assurance that there is any rISItion between the _ Appendix G and H programs and the actual physical espsbility of the reactor as the end of its forty year life is approached.

Are the cyclic stresses you refer to limited to pressurization and depressurization? .

No. There are thermally induced stresses of two kinds, one due to the difference in coefficients of thermal expansion between the ferritic and the austenitic components of the reactor, the other due to the temperature differences, the temperature gradient, between the inside and the outside of the reactor. According to the ASME Code,Section III, Appendix I, Table 1 5 0, the mean- .

coefficientsforthermal'expansionofferrgticandausteniticstgels are, res in./in./pectively, F. Thisfor is the range 70ofto36%.

a difference 600 F,This 7 23 and 9.82 amounts to axdifference 10 of 0.14 % in relative extension and, considering the high moduli, causes a substantial stress. The temperature gradient effect is more complex. It affects the ferritic base metal and the austenitie geladding in similar ways. When the reactor is uniformly cool, say at 70 F, and assuming (optimistically) that it is totally stress relieved at that temperature, there is no differential stress between the inside of the reactor and the outside. That is,the outer layers of the reactor are neither too large nor too sna11 for the inner. Disregarding ,

pressurization, as the temperature inside the reactor is brought up

a temperature difference' develops between the inside and the outside. i The inside, being warmer, expands more. The outside is no longer Isrge enough for the inside. An internal stress develops. And, as the cladding has a larger coefficient of expansion than the ferritic base metal, the austenitic cladding becomes increasingly too large for the ferritic base. A further stress is added. When the reactor reaches a steady state operating condition this temperature gradient persists. Heat is lost from the exterior of the reactor. A stress contributing gradient remains. This gradient, which I shall call positive, increases, adds to the stress caused by pressurisation and is a potential driving force for the growth of cracks on the outer surface of the reactor. fhe expansion forces act to reduce the potential for crack growth st the inner surface of the reactor.

~

In reactor cooldown, if slow enough, a positive temperature gradient can be. maintained. However in a faulted condition, in which cool water is pumped into a scrammed reactor, the temperature gradient can become negative. Now the inside surface is too small for the outside. There is a potential for crack growth. If the temperature is in the ductile range, the leading edge of a growing crack will be -

blunted. But if the temperature is in the brittle fracture range the crack will grow. It is this possibility which makes 1) an accurate determination of the nil ductility temperature of the reactor material i'tself so important and 2) mmxes it essential to determine the effects

.of stress fatigue as well as neutron fluence on the nil ductility temperature of reactor material.

What is the basis for your concern about unanticipated increases in RTg9g?

The earlier versions of Appendices G and H support the egnelusion that tne Staff did not anticipate RT normally exceeding 100 F.

Referring to CESG's Opposition to Aphk,icant and Staff Motions for Summary Disposition of CESG Contention 18/ Palmetto 44, Aug. 15, 1983, see answer to Staff asserted material fact 6 (p. 3) and CESGr a asserted material facts 3, 4, 5, 6, and 7 (p. 7). This unanti'cipated increase with neutron fluence has been attributed to the levels of copper, nickel and phosphorus, CESG material fact 8 (p. 8). A lower level of these three constituents will, the Staff believes, result in a slower rate of increase of RTNDT, CESG material facts 9 and 10 (p. 8).

Do you disagree with Staff?

Not necessarily. It is rather my position that the variability l on which Staff would rely, and the

in the determinations of RT l experience with neutron flu $b,oe on the performance of reactors low l in copper, nickel, and phosphorus is so. limited, that no reliable conclusions can be drawn. The Staff has used several methods to l arrive at estimates of RT including Regulatory Guide 1.99 and l

the Guthrie formula. The kIscrepancies between the two results are frequently 1'arge, see CESG material fact 8 (p. 13). The large variance in the Guthrie formula is characterized bv a standard degiation of 24 F. As a conservatism as to Staff adds two standard deviations, attain a confidence limit of 48 F to its estimates of RT s out of twenty s

the true value of HT 95%. This means that 19 tidST should be below the Staff value. Is a confidence level of 95% soo2T enough? Would not a confidence level of 99.9999% be more appropriate

because we ire'desling with the prevention of reactor breach? If we agree that a higher confidence limit is required, then we are faced with the agsurdity of a 5 standard deviation add-on to the estimated value,.120 F. The result is manifestly of no utility. But tne absurdity of using it pales before the absurdity of assuming that an extraordinarily imprecise, and consequently at least as inaccurate, value determined on samples which do not include weld metal interfaces, which experience no cyclic mechanical stresses, no cyclic stresses due to thermal gradients, and which in no way relate to the weakest part of the reactor will have any predictive value in regard te e reasonable assurance that the reacter will not breach.

Is this the only absurdity relafed to RT EDT

's?

No. Another absurdity is the use of these numbers of essentially no significance on which to base very precisely delineated Pressure-Temperature Limit Curves,- two of which are shown for Oconce, Unit 3 Attachment B. -

What is the safety significance of these curves?

If the reactor is-not in a faulted condition, i.e. if it responds normally to operator actions, it is possible at any temperature in the normal operating range to maintain the pressure sufficiently low so that the mechanical stresses due to pressurization are not likely to cause breach, even though the reactor may be in a brittle state.

If the pressure limit curves were based on accurate determinations of RT w uld they make a contriuttion to reactor safety?

NDT

. Not necessarily. They would, and even now do, contribute to a false sense of security. As I have mentioned previously, the RT does not reflect changes in the material of the reactor.itself br50[htaboutbymechanicalanddifferentialthermalcyclicstresses, l

nor, certainly, the most flawed spot in the reactor. Beside that i they have an additional deficiency. Even if they were completely l appropriate, under certain faulted conditions the control of reactor coolant temperature in relation to reactor pressure falls outside the operator's control. A large leak, one sufficient to turn on the emergency core cooling system, will result in cold water entering l the reactor. If the pressure in the reactor is high enough the l actual condition for brittle failure will be reached by the most l vulnerable--flawed and embrittled--part of the reactor. A PORV which opens and hangs up in the open condition could initiate such ,

a sequence. If the reactor cools appreciably and the block valve in the PORV line is closed there will be combined the condition of i low reactor temperature, rapidly reached, and high pressure.

l Given the ills you describe, have you any remedies to offer?

The unanticipated rises in RT 's at a number of reactors has had the benefit of calling the probIkks of reactor f ailure to the l attention of those outside the industry and the NRC so that the weaknesses in the program become known and receive consideration.

Have you anything specific to suggest?

1 __

Yes. The hidden flaw has the greatest potential for causing reactor breach. The hidden flaw is more likely to give an indication of its existence under operating conditions rather than cold and depressurized. Strain gages are a common, widely used instrument of great sensitivity and reliability. I would recommend covering all of the most vulnerable areas of the Catawba reactors with a grid of strain gages. These gages would track the full len5th of each weld and extend through the heat affected zones to base metal. The outputs of the couples would be continuously scanned and exceptional -

indications would be indicated and alarmed by an appropriately programmed computer. If, for example, over the length of a longitud-inal weld passing through the belt line, the normal deflection at operating pressure and temperature was 7 mils per. inch at the junction of the weld metal with the HAZ, an increase to 10 mils

~

over a length of 6 inches would be cause for alarm, shutdown,-

examination, and repair. The advantages of a system which keeps the potentially most vulnerable zones of a reactor under continuous surveillance are apparent. Both slow and rapid changes can be tracked.

If small changes in the elastic modulus occur during the brittle /

ductile transition it is conceivable that these measurements could give real time information as regards the increase in the actual nil ductility temperature of the actual stressed components of the reactor.

O

, ~

Attechment A-1

t. .

Figure 5-3.

Charpy Impact Data for Irradiated Base Metal, Transverse Direction c.

100 g g g g g , , , ,  ; g a 75 -

.LJ E

5 50 - - - _ - - _ _ _ - - _ _ _ _ _ _ _.__. ___ ______ _ _

h 5 25 - _

p q

^ ' I I I 0 I I I I I I f

r .08 p g I I I I "

i a e i 1 i LJ .

!  !.06 -

• ]

a f

~

w .04 -

, e , ___________________________-

'ys E /

j .02 -

m

~

i

' a o

I I I I 1 I I F"1 I I I i 1 I I i i i i i DATA Sifr%RY

{n 180-T oy --

Lm TeyG5 mLt) 22F 160,, Tc , (50 n-ts) **F yn TeyG0 n-ts) toF _

.Lj Cy-USE (Avo) 100 FT-LBS

y. 140* RTci -16F r~'

La -giu- _

O

.5 100 -  :

U

.5 80 -

O E

r

~ 60-g ._______

L .4 40* -

-__-_ MTERIAL h ECATim TR Alt 3 Y E R S E-~

20

_ FLuth:t 3.12 Ela n yt -

a

. HtAThe. Avs-is2 0 I I I I I I I I e v E

i

-CD -40 0 40 50 120 150 200 240 280 320 350 400 it:T Ten tRATURt, F

~'

5-9 Babcock & Wilcox

~

Attachmsnt A-2 Figure 5-4 Charpy Impact Data for Heat-Affected Zone, Transverse Direction 100 g 3 3 :j i  ; 3 , 3 i

" 15 -

E '

O

= 50 - - - - - - - - - - - - - _ - - . _ . . _ _ _ _ . _

w

o y e e .

a 25 -

e

- I I f I I I I I f I I c

• 08

• i i y

• I ei I I I I I I I e -

E - -

.06 .

j e - I e .-

e

_ g E

i .02 -

• _ =

5

.5 I I I I I f I I l f I o

l 1 1. 1 I I I l DATA Strt%RY 180- Inor -

9 Tey (35 nu) -

g_Tey (50 ri-ts) - _

TeyI30 rT-ts) -

Cy -USE (avs) -

"IU~ RTNDT ~

E e '

i 120 -

i C .

I 8 ,,

.:!100 .

a l 5 l i5 80- e __  ;

E I

- 60 -

40 .

• '" ^*

. _ _ - - _ .a. _ - _ _ _ _ __ _ .

e Ont(NTATiow Traws. "A7 20 . Ftutact 3.12 Ele pvt -

ETAT No. Avs-192 0

I I I I I I I I n t e ll

-m -40 0 0 80 120 ISO 200 240 280 320 3 60 400  ;

itsi TrecuTunt, F 5-10 Babcock & Wilcox j i

. Attachmont B-1

.}

m, p

i .J e

. Iw w

a

- y,

., .=.

c m w w

~ .

• A -

C o w a O

- =

a - . . n a o

- w - .a '

W o E U uw .-

- , _. 2 - . a

> c U .e

.Z am I,,.

gg J K er w - = -,o -

om - ., .

e - < ww C OO x-" ' ' '

, o , x ac a N

- c -

W E C kk E y o mo= c , ce M 0 V .* , = , #9 o er aY z n -m 6 C ,w- o .o

• n , er v .- - w . ,-

c .E w - >- > w o sa

> p. wwoo * = = = z*

y p o en o r ao *= a

- o -

uw

u. w .- v .-,

wo x - .x, o.o .,- o

~

u te o. .

ow

.o-x--- ---o- .

Eu u o o _,. er

-e .., o _x o = , w a, o E '. . Ao e g,,j *= E o er u =oo = e- -

o - - , ec o

..-e

,e a , a w

. - . - e .o.-

2 w x-o .- -

s C w

• e- ~n  ;

cw y

e

,. ws. -

w

.2

, > x er a ck g y er er

• o em C= - t.a c n - EE, m

E .C == v z u e-cc y .o w e-w = s, w w w

.[ - a es u g w ..er 2 - -

o C.

c. - ., . =-

u c. . .-e.o ~ oo 3 C e e< - , , x o e 4 m c a .

N e

. - - - -- o .s

.A, ,E E

SJ

.4 4e u. wo. E, e o . e.r- o .s.

W g a. - w es e- O g z es --.

8D e w o w wa - wo eo gooo ,

o - - - - o -m- .o -

o v- - -

a

> a wu C .x, w

o .o n n

-~~~e

.n . ,

==

~

0 m *

• U O ' '

ow _ - , ".- .

, o eo w , . ~

m .a., .n .n .e ch

m. v.Ioo c: =: . .a o - - ~ ~ o .o

. ~ .n . .

. . . . w ~

K ., J W e.s at

. - - - , ~

< er a o w w er I

CC o = =

• w w e x a , -

-- . , . - c2 w . as - - 4 .uow =

3 ++ w w , a c4 4 e. m u o .o.

g

.r4 f f I f f f f I I f f f o,

o o o o e o o o -o o o o o o C Q Q o o C Q Q Q o o Q o

= , m = = , m o = , m

m. m m ~ - - .- .- -

Ilsd 'aJnssaJd lueloc:) lassaA Jolorag 8-8 Babcock 4.Wilcox ((

. W

m m I 5 i- t t i t._.,_.; i r r r i i e w , ,

Figure 8-3. Reactor Vessel Pressure-Temperature Limit Curve for !1ormal Operation -

Cooldown, Applicable for 15 EFPY (Oconce 3) 2400 ASSUMED RT MOT, F ]

2200 _ BELTLINE REGION I/4T 189 BELT (INE REGION 3/4T 136 ,

CLOSURE HEAD REGION 60 '

2000 -

06TLET N0z7tE so en

" PRESSURE. TEMP, 1800 PSIG F J P0lMT

$ A 210 70 y 1600 -

8 710 220 g C 1940 280 0 2250 315 C u

l c 1400 -

as o j

1200 -

.1 b 1000 - ,

APPLICAstE FOR C00t00WM RATES UP TO 100F/h U

B00 a:

600 - 1 i

THE ACCEPTABLE PRISSURE-TEMPERATURE COMBIMATIONS ARE I!ELOW AMD 'TO THE RIGHT OF ca 400 -

THE LIMIT CURVE (s). THE t.lHIT CURVES 00 NOT m

IMCLUDE THE PRES $tfRE DIFFERENTIAL BETWEEM THE

[ POINT OF SYSTEM PRESSURE HEASUREMENT AND THE c

8 #

200 PRESSURE OM THE REACTOR VESSEL REGION CONTROL-F

" LING THE LlHIT CU4V'E OR ANY ADDITIONAL MARGIM I I I I r SArtu g R r S m lE,iMSuuntMT r,R R O R . bl-l

! O 14 0 80 120 -160 200 240 280 320 360 D, s ,

8 \

n Reactor Vessel Coolant Temperature, F to !

i IV

l

'N?,bf UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION agy CCT

-g All:37 BEFORE THE ATOMIC SAFETY AND LICENSING BOARD 00 UCb, cr hf[hf05l.

BR c  ::

In the Matter of )

DUKE POWER COMPANY, ET AL. Docket Nos. 50-413

) 50-414 (Catawba Nuclear Station, )

Units 1 and 2) )

AFFIRMATION OF SERVICE I hereby affirm that copies of Direct Testimony of Jesse L. Riley for CESG Re Contention 18 in the above captioned proceeding have -

been served on the following by deposit in the United States mail this 3rd day of October, 1963, with the exception.of addressees marked by an asterisk to whom copies will be delivered October 4

• James L. Kelley, Chairman

• Robert Guild, Esq.

Administrative Judge Attorney for the Palmetto Alliance Atomic Safety and Licensing Board P. O. Box 12097 U.S. Nuclear Regulatory Comission Charleston, South Carolina 29412 Washington, DC 20555

, Palmetto Alliance

• Dr. A. Dixon Ca11ihan 21351 Devine Street Administrative Judge Columbia, South Carolina 29205 Union Carbide Corporation

! P. O. Box Y

• Carole F. Kagan, Attorney

, Oak Ridge, TN 37830 Atomic. Safety and Licensing Board Panel l

U.S. Nuclear Regulatory Comission

! *Dr. Richard F. Foster Washington, DC 20555 l Administrative Judge l P. O. Box 4263 Sunriver, Oregon 97702

• George E. Johnson, Esq.

l Office Executive Legal Director l

i lichard P. Wilson, Esq. 1U. S. Nuclear' Regulatory commission l Assistant Attorney General Washirigton, DC 20555 l P. O. Box 11549 Columbia, South Carolina 29211 , William L. Porter, Esq.

l Albert V. Carr, Esq.

,J. Michael McGarry, III, Esq. Ellen T. Ruff, Esq.

Debevoise and Liberman Duke Power Company 1200 17th Street, NW. P. O. Box 33189 Washington, DC 20036 Charlotte, North Carolina 28242

-. - . - . - .= . _.

w e .

' Atomic Safety and Licensing Board Panel ..* Don R. Willard U.S. Nuclear Regulatory Comission Mecklenburg County Department Washington, DC 20555 er Environmental Health 1200 Elythe Blvd.

Atomic Safety and Licensing Appeal Charlotte, NC 28203 Board Panel U.S. Nuclear Regulatory Comission Karen E. Long' Washington, DC 20555 Assistant Attorney General N.C. Department of Justice Docketing & Service Section Post Office Box 629 Office of the Secretary Raleigh,.KC 27602 U.S. Nuclear Regulatory Comission .

Washington,'DC 20555 ,

04AL -

Jfsse L. Riley fo M ESG l

I

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