ML20087B846

Ao:Leak Detected in Steam Generator.Caused by Cracking of Tube Seal Welds at Tube Sheet Due to Separation of Cladding

ML20087B846
Person / Time
Site:	Robinson
Issue date:	06/19/1971
From:	Beatty G CAROLINA POWER & LIGHT CO.
To:	US ATOMIC ENERGY COMMISSION (AEC)
Shared Package
ML20087B746	List: ML20087B773 ML20087B846 ML20087B898
References
NUDOCS 8403090381
Download: ML20087B846 (2)

Text

'

, ,. , ~ ,

q , fm.

• .e \j k )

i

'i. l hi -

\ j.

. Qs e

~=*

~

j 4 n/7/

! 1,

. l lphf Jo sia Dsuks - Di)?.ecrwe 7i. ^ Coinp/ kwc.s Or;=lce l A. .fiq. A% sio.o

, ' ' S o 17.2= F/V

,,:2 3 c)

PEsu/nw.d si ) /V oxni Eest t'

?/lT/-$/0 7s , G a 3osa3 s

l. .

1 Ob O /5 OfkSNf /5$? lf 40/ 0lY

\ t dYA//CE fAJ5 A74?E C)/J4/f.r E / o 3.cNC .4 i

to co n,o.<y eu )7A D PR - 23 faxsax^m . .i c T,sceincri .L S,ascific p ris a 4. G./,o, 4)2

' l /A opsyou Accapr 7%r #swoces:v:sa , w// n I l .4 s .4 .So .c. sri ru 72 . .

I

, /9 Aasa sss /5.e.e,0 r).er.ser:eo / a' n I,37~.en M G s w 5.< n ro.sc, B is /~;4//s o .opaA i '

///$ NO/ZM///- 0 cc-uAAEwcr Dr.=. t=r su ) r/ c. /J o38 /, 8' <2. . T~jiE , O M o , 6 's . G / w 57 d2NJ4- t Ac/ css 77hd am o.

C O s 0 Y A A C- T O M '-< b / [ W8,3 85E. 5O o h f 0 C/ A k IS Sk$ l6 l U5 To U kDk//$

et .

OE &5 Tur35 $Ec'1/ Cd]5 /~Of A/~ 0 5 ?U O 5 '

o ci W

0tr

. d//22=7' Ou5 To /) 32,c044 77 oa of 7~i2 t Ou.:

M [/ A 00/

6~ FA 0M QE Tu/Js S /.sdrY / y As b

?

.$.d2,,0AM/)7Sdd IV.fr A7.2,6,</ /=o esM O rY Tb/o

.t i

- pppp_g ,

9

p. ,- ,

'j ( g'

. C/

. )A ,

\ .'

j 4 ll i

$V&Aff hEr:./L'2-M A70AS

• bO5 AAO f A 6 c G.G O's/J G c o i D> /Cf/?/Rd.

, Ylfd . hod l(j'A K l2/5 M/ /C 0fl0OM

, l l6in er: . H >1.s q.c eJ Kaps- - p=u //,y iNr m 2cc

{ l[SAoM D-17 ~f fr.ts' CA.) 2 fr A C Y J c.e c ,0. 2 C 7 2 D (2 } $454 /E, h 0 2 sm 9 h /~ G D A s2 ,47 0 A Y '(A J / / /

!fovro w w isa.4 s u n ; usw.

il 1 . .

' - s l gciuses / A'o/ ,Y a

i

/

I

%#y? + .

. l kI'sf/

l Af' r

1 f

1 I

l - .

i e is I

' Exhibit C t

e s j ,*

Page 2 of 2

. - p 7 .s 4; o o ,

L June 25,1971 Dr. Stanley Weiss _

4841 North Oakland Milwaukee, Wisconsin 53217 Mr. William C. Seidle Senior Reactor Inspector U. S. Atomic Energy Commission 'i Division of Compliance-Region 11  ;

230 Peachtree Street, N. W.

Suite 818 i Atlanta, Georgia 30303

Dear Mr. Seidle:

In reference to Report No. DC 89, " Inspection of Leaks and Cracks in the (A) Steam Generator at H. B. Robinson II," June 23, 1971, I would like to offer the following opinions and recommendations:

(1)- Westinghouse is primarily placing the blame for this leakage and cracking problem on the inadequacy and reliability pro-blems associated with the explosion bonding cladding process.

There is justification in placing high suspicion on this process.

Experience has indicated that inconsistent bond integrity has ,

been obtained on large, relatively complicated parts. For example, bond adhesion problems might be expected to localize in the central and outer euge portions of a component such as the steam generator. Although the basic principles and po-tential of this process have merit, the application of the process to various components often presents considerable practical difficulties.

The process dates back to the late 1950's and early 1960's and, is generally based on the Cowan, G. R. , Douglass, J. J. ,

and Holtzman, A. H., U. S. Patent No. 3,137,937 ise aed June

~

3,1964 and assigned to E.1. du Pont de Nemours and Co. A ';

typical cladding arrangement, as _well as the basic operating principles are illustrated in Figure 1. After the explosive is' detonated, the prime metal is_ accelerated very rapidly to a-

, high velocity (V nation front (D)p) byacross moves the detonation pressure.

the plate, an As the deto-angle is establishedL h

between its undeflected and deflected portions. When the-deflected portion of the prime metal collides with the backer l

0h gVt plate, the region of high-pressure collision moves across the -

plates at high speed. This velocity (V )c equals the' detonation-

@; . % 07p N2/ . ,

f r.

+

{,

, n v L l'

p ., 2 velocity for the illustrated situation. Ahead of the collision region in each plate, high pressures cause surfaces of both _

metals to flow plastically into the space between the plates.

The jet which forms acts to remove surface films of oxides and of other materials normally detrimental to bonding.

At the same time, the clean metal surfaces are subjected  :-[

to high pressures in the collision region causing plastic  !

deformation. The metals supposedly come into interatomic contact with each other, establishing a metallurgical bond.  ;

t Furthermore, difficulties have been experienced in the 9 D

past in arriving at destructive testing techniques which result in obtaining reliable bond strength information.

Based on this failure and past experiences of manu-facturers of large vessels, it is recommended that past and intended future application of this process be thoroughly reviewed.

(2) It is the writers opinion that the following factors contributed.

to this failure:

1 (a) Marginal clad bond integrity resulting from the explosive cladding process.

(b) High residual stresses resulting from welding

of the divider plate to the clad tube sheet. These stresses probably further deteriorated the clad bonded interface and initiated separation.

(c) Operating and applied stresses then may have led to propagation of the clad bond separation resulting in failure and leakage.

(3) Undoubtedly boric acid solutions and residues have, by capillary action, penetrated up the sides of the tubing and along the propagating cladding cracks. At ordinary temperatures, 750 F,

,' 'the boric acid solution is only slightly acidic, reflecting the weak ionization of the borate ion. The:corrosivity.of the solution at this temperature toward ordinary system materials, e. g. , stainless steels,inconel is insignificant

. ' and can be disregarded. However, the_ existing condition can present problems because of the contact which has

- been made with the underlaying forged steel tube sheet _-

and the potential of higher temperatures. Also, aereated boric acid solutions resulted in significantly higher rates h

em

(')

r

s G U 3 b

l of corrosion as compared to deaereated boric acid solutions.

It is recommended that efforts be undertaken to analyze the potential hazards of this condition and the difficulties j associated with eliminating the condition.

(4) It is recommended that the stress analysis currently being performed by Westinghouse be submitted, as soon as possible, for review by an independent, outside consultant (e. g. , .

Parameter, Inc. ).

(5) It is the opinion of the writer that generic aspects of this problem exist. Evidence of this has already occurred at the Robinson 11 site where two out of three steam generators already exhibit the problem. Furthermore, as a result of our conversations, it was indicated that the problem is imminent at other sites.

(6) The following observations opinions and recommendations are offered regarding the Westinghouse proposed repair procedure:

(a) The removal of defective areas should be per-formed in a careful and planned manner so as to enable a rigorous fracture analysis.

(b) Although the repair completion was estimated as one month, it is believed that this estimate is overly optimistic. The man hours required and working conditions imposed will probably result in a multiplication factor of 11/2 to 2 times the estimate.

(c) No mention was made in the procedure of methods or attempts for removing the trapped boric acid solutions.

(d) Detailed "in-process" and ';' final" inspection pro-cedures were not described. Obviously far more than just the code requirements will be necessary for this repair. Mention was made of a secondary hydrotest as a check for leaks.

(e) Westinghouse is arranging to subcontract the localized stress relieving operation (Cooper Heat was men-tioned). They will attempt to accomplish a localized

. heat treatment of the deposited repair cladding by designing and building equigment capable of heating the area of interest to 1000 F. Westinghouse claims wide experience in accomplishing this type of weld stress relieve. This localized stress relief will result in high thermal gradients and stresses. Efforts I~

}

o 3.+:.

.- /$

9 4  !

i should be made to determine beforehand the effects this operation will have on the surrounding hardware _

(e.g. , remaining explosion bonded interfaces, metallurgical effects in the thermal gradient region, etc.).

O Ver truly yours,

-/ , m I

Dr. Stanley eiss W h cc: DC-89 file ,

S. Weiss '

Lett. Comm "A" file e

T 9

's L

V l'

l

.y

.y

' h' $

w 3

r -_

. _ . . . .. -_ l-

. , _ - _ _ . _ __ _ _ , - _ . . _ _ _ ._ .m._ .. _m _ - . . . . . . . _ _

,8 - j ..

. .* ; e ~ N-

/ %*

-t 4 I f

\

s Fig.$-1n bonding, the cuplos;vo and pnmo mctol oro pinced toacther 1 ond spaced slichtly away frorn tho beckin0 motel When tho exploshe

- tokos ploco chcod of the co!

is;on, ccts to c! con the io nt z ,  ;

, cotes primo metal veloc.ty, V, col,ision vdoc;ty, and D detonct;on front.  ;

V D-Octs.iator

• _r i' t...-,. - .

g '

WEN $$$/2h.0697A74)?7//?) -

Pri c$st r m., m , m ,

1 i

\\ NN\hhhh%%%\\%%\%M- E,0cMn;ircla!

' == \\

wT : + ... ,, ,>

-]. 7[f.4. h 7 01- , ,  :. --Ea;!csive a

Vg Pliot MCtat 1'

'*g =~-

9 g

,9 ,_ i \\\ i Ongi.1 - '

y s\

• l O Ah h ict;\.,.\\/5cf et - - - - -

1---sacung

' I mcta!

r L-4

l. Ve ,

i .\I!'EAI, P; LOC;I7l,%

d' 4

s

?

e 4

e 4

e k-'

9 i

'r 3.

u g..m.veve.v

. n_ _ ., -r97> --

ji _

w 3)q ..