ML20063Q024

Supplemental Rept of Ultrasonic Indications in Top Head Weld VC-TH-1-2 at Ja Fitzpatrick Power Station

ML20063Q024
Person / Time
Site:	FitzPatrick
Issue date:	06/15/1990
From:	Chapman T, Clark J, Mortenson S GENERAL ELECTRIC CO.
To:
Shared Package
ML20063Q022	List: JPN-90-057, Forwards GE Supplemental Rept of Ultrasonic Indications in Top Head Weld VC-TH-1-2 at Ja Fitzpatrick Power Station ML20063Q024
References
B11-0497-DRFT, B11-497-DRFT, NUDOCS 9008170026
Download: ML20063Q024 (16)

Text

'>,1 -

- ., , 1

~ .

a ..

w 4*

o 4

g i i d

. f ,.'

+ 4..

..q.-. .._.--,s_ l. ,' _ -

g i

f '.

s o o.

t' i'

• sp ,

y

, ,;c' .

3 ,.

t 5

\1 1

1 i,  : 0:: ,

. +.

i.. '.

4 i

,, , 4 .

uo  ;- :

z .

gre k#

."  ?

aa o

.1 'I

(. i 1

T s

m mn .

.j 1

' '  :)

'e, -

s,

t. +L,f

. _ 5;

=.

1. s -' i\

'L I J 1

f j .

t .

4 < t >

i:' J, .

s s

--3, u-

-.j m -

1 t 1

A .

$,.l i

+

,. +='il

- 2: . s i

1

'l i ;i.  ;-  !

• * ^

i

.)

n .

a a

q

. ' i,7

, . 1.f } 'k s 1 l'i , iGENuctsar Energy' m

n.

9. . _ ,

1

, .w 4  ; .

. .;4 i

' i'.

1 4,\! '-4i

• wtr' . .

j 2 -

~

y * .

. ,i

o. :900817oO26900814

.PDRL x a' 4

j

? U-~

~

1 P' ADOCK 05000333- .

p v1 PDC- ..

'N a * .

, j., <- '

q W' "

i_

~

1 t

Transmittal 508 9003

-DRF f B11 0497--

i i

'I SUPPLEMENTAL REPORT'0F-

-ULTRASONIC INDICATIONS IN THE TOP MEAD WELD VC-TH 1-2

.AT JA FITZPATRICK POWER STATION June 15, 1990. j i

Ceneral Electric' I Nuclear Energy.

175 Curtner Avenue San Jose,' California 95125 4 i

i l

?

)

Prepared by: (- r - :-

S.f.Mortenson,UTLevelIII NUE Development-Reviewed by: e .

J/JV. Clark, Principal Engineer NDE Technology' Approved by: __-- -

T.L. Chdpman, Manager Reactor and Fuel Examination Technology

l f- -,

IMPORTANT NOTICE REGARDING CONTENTS OF THIS-REre?J PLEASE READ CAREFULLY The only undertakings of the. General Electris Company respecting information in.this document are contained in:the contract governing this work-  !

cnd . nothing contained ' in this document shall be construed as changing said-contract. The use of this information except as defined-by said~ contract, Lor.

for any purpose other than that.for which it is intended, is not authorized; and -

with respect: to any such unauthorized use, neither Ceneral Electric Company nor i eny of the contributors to this. document. makes any representation or warranty  !

(express or. implied) as to, the completeness, accuracy or usefulness Lof the i information contained in this document-or that such.use of such information may not infringe privately owned rights; nor do they'assuma-any responsibility for liability : . or damage . of any kind which may result from . such use of . such information, i d

I t

t a

i

'I I

4

W . . - - - . - - - . - __ . _ _ _. _ _

,+-

'l sl a

TABLE OF CONTENTS-

SUMMARY

1.0= INTRODUCTION 2.0 FITZPATRICK FIELD DATA 3.0 1ABORATORY DATA-3.1 AUTOMATED ULTRASONIC DATA 3.2 MANUAL ULTRASONIC DATA 4.0 DISCUSSION

5.0 CONCLUSION

S

6.0 REFERENCES

i SU1 DIARY During the 1990 inservice inspection at the James A. Fitzpatrick Nuclear .

Power Station, an ultrasonic indication was detected in the top head dollar I plate weld VC TH 1-2 examination volume using standard ASME Section XI manual techniques. In an attempt to characterize this indication, Refracted Longitudinal (RL) sizing techniques were used. What were interpreted to be tip diffracted signals from this flaw, were in fact signals from mid plate segregates located in the base material. This resulted in over estimating the size of the flaw. A considerable amount of additional effort was ' then required to j disposition this indication. It has been shown that the indication was, in fact,  !

I a veld fabrication flaw.

This' report provides a discussion of the RL. examinations performed at Fitzpatrick and of laborator" RL data obtained from a plate containing known l segregates. The data fros thc .s tntaminations, demonstrating the presence of mid-  !

plate segregates in the RPv , top head at Fitzpatrick, and the offeet of the segregates on the RL sitius efforts are presented in this report.

?

-I l-

i i

t 1

1

l a

1.0 INTRODUCTION

During the 1990 inservice inspection at the James A. Fitzpatrick Nuclear Power Station, an ultrasonic indication was detected in the top head dollar plate weld VC-TH-1-2 examination volume using standard ASME Section X1 manual techniques. Considerable effort was used in characterizing this indication.

This included a review of fabrication records and radiographs, taking of new radiographs, automated UT, visual, magnetic particle, liquid penetrant and manual refracted longitudinal-(RL) sizing techniques.

The indication was shown, by a combination of these techniques to be due to a fabrication flaw. However, the data from the RL sizing technique was misleading because what was interpreted to be tip diffraction signals from the flaw was actually mid plate segregates in the base material.

Mid plate segregations are concentrations of alloying elements in specific regions, that form differer.ces in metallurgical structure due to variations in chemical composition. They are formed during solidification of the original ingot. These concentrations of alloying elements form bonded interfaces with different acoustic impedances that, from an ultrasonic point of view, are similar to the interface in a dissimilar metal weld. When an ultrasonic beam interrogates this interface, portions of the beam are reflected due to the acoustic impedance mismatch at the interface. Similar ultrasonic responses are also caused by small inclusions, such as Manganese Sulphides (MnS). These inclusions are found parallel to the plate surfaces and are often called stringers. Segregates and stringer inclusions differ from laminations in that laminations are larger areas consisting of unbonded material and are'normally contained in one plane near the center of the plate thickness. Segregates on the other hand are usually scattered through out the middle one third of the plate thickness.

This report is not meant to provide the evaluation of the indication in the Fitzpatrick top head which is contained in reference 1. Instead the purpose of this report is to supply supplemental data on the findings related to plate segregates.

2.0 FITZPATRICK FIELD DATA Af ter detection of a high amplitude indication using 45' and 60' shear wave transducers, the indication was further characterized with 60' and 70* RL's from the inside surface of the top head and a 45' RL from the outside surface. The sizing signals were not clear, distinct, tip diffracted signals and at times the screen contained additional signals. The indication, when finally plotted with a 45' RL, showed a flaw configuration that was not consistent with any postulated in service flaw. The base material was scanned with a O' manual examination at 5 MHz at 30 dB above reference and revealed mid-plate indications typical of segregates.

The base material on each side of top head weld VC-TH-1-2 was scanned with the General Electric's Reactor Inspection System (GERIS), an automated ultrasonic system. This base material was scanned for a distance of approximately 1-T from 1

. , 1 the edge of the veld with a.1" diameter, 2.25 MHz transducer. The calibration ssnsitivity was determined according to ASME Section V, Article 4 T 441.6 which utilized the side drilled holet in Fitzpatrick's 5" thick ASME Code calibration block PT No 1. This is the procedure used for the detection of planar reflectors. The recording sensitivity of the data was 4 dB above reference gain, j GERIS recorded base material, mid plate, O' indications sporadically over ,

the entire area of the scanned area. Figure 1 is a top view (C-scan) of the data l

recorded with GERIS. The CERIS data does not show the reflectors to be adjacent j to the weld where the RL sizing was performed, however,. as mentioned before, a i supplemental high sensitivity manual O' examination did. The area where the sizing was performed was from 130" to 155" on Figure 1. Figure 2 illustrates a cross sectional view of all the O' GERIS data recorded along 80" of length of I top head weld VC TH 12. The base material to the right side of the weld is the i dollar plate, and the torus plate is on the left side. The reflectors vary in depth from 0.85 inches to 3,45 inches from the outside surface of the top head.

3.0 LABORATORY DATA To confirm the effects of segregates on RL flaw sizing techniques, manual ultrasonic tests were conducted se the GE Nuclear Energy's NDE Development Lab in San Jose, California. Contained in this section is data from previously ,

performed GERIS ultrasonic tests on segregates. '

3.1 AUTOMATED ULTRASONIC DATA Figure 3 illustrates the cross-sectional view of General Electric's Cracked Implant Block containing GERIS recorded ultrasonic reflectors that have been characterized as mid plate segregates. These reflectors were similar to reflectors from an operating _ plant that have been characterized as mid plate segregation through metallography. The recording sensitivity of this data was the typical ASME calibration sensitivity. The Cracked Implant Block is a 26" long section of 6.44" thick vertical shell weld from a cancelled RPV. Since this-plate is from a real, as fabricated pressure vessel, and since segregates have been detected at several operating plants, it is believed that this condition is typical of many other plates in pressure vessels in the field. ,

The proximity of the individual reflectors and the distribution of the reflectors of the indications within the volume of material are consistent with plate segregates or small inclusions. To illustrate this fact, Figure 4 is a cross sectional view of one scan line of data recorded at 20 dB (10 times) above calibration sensitivity. Note that areas exist that consist of densely packed reflectors of varying depths. The ultrasonic data plots of these segregates appear larger than their actual size due to transducer size and ultrasonic beam divergence. Although these segregates do reflect sound due to their metallurgical properties, they also allow some sound to pass through.

2

F1 ht: TCaett.det- -- 230* ';i A -

_ [ch # 1]

GE  :

- "^^ 0*

[ch # 2] }; -

r 5- Figure 1

[ch 8 3] g' Top view of CERIS IJT Data i- - 210, in veld VC-TH-1-2.

p -

L

- 200~

e-

[ omit] [ie sat] [x-sec]

iT')[ se s ][come z ] - 190' w

- 180' Data Ch # 1 No Ind: O e t

No Ref1: 1158 U me sytes: ES5ee P - 170' Time: Jun ~ 5 11:99:99 1 M

- 160' COLOR AMP (% DAD = - 150' S

E 100 - ,

r B 90 .99 E ~80 - 89 ..- E - 140' ES .70 - 79 E, . Veld - VC-TH-1-2 Top Head Veld

~

60 - 69 D 50 - 59 .

O Degree Reflectors

- 130'. Ex6n perforned 4dB over Reference-.

E 40 - 49 E .30 - 39 E 2d - 29 3 elect' objects: Base point or displacement: Second point of displacement:

Command: regen Reoeneratino drawing.

_-_ -__=

. . _ . . _ , _ . . _ - ~ _ _ _ _ _ _ _ _ . - - . _ - . _ . . - . - . - -___

File: TCaett.eet (Ch # 1]

GE

[Ch # 2] _

FWe 2

[Ch # 3] Cross-sectional view of GERIS UT Data in Veld VC-TH-1-2.

[ Quit] [10 58E] [x-sec]

Lies % ][ 6e % ][ Code Z ]

v Ch # 1 S No Ind: 0 t U _ .- __ ,:-

No Ref1: 1158 p - ~~

No Bytes: 63590 -d Time: Jun 4 19:18:52 1994 IM"CD T4?L c -

~

-E=-:_-6~

_ _ - - ~ -__ ,

. ,M __ ~ ~ ;= := GIf_D._-

\

\

COLOR AMP (%DAC) _

E 100 -

B 90 - 99 \

E 80 - 89 70 - 79 60 - 69

! )lIlll 4

~:~ -

w a

t' a .

D.

T

. U .

Sc k g_

I o Rl EB

.C t _

f n 3 oal ._,

e wp rtiIem a

g iv d F le ak nc oa 6 _

ir tC c

en _

si s

s o __

r _

C 3

~ _.

+

'T

~

d '-

-  : e. _

sa .

tr s

ct t eo jr b'

E o '- i _.

$ tn cs eau a

k'. . _ '

l a

=

j I

] s U Mi C ._

E G

c e

s s

, A

)

I w9999999 87 654 32 M

[

c

[

o 5 4:  %( - - -

7 o0 0

]

E j

s s: 4 P

0 0 m s 7 6 m 40 k'2 3

0 -

t B s7 e . S e 1 e0 M ~

d. s 7 3 A -

o 1 g 2 1 3 4 4 m

u [

l

[

s t j  : y -

i c

]

s a 1

s eM v

] t y :

R t # t e ~ O e S e h " EEEE LE E E s~

i -

h m 1 m i 1 c 9 oi m_

F { [ [ N T O

-_ C h

File: cibl.dat

[cn # 1] -

GE J .

Figure 4 Cross-sectional view of one scan of GERIS UT data from'tne Cracked Implant Block.

[$ sit] [le SSE] [M-sec]

.I soo 5 ][ se Z ][ cede s ]

• *8i Se --

~~

8

~

ty .. . ~ , , , .~~ . . . .

setI: 19898 p, _ _

_~..

me nytes: 114 esse Time: Jun 4 16:12:43 199eE

~

COLOR AMP (%I)AC)

E

~

100 -

B 90 - 99 E 80 - 89 E3 70 --79

60 -- 69 E 50 -'59 E 40 - 49

'E 30 - 39 E 20 - 29 1 1/ Cent er /Dynami c / Extents /Lef t / Prev i ous /W i ndow /< Sca l e( X) > :

egenerating drawino.-

~ . - . .-. -- . - . - . . - - - - _ . - _ - _ . _--.-

^

l l

l

\

i There is no evidence that these reflectors present any significant interference  ;

to ASME Code required angle beam examinations, In-fact, it nas been verified- [

that small underclad notches in this' plate.can be detected with a shear. wave  !

cngle beam passing through these segregates. I 1

3.2 MANUAL ULTRASONIC DATA Manual UT data using a 5 MHz, 1" diameter,. straight beam transducer, was  ?

taken at 3 locations' ranging from .3 inches to 1.0", from the right side of!the weld centerline. _ Figures 5 thru 7 are the digitized A scans taken at these i locations. Mid-plate segregates that were detected ranged in depth from 1.5"  !

to 3". The sensitivity for this test was 20 dB over a backwall signal' set at ,

80% FSH where no mid plate segregates were present.-- The amplitude of these 1 segregates in this- test was less than one half of the amplitude of the backwall-and in nc case observed resulted ~ in a complete loss ; of back reflection. A l completo loss of back reflection is a part of the Code basis for' rejection _of-such indications. , q l

i i

I

~"

h

• backwall l

• backwall- '

segregates segregates 4 4-bdl 1 e u M L m. . Rh, ,s bk. L ,

Figure 5 Figure 6 0* A scan .3"'from Ref. O' A-scan. 8" from Ref.

i

'7 I' ]

'l I

i I

I l

l

• backwall segregates

.4 l

Iulh b.i _ E kd !- uuA. .

Figure 7 0*=A scan 1.0" from Ref, i

d 1

The ultrasonic-instrument was then calibrated with a dual element 2 MHz  !

45' RL transducer. The sweep was set at 5" in depth using the side-drilled holes in a 4" ASME Code calibration block. The sensitivity was set with the 2% .

of T notch at 80%-FSH (full screen height) as shown in: Figure 8. l l

' Notch

• l Yb EYEhAA) Mh, Figure 8 45' RL A-scan of-Calibration Block Notch 8

i

The area, where the O' manual ultrasonic data (Figures 5 thru 7) was taken, was scanned with the 45' RL.. At reference sensitivity a segregate was detected .i with the transducer at 1,6" from the weld centerline. The sweep position on the i CRT screen showed_4" of depth as shown in Figure 9 With a 45'-RL transducer  ;

o 23' shear wave component exists. The segregate detected was detected by the shear wave component. Since the velocity of longitudinal waves. is 1.8 times greater than that of shear waves, and the refracted angle is twice that cf %e '

2 hear wave, the reflector appears to be 4" deep instead of the actual cepth of 3". Figure 10 is a cross-sectional plot of the segregates showing longitudinal j cnd shear wave components, i i

i i

l IRAN $ha"(R

(

segregate

• t<An vavt c

l LDCITuttwL vAvg -

5thtGatts L w L ) &nk_k %) & h.k. .

Figure 9 Figure 10 45' RL 1.6" from Ref. Plot of Figure 9 A-scan using Multiple Beam Components, I

4 i

9

e .. ..

The transducer was moved forward 0.5" to a distance of 1.1" from the weld i centerline. The shear wave component then detected a segregate at a different l depth as shown on the sweep of the A scan in Figure 11. The depth of this segregate was- at 2.4" . Therefore, the assumed depth of this reflector using the -

45' RL component was 3" as illustrated in the _ cross-sectional plot in Figure 12.

i 1RAN$DUC(R 5

segregate 4

wenum% vavt ,,

syan vave

}

s

. dLNJdVF O Figure 11 Figure 12 Plot rx Figure 11 using 45' RL 1.1" from Ref. Multi 1 1e Beam Components 4.0 DISCUSSION l

The GERIS data from the Fitzpatrick top head base material of weld VC .

TH 1-2 was characteristic of the segregates in the base material of General Electric's Cracked Implant Block. A review of the vessels' fabrication ultrasonic test (reference 2 ) reported mid plate spot indications with no length or width, which is typical of segregates.

The indication at Fitzpatrick was parallel to the weld and believed to be at, or near the base metal to weld metal interface, on the dollar plate side of the weld. To size such an indication, using RL techniques, the sound beam would be directed into and perpendicular to the weld from the base metal side. It becomes readily apparent that the lower angle shear wave component from the RL trannducer could reflect from segregates located almost underneath the transducer in the base metal. The higher angle RL component is much less sensitive to segregates which are oriented for the most part, parallel to the surface.

Because the shear wave travels at a lower velocity than the longitudinal wave, 10

o- ,, ,

the indication would appear to be iwated at or near the weld metal, when plotted-et the_ angle. and the velocity of the longitudinal wave. This explains what happened at Fitzpatrick since it has been duplicated on a plate which is known to contain segregates along the side.of the weld.

5.0 CONCLUSION

S The indication in Top Head weld VC TH 1 2, as characterized by RL sizing-.-

techniques, was of a nature' unrealistic of crack propagation from any expected surface or subsurface type flaw. The' larger indication, characterized by RL sizing techniques, could not be confirmed by either 45' and 60' shear wave (manual or automated CERIS) or by radiography.

Mid plate. indications with characteristics typical. of - segregates were detected at higher-than normal sensitivities with both manual and CERIS systems.

These mid plate -indications were -located in <the base material near.. the indication. These would affer,t. the RL ' sizing data, which requires' high sensitivity to detect tip diffracted signals. The RL sizing data taken on this indication, that was affected by segregates, was mis leading.

The results of all examinations and tests have shown that the RL ultrasonic data affected by segregates, is=not related to the fabrication flaw which was detected and sized (rer'erence 1) for disposition by analysis (reference 3).

~

The. tip diffraction method, using RL transducers, is one of the mos_t accurate methods for determining the depth of a flaw. 'However, caution is -

recommended when RL transducers are used on rolled plates'. A complete examination for segregates, along with accurate data plotting is required..

11 i

s o,,, .

6.0 REFERENCES

i

-1. Ultrasonic Examination. Data Package for Wold VC TH-1 2, copy j contained in CE'DRF B11 00497. j

2. Report of Ultrasonic Testing, Combustion Engineering, Inc. , Contract  !

1. 21566 dtd 4 10 1971, copy contained in CE DRF B11 00497.-

3. Structural Evaluation'of Indications in the Reactor Top Head at the~.

JA Fitzpatrick Power Station, Report SASRf 90-43, dtd May 1990, j

..)

-l i

e I

)

\

i I

12

- t:

I ae .. r 4

a

-i

\

Jl  ;

i l

s 4

I I.

o,n .c:r , .

(

...u- .1

, 3/ r. [..

h. . ),, . ,)

y *, ,

. .. 2 a . ,,

A;;.

s .,,

l-- .,

,wk' n; ;. 7 N

.

t

. . -

• O,. . ' 2.,4. , ..

'.,- % ,*yy?,

,l',,.

,- ... . r  !

. '7, . , ' . ,' 4* -

v.s..

g.A .* *. . .,* , ,-

j GE Nuclear Energy i 175 Curtner Avenue i

San Jose, CA 95/25

. ... .. . . . . . . ... . . . . . . . . .