ML20114F273
| ML20114F273 | |
| Person / Time | |
|---|---|
| Site: | Oyster Creek |
| Issue date: | 03/28/1968 |
| From: | Doan R US ATOMIC ENERGY COMMISSION (AEC) |
| To: | |
| Shared Package | |
| ML093631134 | List:
|
| References | |
| NUDOCS 9210120425 | |
| Download: ML20114F273 (3) | |
Text
.-_- _
c M.AR 2 81300 i
'4 y t
c-m,
~Z.Z&,/ -[
C0!4GTPS ON JERSEY CFJTIRAL F%'ER & LIGiff COMPMIY'S REPOffT ON OYS1Ut CREEK PPESSUF5 VESSEL FEFAIR PROGRAM e'
( Arendnent tio. 35 dated March J4,1968) j BY
/
RICHARD L. LO/Ji W
The tr:;t sigilricant revelation in the subject report (hereafter called oc-35) is the apparently conplete failure of the applicant and his contractors to fulfill their process control and inspection obligations during the initial field welding of the control Ivi drive housings to the stub tubes in the bottom of the pressure vessel.
This seems to be the only possible explanation of the very large number of welding flaws that could readily have been detected early in the welding operation but actually were not observed until after the hydrostatic testing of the vessel had shown a leak that indicated something was wrong. Subsequent examination of the welds by nethods (dye penetriun and ultrasordes) that should have been used to verify the integrity of the welds durinC the course of the welding operation led to the discovery for tle first tire of the following amasing list of welding defects as set forth on page 2-1 of the report.
1.
Localised intergranular cracking in 123 of a total of 1'l stainless steel stub tubes. These cracks were entirely in the stainless steel adjacent to the Inconel shop welds between the vessel bottom head and the stub tubes and varied to the full stub tube circumference in length and up to 5/16" in depth. Ibese are shown in Fi;2re 1.
2.
Randomly oriented, shallow linear and non-linear defects on the outer surfaces of 43 of the stainless steel stub l
tubes. These defects showed intergranular attack,.
penetrated to a maxJrlum depth of approxinutely 5/32" and were extremely fine.
These are shown in Figure 2.
I 3.
Porosity arNor lack of fusion welding defects in each of the 137 field welds joining the stub tu' eses and the l
control rod drive mechanism housings.
The defects in one of these field welds accounted for the small leak detected during the hydrostatic test.
i 4.
Minor slag inclusions in a number of the Inconel field welds attaching the flux monitor tubes to the reactor vessel head.
l l
99V 2-7
/
9210120 m 920sco PDR ORG NRCHIST PDR
'7
. 5 Appmximtely 19 smil and scattered penetrant indi-cations at the junction of the stub tube shop weld and the reactor vessel cladding. Rese indications am considered to be fabrication defects.-
As indicated above, the infomation presented in OC-35 seems to m to constitute strong evidence that if there was any process control, supervision or inspection at all during the field welding amund the stub tube areas in the bottom head of the Oyster Creek pressure vessel, it was completely ineffective.
The question then arises as to how much credibility can be placed on the plans and pmmises set forth by the applicant on page 2-5 of 0c-35 for assuring the quality of the proposed repair pmgram. Rese 'tquality assurance" plans are listed on page 2-5 as rollows:
1.
All repair work in the vessel will be perfomed under the direct supervision of experienced GE arid Combustion Engineering personnel.
2.
In addition to ASE Code qualification of welders and welding procedures, welders will be qualified for all mnual welding in a rock-up which simulates the accessibility conditions in the vessel.
In this regatxb about 50% of the CRD housings and the upper portions of all flux monitor housings will be renoved fmm the vessel to provide adequate accessibility for these repairs.
3
%e perfornance_of automatic welding machines for weld--
overlay cladding and CRD housing field weld r7 placement will'be nonitored to assure proper operation. In addition, _ during the production welding, periodic weld test samples will be made for destructive examination.
4.
In process inspections will-include:
i
.- Liquid penetrant _ examination of all ground or machined areas, final welds, and when deeper than-3/16", intercediate weld examination before completion.
Ultrasonic inspection of al1~ CRD housing field welds.
Ultrasonic inspection of the shop welds on a large number of stub tubes (40 to 50) from which CRD _ _
housings are mroved.
%ese inspections will be performed before and after the cladding operation to verify thht the overlay welding has not affected the shop weLis.
5.
Acceptance Nandards for this repair work will be in accordan% with the requirements of the ASE Code,Section III.-
l
- e.
e d With regard to the foregoing program for assuring the quality of the proposed repair job, I would guess that the applicant and vendor have gotten into so much difficulty over the stub tube welding defects that they will use all of the diligence in process control, superviaion and inspection that should have been employed in the first place to assure that the welding and repair job is top quality. But it is disturbing to contemplate the possibility that such diligence my not be exereised except in tims of trouble. What about other areas of construction where potentially seriot.s but as yet undiscovered defects of one kind or another may still exist? These and other similar considerations seem to me to overshadow in importance the mom direct problem of whether or not the welding repair program proposed by Jersey Central is acceptable fm m a safety viewpoint.
It is ray own feeling that anything short of replacing all of the stub tubes in the bottom of the pressure vessel will substantially enlarge the area of uncertainty regarding the long-term K
integrity of the pressura "9ssel.
If, as is claimed in 00-35, the weld $ng difficulty originated with the "sensithed condition" of the stub tubes resulting from the vessel heat tmatment at combustion Engineering it might be possible to chango the heat treatment or the traterial of the stub tubes so that they would not be left in a sensiti::ed condition.
If so, that would give rise to the larger problem of what to do about all the other BWR pressure vessels now in various stages of fabrication that may also have sensitized mterial in the stub tubes. 'Ibe 308L weld mtal overlay proposed for the repair program on the Oyster Creek vessel my be a successN1 way of avoiding the necessity of replacing the stub tubes, but a multi-million dollar pressure vessel could be an expensive experimnt.
It seems reasonable to ask whether the rapid escalation in the demand for large pressure vessels may be bringing the industry face to face with the possibility that they really do not have all the -infomation they need to cope successfully with all of the complex fabrication and installation problems on pressurc vessels.
Distribution:
HLPrice C}Teck
!Wann Cmenderson PAMorris LDlcw
)