ML20211D421

Preliminary Steam Bypass Line Structural Evaluation

ML20211D421
Person / Time
Site:	Fermi
Issue date:	10/03/1986
From:	HOPPER & ASSOCIATES, ENGINEERS
To:
Shared Package
ML20211D357	List: ML20211D351 ML20211D360 ML20211D396 ML20211D421
References
HA-10-86-532, NUDOCS 8610220191
Download: ML20211D421 (7)

Text

. ..

Attachment

~ VP-86-0146-Page 17 of 23 HOPPER AND ASSOCIATES HA-10/86-532 ENGINEERS PRELIMINARY STEAM BYPASS LINE STRUCTURAL EVALUATION Prepared for: Detroit Edison Company Enrico Fermi 2 Job Site 6400 North Dixie Highway Newport, MI 48166 Prepared by: Hopper and Associates

• Suite E , 210 Avenue I I Redondo Beach, CA 90277 October 3, 1986 8610220191 861016 1 PDR ADOCK 0500 P

I r

- - , , . , , - - . - , , - - , , , , , - - . _ - , , , - - - , - - - - - - . , -~-..n-,--- -. . - - . - _ - . - - - - , - - - , ,,-~.r-.- - - - - - , , - . . . , - . . , . - -

.. k'ttachment VPs86-0146 Page'18 of 23 HOPPER AND ASSOCIATES HA-10/86-532 l

~

ENGINEERS l

A catastrophic respnnse to flow generated noise was observed in i the Fermi facility main ~ steam relief _ valve dumplines during operation in 1985. As a result, this 30" diameter piping was redesigned to en- ,

hence its operational response characteristics. The material thick- 1 ness was increased from 3/8" t o .1 " , and support lugs were removed from  !

the piping to reduce wall stresses and stress concentrations. In ad-dition, flow orifice plates were installed downstream to mitigate the noise levels at high mass flow rates. The newly configured piping was thoroughly instrumented to provide immediate feedback on the opera-tional response of the system. .

I as predicted, the I Evaluation of the resulting data shows that, most severe structural response in the form of wall flutter occurred in the west line just downotream from the valve. Significant random vibration was observed at mass flow rates between 30 to 40 percent of the design limit. A fatigue life evaluation using a correlated frac-ture mechanics approach indicates that the system is not susceptible to structural degradation even in the most critical mode of operation.

The wall flutter observed in the dumplines was noted to have ran-dom vibration characteristics. The fatigue life evaluation was based upon a' detailed fracture mechanics analysis in which a maximal permis-cible weld flaw was assumed at the location of critical stress.

Crack growth was evaluated in accordance with the relationship presented in Exhibit A.

As anticipated, data from the instrumentation revealed critical atrains in the west line just downstream from the valve. Exhibit B indicates maximal half amplitude (0-P) microstrain recorded at this location as a function of mass flow rate. For the postulated weld flaws to grow, the strain . (0-P) must exceed a 100 microstrain threshold level. These strains were noted only at 30 to 40 percent design mass flow rate levels. At other mass flow rates, crack growth is improbable.

Statistical analysis of maximal and minimal peak strain values measured from expanded scale strain traces yielded a normal distribu-tion for number of occurrences as a function of peak strains. The i; number of occurrences for alternating stresses is determined by the difference (maximal minus minimal distributions) as a Rayleigh dis-tribution. The Rayleigh distribution of occurrences for' alternating i.

atress ranges at 36% mass flow rate determined for a one-second time interval is shown in Exhibit C. A minimal occurrence of detrimental stresses (i.e. A S>5. 85 KSI) is observed from the Rayleigh distribu-tion. These stress-occurrence value sets are then used to predict fatigue crack growth.

• g:

Attachment VF-86-014&

Page 19 of 23 HOPPER AND AssocirrES HA-10/86-532 ENGINEERS The life of the west dumpline in the most critical operating mode at 36% mass flow is estimated to be in excess of 130 days as shown in Exhibit D. This represents the total accumulated time in the critical mode, therefore, brief transitory periods through~the critical operat-ing mode range are not detrimental to the overall life of the pipe (this phenomenon is akin to the transition of a turbine through its critical speed during synchronization). Furthermore, radiography of the welds in .the critical regions has indicated that flaws with the postulated size are not present. This means that the critical fatigue life prediction is extremely conservative and that structural degrada-tion is improbable even in the worst-case operating situation.

However, avoiding operation of the west dumpline between the 30 and 40 percent flow rates can do no harm.

Larger peak stress intensity ranges are probable at other loca-tions in the line as a result of three-dimensional stress combination end stress concentration effects. Such stresses do not appear to be detrimental since significant material flaws are not observed in the parent material.

l I

i .

0 0

X .

CRACK GROWTH RATE lo'g h  :

2 0 W t e5 4 m, AK

_ _ _ _ _ _thr_esho_ld = 3.1 ksi 6 7

w N

> a Z.

O Z ._,

Ei o n O

3 EM 'y%

g y E " g> ,

- oZ ,

% 9 EU

> d ..

E

>i Q E 4 s ._ _ _ _ _ _ _ _ _ _ _ _ _ _ 1 C

" ^K >

@ critical h'

lM _.

e W

% > Q

• E M

g V O

3 .

M

z V 4T9Tyxa E2 Je 02 a6ed EES-98/0I-VH -9tLO-98-dA

'4uaWygeig -

HOPPER AND ASSOCIATES 2 7 ENGINEERS . $ .b#R-

! Nei

'l e., % @

-ee N

t. ca L -

I 100-  :

)

!  ?

e j z e 0

. e I G i 2

) e x J 50-l

. Recorded Data

l. o On Line Observed Data 1

1

, ,. . . M2 20 4,0 60 80 100 g.T M (%)'per line pg

! P-N j st to m

CD I j . . ALTERNATING STRAIN VS. MASS FLOW RATE N 1

I

Y.$nr.a$e -

o i

@$m..o Ae

' mt HoNtmIvWM o rt tPrHtPt O 2 $ m~ e% Nw

=

S E

S S

E R

T S

.0 G 5

N

)

I I

S T K

1 A 7 ,0 5 4 N

= R S X ) E E A I T M S K T A S (

I C (

A L

O ,0 S S 3 2 A

SS R AE E 6

=

DNGI 3 S A

F N N O AE R .0 E N P 2 P O O

H I T

J L

0 1

B I

R T

S I

0 0 0 0 0 . D w 0 0 0 3 2 0 1

H w az8wm *E 0o*$8oo mo xgro' G I

• E .

L Y

A e R

1il

1 i11 1 ,:!i1l ' iI

HOPPER AND ASSOCIATES o ,-

ENGINEERS ' to cn su on N

100 --%+ + w-o' m 7 a.

il l _ ._ . .

_.__ _ .... l..,_ _._..~.. .4 .m. . _ . , l e A :3

% e c.

- q rin :ta D' i- , t.0q n n: -

01: .- t yv j;- ,  ;

._-. .r

2.. m aj t:5 t.. ; 4 .0.: .

3. . . ,._, .

. . . . .. 4 .. . .4

'cg;..

as yj 4. ., ..; g

_ _ _ . . 1 1 . . . . .. ._ . . .t . .

[. 8'. -

E 6 L' 'J - 4 i -

N!

!!h '*-

I i .'=f ic t' r I t -

E -

@  :.4 ! ! - .

10

EN I I ##E ' '

~

E i  ! ! I I- EM 5

% %__4

..f _. . . . . ...

+ s -

. . , c pi + , in, i i o % - , i + g ..

.m un ,- ,

f -.,

.2._ u_ . . a. ag  ; a 4 gg .g g . . u_ _. .: u. . . :. .

4 .g ut -

g.g

. :i t' r?

• - i i th ei- ,,1 i.

as J.-

r 1 ii.'!!iz iiq!E

!!i = ' 2

!/. 1 .ii g

Ei.ii .- -

. i

_ i iii::

I '

a Elf .l,l g L f 8 T h h1 Ekl f - .

!lf. d.'

{

L3

, r- 25.

.j .  ;

{

. g.

,, ... . .

i .. . .

MG JA $1 7 e m -

6 -

m- fi a -; g i ; [.

o a = == . cht  ;- _ m  :. i! _

A. .:.. .::. y. -

= ..

u::. i t F--

=

w ..,._,

g .

.-. .6.-..-. ../

CX" >

4 --/ n l

!!': i W

p. y.

.-= . .

=.. 3

@ r o.01

/. -

f i 4.. -

9... , . .I..

^f il- j

/

f- t j- a- i-

/ - . ..- i-- : ~;-- -9 -- -

," . i i. i

% f.

. . .#~

~

~ ~ l."

0.001

, 4- . - . .

/ r i /

i fy . . . . , , up ; >

J::n .

4 5 6 10 100 1000 10 10 10 M lI:

TIME (HINUTES) s.y H< H

. tf o C1" WALL, FULL NOISE > Pg) cn t2 :

CALCULATED FATIGUE CRACK GROWTH m

. . W E4 VERSUS TIME AT 3 C> % MASS FLOW RATE

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _