ML20237K142
| ML20237K142 | |
| Person / Time | |
|---|---|
| Site: | Calvert Cliffs  |
| Issue date: | 06/22/1987 |
| From: | Lippold W BALTIMORE GAS & ELECTRIC CO. |
| To: | Johnston W NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| References | |
| NUDOCS 8708270070 | |
| Download: ML20237K142 (6) | |
Text
-_--
o BALTI M O RE
'I GAS AND
\\
CHARLES CENTER.P. O. BOX 1475 BALTIMORE, MARYLAND 21203 W. JAMES LIPPOLD MANAGER NUCt. EAR ENGINEERING $ERVICES DEPARIMENT June 22,1987 U. S. Nuclear Regulatory Commission Region I 631 Park Avenue King of Prussia, PA 19406 ATTENTION:
Mr. W. V. Johnston, Acting Director Division of Reactor Safety
SUBJECT:
Calvert Cliffs Nuclear Power Plant Unit Nos.1 & 2, Docket Nos. 50-317 and 50-318 Failure Analysis Report Summary for Low Pressure Safety injection Relief Valve RV-439 Line Gentlemen:
During an NRC inspection the week of May 11, 1987, your. staff requested that we forward a summary of the subject Failure Analysis Report. This summary is provided as an attachment.
Should you have any additional questions, we will be pleased to discuss them with you.
Very truly yours, s
WJL/LSL/ dim Attachment l
1 i
8708270070 870622 dog }
PDR ADOCK 05000317 G,
PDR g
oss y<m
Mr. W. V. Johnston June 22,1937 j
Page 2 l
cc:
D. A. Brune, Esquire J. E.
Silberg, Esquire R. A.Capra, NRC S. A.McNeil,NRC i
W. T. Russell, NRC T. Foley/D. C. Trimble, NRC H. J. Kaplan, NRC 1
I' ATTACilMENT i
CALVERTCLIFFS UNIT 2
SUMMARY
OF MATERIALS AN ALYSIS OF TIIE LINE LE ADING TO RV-439 i
'A visual examination of the failed 1/2 inch. pipe leading. to RV-439 (Low Pressure l
Safety injection Relief Valve) revealed a crack on the 1/2 inch pipe that is at an
.j angle of. approximately 45 0 with respect to the fillet weld.
The 45 crack i
intersects the fillet. weld and fo!!ows, the toe of the weld for approximately 3/8 of an inch. The cracked-pipe was removed from the line leading to RV-439 for further examination (see - Figure No.1)
Microscopic examination of the. pipe spool fracture surface in the ?aboratory indi-cated two cracks ' initiated at the toe of. the fillet weld approximately 1/8 of an inch apart. These cracks grew in slightly.different planes creating the step (or ratchet mark) observed at in Figure No. 2. The crack to the right of the step in Figure No. 2. continued to propagate along the toe of the weld while the crack to the left veered off at a 450 angle to the toe of the weld. A schematic is shown in Figure No.3 that highlights the features discernible on the fracture surface. The radial lines extending. from points A and B indicate these were the sites of crack initia-tion.
There was no evidence of beach marks on the fracture surface of the crack that initiated at point B and only faint beach marks near the very end of the crack initiating from point A.
A scanning electron microscope examination of the fracture surface around the initiation sites did not reveal any latent defects that could be responsible for the initiation of these cracks. An interesting area found near initiation site A is shown in Figure No.
4.
This area has a faceted appearance and river patterns indicating crack propagation along preferred crystallographic planes.
This is typical of stage I fatigue often observed near the point of initiation. No discern-ible striations were observed in the through wall area near the initiation sites but striations were ~ observed from the middle to the end of the 45 crack (see Figure No. 5). If the spacing 'for the striations in this area is projected over the entire I'
surface the number of cycles involved in the propagation of the crack would be greater than 6000.
1
('
An axial cross section of the pipe containing the crack was prepared for metallurgical examination. This examination revealed that the through wall crack was transgranular. The microstructure was that of a typical 304 stainless steel.
The chemistries of the' weld and base metal were within the materials specifications.
Hardness of the weld and base metal was HRB 9) and 87, respectively, which are typical and acceptable values for these materials.
Location and mode of crack propagation indicate the failure was due to cyclic torsional / bending. loads. It has been estimated that greater than 6000 cycles were required to propagate the crack. Our visual inspection and assessment of the line during operation did not reveal any unusual vibration that would explain the fail-ure.
We believe it is possible that a transient high frequency vibration condition may exist that is responsible for the failure. The lack of distinct beach marks over. most of the fracture surface seems to indicate that the propagation of the crack' was not interrupted. This strongly suggests that a single transient vibration condition would be responsible for the failure.
l
[.
~~ W:7.;
MW 1
. y.
s...
..,x t c'>e.; n:n v net,
- .grJ.
i **'WN DW ?Bii. Myth,&5 1%V :. ins 41.:% rs'. VfW i.
\\
Figure No. 1 Section Containing Crack in as Removed Condition
. -r-m-I[
('
2 2 a f; 3 T % y-
-Q _, ' ' '
,, 0.. " * '
4 I
i j
l' I
i Figure No. 2 Fracture Surface of Cracked Pipe Showing a Ratchet Mark at Point 1
{
l
STRIATIONS l ji[ g { lll{}'
f c
l
- FILLET *' ELD p
W l
\\vl j
\\
\\
f.
i L5 V
,Q Q p ',A
/}W
,.\\
FAINT 3EACE. MARKS Y
'{
STEP Figure No. 3 A 5che=ati: cf the Fracture Surface F. highlighting D:,.scern ble Features f r== the Examination
~
' ' h ^
?/*
^ -
.;, s yf l
. ji lI,ll 3-
' f,.
.3, 14..
Q l
.~
l'
//*
,a w~
\\
l J!
e
'j s.
[2
.#' 4 !
l t
(,,'//
'(
bg i p
- N:'
<]
N:'
,: 4
\\
f i 4,
.'i..;..
/,,
jf e-
'g i,
3,. /p ' E) it,
(f
/, Qi r
' \\
r,,,j - (
4cs g.
s
/ll-f,$,,:
.b
.g
's',,,. [
'I h.
d
~
0.1 m'2 9.9 k V 372E2 01N 01.BGLE A
Figure No. 4 Faceted Area Near the :nitiation Site
"T W.
y,
,, e
',k s.
.,.,w s
. ;**M i
T
'Q~..,.
+
'. a l.
4 s.d
~
> >'~,.,..w!
.y.: -
T
/
f'
.1
_ k) L ',
r-
^
4
.s t.
l k
.7 c
'x j,
t 1 0 u m 2 9 9 6.i.i 131E? O009 01 BGtE m,
Figure No. 5 Striations observed on the Fracture Surface Near the End of the 45 Crack i
.