ML20084K810
| ML20084K810 | |
| Person / Time | |
|---|---|
| Site: | Quad Cities  |
| Issue date: | 06/18/1974 |
| From: | Kalivianakis N COMMONWEALTH EDISON CO. |
| To: | Oleary J US ATOMIC ENERGY COMMISSION (AEC) |
| Shared Package | |
| ML20084K663 | List: |
| References | |
| NJK-74-121, NUDOCS 8305190548 | |
| Download: ML20084K810 (4) | |
Text
[
A.
Comm zith Edison
(
/
h (C1
^
/
Quad-C 3en:. rating Station
( _
/
~
g,D ro
~
Post Offics Box 216 Q*M,/ Cordova.Ilhnois61242 Telephone 309/654-2241 fo.,
HJK-74-121 h~'y UUN24 tg74s Lts. m v ~ n:ny June 18,1974 9;dcy m.1 ff r e.
4 W Sect.sa j
/
Mr. John F. O ' Leary, Di rector Directorate of Licensing Regulation U. S. Atomic Energy Commission Washington, D. C.
20545
Reference:
Quad-Cities Nuclear Power Station Unit 2 Docket No. 50-265, DPR-30, Appendix A, Section 1.0.A.3, 6.6.B.l.a
Dear Mr. O ' Leary:
The purpose of this letter is to inform you of the details of an abnormal occur-rence which was discovered at 0225 hrs on June 10, 1974 With Unit 2 operating at 200 MWe, the feedwater low flow regulating valve, A0-2-643, f ailed at the junc-tion of the tubular portion of the valve body on the inlet side with the spheri-cal portion. The severed valve body caused a decrease of feedwater flme to the reactor; a reactor scram from vessel 1 me level resulted. This occurrence was reoorted to the Reaion t il Directorate of Regulatory Doerations by telephone and t e i 3,..,
n t,
- r..
n s.,,..,5..,3
,,,,, t. 3, y.
7,. a.
m.,...,.
PROBLEM AND INVESTIGATION The Unit 2 load had been reduced to 200 MWe at 2115 hrs on June 9,1974, due te high reactor water conductivity. At the time of the occurrence, the shift engi-neer was in the reactor building on a routine inspection. He heard a strange noise and called the control room to inquire. The control room operator reported that conditions were normal, but that feedwater flow was increasing, and that vessel level was experiencing small oscillations. The shift engineer oroceeded to the Unit 2 reactor feed pump area and found water coming through the door to the reactor building. He noticed steam and water thrcughout the feed pumo area.
He then returned to the reactor building and proceeded toward the feed cump area from the turbine building. When he arrived, he again found large quantities of water on the floor and steam filling the area. At 0225 hrs on June 10, 1974 Uni t 2 scrammed f rom low reactor vessel water level. At some time during the occurrence, the fire deluge systen in the vicinity of the feed pur.ps was activated due to the heat of the released steam and water. This released additional water into the area.
The source of the steam and water was the failure of the feedwater Icw flow regu-lating valve, a 4 inch, air operated, ASA 900 lb. class globe valve nanufactured by the Cores-Vulcan Division of the Blaw Knox Company. The valve had severed on the inlet side where the tubular and spherical sections meet.
)
(!
t'
\\
l
'i i, l' O r) 8305 r
?~198EA"OMg;g
('
y, i
y
hs Juna 18, 1974 f
' Er. John F. O 'Lstryd,
On the instruction of the shif t engineer, the feedwater flow regulating station was isolated and the reactor core isolation cooling system was placed in operation to control vessel level. This was completed at 0245 brs. Rad Protection personnel were immediately called to survey and rope off the affected areas. The total amount of water released to the plant as a result of the occurrence has been esti-mated at 70,000 gallons; 55,000 gallons came from the feedwater line, and 15,000 gallons came from the deluge system.
Initial estimates were that 360 gallons of water went to the Unit 2 oil separator which discharges to the spray canal return to the circulating water inlet bay, it was later determined that 800 gallons of water actually went to this separator. The remainder of the water went to rad-waste or to the condensate pump basement via the floor drain system. Water samples taken at the plant circulating water inlet, and the Davenport, Iowa, and East Moline, Iilinois, water works indicate that no contamination was released to the environs.
The sample results are as follows:
4 GROSS BETA Feedwater Sample 2.6 x 10~ uc/ml Condensate Pump Basement 2.4 x 10 uc/ml
-6 U-2 011 Separator 1.9 x 10 uc/ml
~0 Circ Water Inlet 1.1 x 10 uc/mi
~9 n3yenport w, gor vori,=
9.6 v in oc 41
~9 East Moline Water Works 2.3 x 10 uc/ml
,131 Davenport Water Works 0
' East Moline Water Works 0
U-2 Oil Separator 4.5 DPM/mi Feedwater Sample 235 DPM/ml Condensate Pump Basement 189 DPM/ml EVALUATION AND CORRECTIVE ACTION A meeting was held on June 11, 1974, to discuss the mode of failure of the valve and to determine the cause of the failure. Four independent examinations of the failed valve were made by the metallurgical specialists called to the site. Their findings were:
1.
The fracture was located at the junction of the spherical portion of the valve body with the straight tubular section.
Jun3 18, 1974
- Mr. ' John F. O 'Lzryn n b
2.
Machining of the outer surface of the tubular section resulted in wall thicknesses reduced to values ranging from.315 in. to.447 in.
The design required thickness was.562 in.
3.
Machining produced a sharp corner at the point of the failure.
It was concluded that the point of failure represented the point of maximum stress.
A zone of progressive crack growth preceded the failure. The progressive cracking occurred primarily as a result of the stress concentrations caused by the sharp corner, and was facilitated by the reduced wall thickness. This was compounded by cycIlcal loading and vibration encountered during normal service. When the crack reached critical size, the failure occurred.
The reduced wall thickness and accompanying sharp corner were the result of a machining error whereby too much metal was removed from the valve wall while mak-Ing the weld preparation for installation of the valve.
A complete magnetic particle and visual inspection of the 6 inch low flow piping, including the saddles connecting the low flow piping with the 24 inch feed header, was conducted.
In addition to the failed valve, a 6 foot section of vertical pipe 0
and a 90 elbow downstream of the valve have been replaced. The vertical pipe section had a permanent set of approximately 10 and the inner radius of the elbow was cracked as a result of the impingment of the water and steam released from the severed valve.
Copes-Vulcan has provided a replacement valve body of ASA 2500 lb. class machined to accept the existing 900 lb. valve bonnet and Internals. Magnetic particle inspections of both full flow feedwater regulating valves ware performed. Additionally, all welds 4'
!!ne ruil !!cer v:!ve ??ere size n ; net h P t W en the p!peng ::: : ::ce eritn :n:
Inspected. Following installation of the replacement valve and piping, all welds were radiographed, magnetic particle tested, and leak tested under operating conditions prior to startup of the Unit. All tests were satisfactory.
An ultrasonic test was performed on the Unit I low flow regulating valve to verify that its wall thickness was within specifications. All measurements were.500 inches or greater, which meets minimum wall thickness requirements.
SAFETY IMPLICATIONS At no time during this occurrence was the safe operation of the reactor threatened.
All reactor parameters responded satisfactorily.
It is felt that the actions of the operating crew were completely adequate to maintain the safe condition of the plant.
There was neither excessive exposure to plant personnel nor contamination released to the environs as a result of this occurrence. Plant contamination was considered low level and was easily decontaminated. Decontamination of the condensate pump basement was completed on June 15, 1974, all other affected areas were decontaminated within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
All replacement parts meet or exceed the original design requirements. The operat-Ing leak test and operability test verified the integrity of-the repair and demon-strated that the valve is safe for operation. Unit 2 was made critical at 2116
=
4-Juna 18, 1974
.. M,r. John F. O 'Lscry hrs on June 16, 1974; the valve performed satisfactorily during the subsequent startup and power ascension.
CUMut.ATIVE EXPERIENCE No failure of this type has occurred previously at Quad-Cities Station. The in-vestigations made and the actions taken should preclude further failures of this type with these valves.
Very truly yours, COMMONWEALTH EDISON COMPANY QUAD-CITIES NUCLEAR POWER STATION
/?h',
l'Oh('
N. J. Kalivianakis '-
Station Superintendent t
NJK/ REQ /Jeh cc: Region 111, Directorate of Regulation Operations J. S. Abel T
i i
i L.