TENNESSEE VALLEY AUTHORITY CHATTANOOGA. TENNESSEE 37401
51 1573 Lookout Place
FEB 27 1987
WbRD-50-390/84-17 10 CrV 50.55(e) WMRD-50-391/84-17
U.S. Nuclear Regulatory Comaission / Attn: Document Control Desk V Office of Nuclear Reactor Regulation Washington, D.C. 20555
Attention Dr. J. Nelson Grace
MATTS BAR NUCLEAR PLANT UNITS 1 AND 2 - DEFICIENT WULDS FOR HANGER LUGS O
ASMN CODE PIPING - WUBD-50-390/84-17, USD-50-391/84-17 - ITwIT RPOR
The subject deficiency was initially reported to IRC-Region II Inspector Steve Elrod on January 27, 1987, in accordance with 10 CFR 50.55(e) as ICR W-S18-P-8 for unit 1. 8CR WN 7192-S Mas initiated to document the potential for this deficiency for unit 2. A similar deficiency was reported previously (W•BD-50-390/84-17, UBRD-50-391/84-17) but was downgraded to nonreportable in our May 18, 1984 report to NC. Because of the similarity to the previous deficiency, we are reopening that itm rather than reporting this as a new construction deficiency. Enclosed is our interia report. We expect to submit our final report on or about January 15, 1988.
If there are any questions, please get in touch with R. D. Schuls at (615) 365-8527.
Very truly yours,
TENNESSE VALLEY AUTHOIITY
R. ridley, Director Nuclear Saf ty and Licensing
Enclosure cc: See page 2
B '
An Equat Opportunity E imployei
U.S. Muclear Regulatory Comission FEB 27 1987
ec (Enclosure): U.S. Nuclear Regulatory Comission legion II Attn: Mr. Gary G. Zech, Director, TVA Projects 101 Narietta Street, NW, Suite 2900 Atlanta, Georgia 30323
Reeccrds Center Institute of Nuclear Power Operations 1100 Circle 75 Parkway, Suite 1500 Atlantas, Georgia 30339
U.S. Nuclear Regulatory Conmission Watts Bar Resident Inspector P.O. Box 700 Spring City, Tennessee 37381
ENCLOSURE
WATTS BAR NUCLEAR PLANT UNITS 1 AND 2
DEFICIENT WELDS FOR HANGER LUGS ON ASKE CODE PIPING
SCR W-518-PS, SCR WRI 7192, SCR WBN 5559
WBRD-50-390/84-17, WBRD-50-391/84-17 10 CFR 50.55(e) INTERIM REPORT
DESCRIPTION OF CONDITION
During rework activities on unit 1 supports on the Essential Raw Cooling Water
(ERCW) system, it was discovered that welds joining the piping shear lugs to
the pipe did not achieve complete penetration in conformance with the design
drawings which specify full penetration welds. In addition, the welds on some
of the shear lugs did not extend the entire length of the lug. Discrepancies
were also found with the specified root gap on a small number of the known
discrepant welds. On unit 1, weld deviation reports have been written which
identify other full penetration shear lug welds in various other systems that
have a lack of the specified full penetration. This nonconformance was
reported under SCR W-518-P for unit 1. SCR WBN 7192 was written for unit 2 to
facilitate evaluation for generic nonconformances. A photograph of a section
of a discrepant unit 1 shear lug which has been polished and etched is
attached.
In March of 1984, NCR WBE 5559 was initiated to address a generic problem of a
lack of full penetration on shear lug welds due to the welds not being
backgouged after the initial weld pass. The May 18, 1984 final report
dispositioned the nonconformance as use-as-is. TVA has since discovered
errors in this disposition for Class 2 and 3 pipe and thus the use-as-is
disposition was not adequately justified and will be reevaluated.
The most likely cause of the lack of full penetration in the welds is the same
as that identified in NCR WBN 5559, i.e., the welder did not backgouge before
welding the second side of the lug as required by General Construction
Specification G-29M, drawing l.M.1.2-12, although improper fitup between the
lug and pipe could have also contributed to the condition. The lack of weld
along the entire lug length and the lack of the specified gap apparently
resulted from poor craftsmanship and inadequate inspections.
This deficiency was discovered and reported by a welder who was performing
unrelated rework activities for the purpose of resolving zero period
acceleration (ZPA) concerns. His action was indicative of his alertness and
comuitment to quality.
SAFEUY IMPLIrATIONS
The lack of a lull penetration weld results in an increased stress in the weld
and the pipe w-l1. Experience shows that the stress ratio (applied
stress/allowatle stress) will normally be more critical in the pipe wall. The
increased striss could, in some instances, result in some shear lugs inducing
stress in th,' pipe wall that would exceed the allowable stress. The
overstressed condition in the pipe wall could potentially result in some local
yielding at ýhat point. However, the stress intensity would not likely be
sufficient to cause a loss in the pressure boundary, thereford, the pipe and
system would continue to fulfill its design function. However, a complete
analysis has not been performed on all discrepant welds to determine the
likelihood of a t•ilure under all design loading conditions.
INTERIM PROGRESS
Five supports have been identified on unit 1 with ~repartial penetration
shear lug welds. An analysis was performed to determine the acceptability of
the induced stresses in the weld and pipe wall. Of these, four supprts were
found to be within the acceptable limits of the ASHE code. However, the
calculations for the fifth support show that the stress in the pipe exceeds
the allowable stress by 46 percent. These calculations were based on very
conservative assumptions and it is highly unlikely that the overstress would
result in a loss of pressure boundary. These five supports have already bad
their shear lugs removed due to other rework activities and new shear lugs
will be installed using the proper procedures. This action will be sufficient
to correct the deficiencies for these supports.
TVA is currently evaluating the existing welds on shear lugs by ultrasonic
examination. For this ultrasonic examination. acceptance criteria will be
established for ful'1 penetration welds. On unit 1. 100 percent of the shear
lugs on the safety-related systems will be examined ultrasonically. Welds
which do not meet the acceptance criteria will be reanalyzed, using
conservative design assumptions, to ensure that the existing welds are within
ASKE Code requirements. For those shear lugs which do not meet the ASHE Code.
modifications or repairs will be performed.
From the unit 1 results, a trend can be identified to aid in the development
of an evaluation program for unit 2. It should be noted that 40 shear lugs oau
unit 2 have already been examined ultrasonically. All 40 shear lug welds were
found to have the specified full penetration.
A final report will be submitted to NRC on or about January 15. 1966.
WATTS BAR NUCLEAR PLANT - UNIT 1
Shear Lug Removed from ERCW System
Polished to 600 Grit and Etched