ML19269C868
| ML19269C868 | |
| Person / Time | |
|---|---|
| Site: | Brunswick  |
| Issue date: | 02/05/1979 |
| From: | Gieske J SANDIA NATIONAL LABORATORIES |
| To: | Stello V Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 7902200004 | |
| Download: ML19269C868 (4) | |
Text
.
Sandia Laboratories Albuqum que. Ne.v hco 8 7115 February 5, 1979 Mr. V. Stello, Director for Operating Reactors Division of Operating Reactors
$O~ N USNRC Washington, D. C.
20555
Dear Mr. Stello:
Subject:
Brunswick I Ultrasonic Examinations, Safe-Ends September 1978, November 1978 and January 1979 A review of the ultrasonic data of all safe-ends obtained fram the September 1978, November 1978 and January 1979 ISI examinations has been made.
Of particular concern was possible ultrasonic detection of intergranular stress corrosion cracking at the safe-end to thermal sleeve weld. The UT scans as they were being performed in January 1979 were witnessed so that a complete understanding of the recorded data could be obtained.
Special attention was given to the examination of safe-end N2D where the maximum indications were recorded in September 1978.
After normal scans of N2D were accomplished, supplementary scans were requested and performed on N2D to further characterize the indications recorded.
Each safe-end was scanned with a shear wave angle beam at 16 db above reference sensitivity where reference sensitivity was set to 100% DAC using a 10% through wall calibration notch. Amplitude and metal-path data were recorded on a strip chart for all scans. A typical axial scan is shown in Figure 1 where a 45 shear beam scan of Zone 2 is depicted.
Full coverage of the safe-end was obtained by axial and circumferential scans of each zone shown in Figure 1 by moving the search unit back and forth over the zone and proceeding in a clockwise fashion around the safe-end. The search unit was also rotated through approximately i45*
as the scans were made. Four scans of each zone were made with the beam directed in opposite axial and circumferential directions. An additional zone 5 not shown in Figure 1 at the dutchman to safe-end weld was also scanned.
In general most indications were just above the noise level at the 16 db scan sensitivity.
Certain scans yielded spot indications on all of the safe-ends which were above the noise level and those were the scan of Zone 2 as shown in Figure 1 and the scan of Zone 4 and Zone 5 of the safe-end to dutchman weld.
Every safe-end displayed characteristically the same indications for these zones.
From the recorded metal-path data, the reflectors were located at the crevice ID as shown in Figure 1 and at the ID of the dutchman to safe-end vreld.
\\
t(
t i
7 9 0 2 20 0 0dY9
Mr. V. Stello February 5,1979 The amplitudes of several indications in Zone 2 for safe-end N2D were the largest recorded of all the safe-ends.
Figure 1 shows a plot of the positions of many of the indications recorded whose amplitudes were above the noise level at the 16 db scan sensitivity for several of the examinations of N2D. The indications were plotted fr the metal-path data assuming that all reflectors were at the safe-end to thermal sleeve weld. As a result all indications plotted near the crevice at the thermal sleeve weld to within the experbnental accuracy of 0.1 inch.
The area of maximum signals at 10:00 to 11:00 o' clock are located at the crevice. As most probable from the geometry of the scan, the indications were also assumed to be caused by the leading edge of the ultrasonic beam which was taken to be 50.
From the calibration data at 16 db sensitivity, the beam covers the angles from 35 to 55*.
Figure 1 also shows data plotted from separate 36* scans of the same area where a 45 leading edge beam was assumed.
A reasonable correlation of amplitude data was obtained by comparing September 1978, November 1978 and January 1979 data for the 45 scans.
There is also a good correlation of the mnplitude data for the 36*
scans perfonned in September 1978 and January 1979 Positions where the maximum amplitudes of the 45 and 36 scans occur also correlate reasonably well.
Two characteristics of the ultrasonic recorded data were looked for at the high gain sensitivity of 16 db above reference.
Evidence of a short metal-path indication coming fran the tip of a possible crack and e'ridence of steps in metal-path data due to reflections from the base of a possible crack and again delayed reflections fram the ID surface to the tip of a possible crack and back to the search unit.
No evidence of either of these possibilities was observed in the recorded data.
From a review of the examinations the following conclusions are made:
A good coverage of the safe-end welds by different angles and directions has been made; Correlation of amplitude and metal-path data between September 1978 and January 1979 shows no apparent change of reflectors; Data from each safe-end has the same general ultrasonic characteristics of low amplitude indications at the crevice weld and at the dutchman to safe-end weld.
Therefore the most likely cause of the indications at the crevice is the veld to base metal interface of the first bead weld at the thermal sleeve to safe-end weld.
Mr. V. Stello February 5, 1979 By the way the data is recorded and examinations conducted, it is highly unlikely that large-scale cracking could go undetected.
However, very small localized areas of cracking may not be evaluated as such due to ambiguous signals of very low amplitude.
The high gain scanning level is considered impcrtant for maximum infomation to evaluate the condition of the safe-ends but occasionally couplant noise interferes with possible signals of interest. The couplant noise problem could be improved somewhat by using a time corrected gain control on the ultrasonic instrument.
Sincerely, eb,, [ ' N 0J
' John H. Gieske NDT Technology Division 1552 JHG:1552:mek Copy to:
V. Noonan, Engineering Branch, Division of Operating Reactors USNRC, Washington, D. C.
20555 W. Hazelton, Engineering Branch, Division of Operating Reactors USNRC, Washington, D. C.
20555 R. Johnson, Engineering Branch, Division of Operating Reactors USNRC, Washington, D. C.
20555 J. Smith, Metals and Ceramics Division, ORNL P. O. Box X, Oak Ridge, Tennessee 37830 1552 J. H. Gieske
r e
kl
%2 Fc.ic h
?~.7 4'
A l
y 'c -
N s',
U'+h %2 )
v l.l2f" N
~
\\
x SRFE END Q
,'s-j)<.~,of Thr o t.t TagemnL SLEEVE 5
0.'V fif:
Y Tiq {=n)l;L ELEE'IE
+
N
'//E L D
+ St18 45 %
f g +. O.l" R rw r.., L 1c tcwt q das 'll 05" Scw 0 i'm h j,, f,,.. r
,y, (,G,,,,.,,
h Seft 78 3{ Scan S n '" k,
syt,78 78% Dnc,Jw 71 to %, Dnc.
o jan /'l
,4
<a co.
ozio(y pcc i r 10 r,) ARovND.shFE E ND
\\
no t,cv 9:pc IW n,,,
OD l
~
112 o Si1FE ENp WALL ]IllcnNE55 0
+
(P,Evic E,
?
c e^,&y
+ f,,
5 6 5 26 3
p 7,
ll/,RERMI) L SLEEVE WElL- ^
^ '
/
/
i'
/
I j
t o.l h (cosede Ecua c5 I
/
A Mda Pa!I'GICNahNn R
<e 1 %zL WJI +L;4xz # AeI s!ceve wek g
UT iaAcah'ows for /;/ r u n.,.
ofN'aD