ML20198F587
| ML20198F587 | |
| Person / Time | |
|---|---|
| Site: | Columbia  |
| Issue date: | 03/20/1984 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20197G328 | List: |
| References | |
| CON-WNP-0745, CON-WNP-745 NUDOCS 8605290013 | |
| Download: ML20198F587 (5) | |
Text
.
)
ATTACHMENT SAFETY EVALUATION REPORT WASHINGTON PUBLIC POWER SUPPLY SYSTEM NUCLEAR PROJECT NO. 2 DOCKET NO. 50-397 MAGNETIC PARTICLE LINEAR INDICATIONS ON CONTAINMENT WELD PADS MATERIALS APPLICATION SECTION MATERIALS ENGINEERING BRANCH BACKGROUND Purpose of Weld Pads There are 19 containment shell weld pads at the 512' and 524' elevations which are steel plates, two inches in thickness, round or rectangular in shape, and approximately five feet in diameter or length of side.
Beam seats are welded to the pads, and the seats support the radial beams between the Mark II boiling water reactor containment shell and the drywell.
Each pad attaches to the containment shell with a 3/8". fillet weld for the entire periphery of the pad.
This is the minimum allowed fillet weld size for joints with material two inches in thickness as required by the American Society for Mechanical Engineers Boiler and Pressure Vessel Code,Section III, Subsection NE, " Class MC Components."
T.he peripheral welds are adequate by themselves to take all loads.
There are redundant plug (circular fillet) welds within the pads that also attach the pads to the containment shell which are also capable by themselves of supporting all the loads.
The plug welds are a little over two inches in diameter and range in size from 1/2 to 5/8 inch.
8605290013 840310 DR ADOCK 05000397 PDR
i
, History of Events The Quality Verification Program required reinspection of a sample of attachment welding to the containment vessel under Contract 213.
Rein-spection by magnetic particle testing (MT) revealed linear indications beside the thick weld pads' peripheral fillet welds to the steel containment shell.
The reinspections were expanded to all 19 two inch thick pad attachments and all had linear indications.
The indications have an average length of 3.3 inches on the pad side and grinding showed average depths of approximately 1/4 inch.
Metallurgical investigations determined the indications in the weld pads and shell to be hydrogen underbead cracks.
In the weld pads, some of the underbead cracks ended as lamellar tears.
For the linear indications in the containment shell plating, metallurgical examination and ultrasonic testing (UT) showed no evidence of lamellar tearing, and the shell MT indications' lengths averaged about half of thosa in the weld pads.
All peripheral fillet welds around the weld pads at the 512' and 524' elevations had their magnetic particle indications removed by grinding.
These fillet welds were weld repaired and weld built up to 1/2 inch, exceeding their original size of 3/8 inch, and the final welds met MT inspection criteria.
l The 19 weld pads also have 405 plug or circular fillet welds which also attach the weld pad to the containment shell.
A statistically significant (55) of these welds were inspected by liquid penetrant testing and 13 had linear indications.
The indications were removed by grinding to sound metal demonstrated by liquid penetrant inspection, and the excava-tions were allowed to remain as is.
l l
. These welds were not restored to their design configuration because:
design did not require them as they were redundant to the peripheral o
fillet welds, repair of these welds in these small holes would be very difficult.
o fracture mechanics analysis determined that the flaws were innocuous.
o The inspections were further expanded to include ten other field welded pads of smaller size in the drywell and a sampling of all other attachment welds to the containment shell.-
A few magnetic pafticle linear indications and one UT indication where (MT was not practical) were found.
The total length of the indications is less than'E4 of the lengths of welds reinspected.
This ratio is normal for reinspections of structural steel weldments.
The indications were short in length, and, where depths were determined, very shallow (maximum of 1/8 inch).
Most MT linear indications were removed by grinding and weld repaired.
Fracture mechanics analyses were performed based upon the largest size of the flaws found which far exceeded the size of linear indications not repaired and the analysis demonstrated that the flaws remaining would not propagate.
, DISCUSSION / EVALUATION The extent and severity of the linear indications found on the original 19 thick weld pad attachments were not repeated in the expanded surface and UT examinations.
This indicates that the original causative factors of cracking in the two inch thick pad weldments were unique, a local problem, and not generic to other welded steel structures.
The expanded reinspection sample size appears adequate.
The small per-centage of linear indications length found vs. total length inspected is normal for reinspections of steel weldments which had previously been inspected.
The AC yoke magnetic particle inspection method is an acceptable surface inspection +.echnique.
There are other magnetic particle inspection techniques which are more effective for detecting defects which do not break the surface, such as underbead cracking and lamellar tearing, the crack causing mechanisms in this case.
- However, the configuration of the assembly, and the sequence of fabrication and loading caused the underbead cracks to break to the surface or caused lamellar tears to initiate at the surface.
Accordingly, the AC yoke method is viewed by the staff as acceptable for this application.
At the present time, we do not have a demonstration that would convincingly justify our establishing particular surface examination techniques for structural steel constructions.
The fracture prevention approach by the applicant is a standard approach, often used and accepted by the staff for many other applications.
a
. CONCLUSION Based upon the data presented, we recommend acceptance of the applicant's actions.
4
(
l L
~..-., p
~3
$E WQ R$${h.-
$Lk,*,,
DISTRIBUTION:
DOCKET FILES RAB R/F Docket No: 50-397 tgAR 2 8 BB4 FJCongel/ PLANT FILE MEMORANDUM FOR:
Albert Schwencer, Chief Licensing Branch No. 2 DL FROM:
Frank J. Congel, Chief Radiological Assessment Branch, DSI
SUBJECT:
tiOTICE OF DEVIATION, UNSHIELDED SAMPLE LINE - hWP-2 This letter is in reference to the Notice of Deviation contained in the WNP-2 InspectionReport(No. 50-397/83-58) dated January 25, 1984. This Notice of Deviation concerns the routing of an unshielded reactor coolant sample line through occupied portions of the reactor and turbine building at WNP-2.
Regulatory Guide 8.8, Revision 3, paragraph C.2.b.(6) states: "The exposure of station personnel to radiation from pipes carrying radioactive material can be reduced by means of shielded chases." Paragraph 12.1.2.1.b(2) of the applicant's Final Safety Analysis Report states that:
" Radioactive pipes are routed so that radiation exposure to plant personnel is minimized.
The extent to which radioactive pipes are routed through normally accessible areas is minimized." Contrary to this, a 3/8" reactor coolant sample line runs unshielded and uninsulated in close proximity to personnel walkways l
for most of its transit'from the X77 penetration at level 501' of the reactor building to Sample Room 116 on level 441' of the turbine building.
In response to NRC concerns, the licensee performed a shielding calculation to evaluate the potential dose rates from the unshielded reactor coolant sample line during normal plant operations. The licensee utilized the long-lived activities from corrosion and fission products, as well as the coolant
. activation products, in their evaluation. This evaluation concluded that the exposure rate from the sample line would be 0.66 mR/hr at one foot. Since the areas which this line traverses are designated less than 2.5 mR/hr, the licensee concluded that shielding this sample line for "ALARA" considerations is not necessary.
The staff disagrees with the licensee's conclusion not to shield the reactor coolant sample line. The licensee's calculation of 0.66 mR/hr at one foot from the sample line does not account for plateout and potential crud buildup l
in this line, both which would result in higher sample line dose rates ever time. The presence of high percentages of failed fuel could also account h
i 840328 giBacIOS000g c"'cca P"*"c>
pe reau aie no semaco ano OFFICIAL RECORD COPY
- ".5"""
'2"
c --
i
- O 27
.7
~
Q
%d i
- whg A. Schwencer for dramatic increases in the dose rates of this reactor coolant sample line.
Following the accident at THI, Unit 2, general area garana exposure rates of 500 mR/hr to 1 R/hr were measured several feet from the unshielded sample lines located in the Unit 1 hot machine shop. Because there is a potential for the radiation field from this sample line to greatly exceed the licensee's estimate, the staff reconsnends that this sampling line be routed so that the potential radiation exposure to plant personnel is minimized. This can be done by shielding the present line, or relocating the line to a shielded pipe chase. Conducting regular health physics surveys of this sample line to detect increases in radiation levels is not an acceptable alternative to shielding the line.
On the basis of the above,' the staff concurs with the I!otice of Deviation concerning this issue which is contained in Inspection Report ido. 50-397/83-58.
This review was performed by C. Hinson, RPS/RAB.
pristaal signed by F. J. Consel Frank J. Congel, Chief Radiological Assessment Branch Division of Systems Integration cc:
R. hattson D. Muller
- 0. Lynch R. Auluck G. Yuhas, Reg. V C. liinson l
l l
l l
l A
,/
"'c ' >.DSl;,RAB,.
.D.S fdA.[.,,,,,,
. p.S.I.:.R.AB....
Chins.. sj DDTLWch FJCoogel pmcusp o=> o3/ W 84 o3M/84 03/M4 we roau am no semacu o24o OFFICIAL RECORD COPY
- as "o " 8 3-'* "
--