ML19296D326

Final Deficiency Rept Re Leak in Bottom Plate of Condensate Storage Tank Caused by Crack Near Circumferential Weld Seam. Deformed Areas in Bottom Plates Cut Out & Replaced W/New Plates

ML19296D326
Person / Time
Site:	Summer
Issue date:	02/04/1980
From:	Crews E SOUTH CAROLINA ELECTRIC & GAS CO.
To:
References
NUDOCS 8003030495
Download: ML19296D326 (3)

Text

'

% ...'., 3 %

~

SOUTH CAROLINA ELECTRIC a GAS COMPAN,Y - , j CoLums A,Sourn CanottwA seria

t. H. Ca t w s. J a.

"i.'.".'I*i' '",! $.*!,*.7'

/f '

. February 4, 1980 a 2/

United States Nuclear Regulatory Commission ATTN: Mr. James P. O'Reilly, Director Region II, Suite 3100 101 Marietta Street, N. W.

Atlanta, Georgia 30303

Subject:

Virgil C. Summer Nuclear Station Reportable Item in Accordance with

• 10CFR50.55(e)

Gentlemen:

The purpose of this letter is to provide a final written report for the item previously identified as " Condensate Storage Tank". This item was identified to Region II on July 10, 1979 with an interie report sub-mitted on August 17, 1979.

Nature of Condition As described in the report of August 17, 1979, during start-up and testing activities, a leak in the bottom plate of the condensate storage tank was identified. This tank had been accepted from the Contractor, Pittsburgh-DesMoines Steel Company (PDM), after testing and inspection had confirmed acceptability. A subsequent in-tank inspection revealed a crack, about 6" in length, near the circumferential weld seam between the bottom plate and the 8" diameter drain pipe at the bottom-center of the tank. In addition, at the center of the tank, the 8" drain line had apparently risen some 5" from its initial position, with the floor plate assuming a " draped" configuration at the raised portion. The plate, as part of the bottom shell, was resting on a prepared and compacted sand fill retained by a perimeter concrete ring under the tank wall, with a 4-foot thick concrete slab under the entire foundation.

Investigation also revealed that the vent screen, provided to shield the vent intake, had been specified as a fine mesh screen (80 mesh) and had been inadvertently painted over, thereby restricting air flow through the vent.

80!?

bc lll 8003000

Mr. James P. O'Reilly

• February 4,1980 Page Two Cause Several possible causes are noted here. No one cause can be singled out as being totally responsible for the condition described, and it is considered that at least two of the causes noted here could have contributed to the failure.

(1) After completion of the tank and prior to construction of the valve pit, the drain line projected horizontally beyond the perimeter ring wall. It is conceivable that a piece of construction equipment, such as a bulldozer, could have ,

inadvertently hit the projecting portion of the pipe causing .

it to move inward and/or rise. Examination of the pipe, uncovered during the repair work, yielded no evidence to confirm this conjecture.

(2) Settlement of the sand fill'due to inadequate compaction may have occurred. Calculations indicate that under certain assumed conditions of non-support under the plate, a shallow depth of water in the tank would be sufficient to cause an overstressed condition in the floor plate.

(3) A partial vacuum in the tank, cause by restricted air flow through the painted-over vent screen could, under total and rapid drawdown conditions, contribute enough uplifting force to raise the floor plate. A loop seal in the overflow line would limit such a vacuum to about 21 oz/in2 , Thus when the level of water in the tank dropped below about 3 feet in depth the uplifting force would exceed the downward pressure of the water, thereby tending to raise the floor plate. This could have occurred during piping system and tank flushing operations.

It should be noted that the sides and roof of the tank, (of heavier construction) exhibited no evidence of distress.

Safety Implications Although it is not considered that the leakage through the approximately 6" long tear in the floor plate was sufficient to lose enough water to directly affect the safety of the plant, it is reasonable to assume that, with time, the tear may have increased in size during an OBE or SSE with the resultant loss of condensate water from the tank. Thus, in the event of a simultanous loss Of ?eedwater accident, water 'rse the condensate storage tank would not be snailable to the emergency f2edwater pumps. Water from the service water system

. oula, however, be made available to the emergency feedwater pumps by remote

.alve alignment from the main control board.

Mr. James P. O'Reilly -

February 4, 1980 Page Three Action to Correct Condition Permanent corrective action was initiated in October 1979 and completed December 1979. Deformed areas in the bottom plates were cut out and replaced with new plates. Sand fill was replaced as required and compacted in the areas exposed by the removed plates. The new plates were welded in place and then grout was pumped under the plates, creating a grout layer which provides close contact between the underside of the tank and the sand fill, for the entire bottom of the tank. A tear, discovered in the rubber dia-phragm, was vulcanized in place. All work was performed and inspected under approved site procedures. Final interior painting was performed over the repaired areas, with further touch-up painting still remaining to b'e performed later this year.

As an additionnl precaution, the painted-over vent screen was removed and replaced with a 1/4" mesh screen.

Action to Prevent Recurrence It is considered that the actions described above are adequate to bring the tank into a satisfactory condition for long term operation and no recur-rence of the problem is anticipated. As a further precautionary step, the fine mesh screens in tanks of similar design, the Reactor Makeup Water Storage Tank and the two Boric Acid Tanks, have also been replaced with 1/4" mesh ,

screens.

SCE&G believes that the actions outlined above has' adequately resolved the reported conditions. This is considered a final report of this item.

All records with respect aegions taken are available at the Jummer site for NRC review, r l

Very rt ly yours,

," v

~

E. . Cr vs, Jr.

EHC:rh

,