ML19317F187

From kanterella
Revision as of 17:02, 21 February 2020 by StriderTol (talk | contribs) (StriderTol Bot change)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search
AO-287/74-01:on 740903,weld Failure Dectected on Reactor Coolant Pump Seal Injection Piping During Leak Test.Caused by Lack of Fusion Between Weld Fill Matl & Elbow Fitting. Defective Weld Repaired
ML19317F187
Person / Time
Site: Oconee Duke Energy icon.png
Issue date: 09/17/1974
From:
DUKE POWER CO.
To:
Shared Package
ML19317F183 List:
References
NUDOCS 8001080901
Download: ML19317F187 (2)


Text

_

a 4

  • *
  • p DUKE POWER COMPANY OCONEE UNIT 3 Report No.: A0-287/74-1 Report Date: September 17, 1974 Occurrence Date: September 3, 1974 Facility: Oconee Unit 3, Seneca, South Carolina Identification of Occurrence: Weld failure on reactor coolant pump seal injection piping .

Conditions Prior to Occurrence: Unit in hot, shutdown ,

Description of Occurrence:

On September 3, 1974, during a 2285 psig leak test of the Oconee Unit 3 Reactor Coolant System, a weld failure was detected on the seal injection line to the 3B2 Reactor Coolant Pump. The leak, located on a 115 inch socket welded joint between valves 3HP-286 and 3HP-67, was detected by visual inspecticn. Cooldown of the Reactor' Coolant System commenced. An attempt to isolate 3B2 Reactor Coolant Pump by closing valves 3HP-146 and 3HP-67 was not successvul due to valve leakage. The leak was stopped by isolating seal water to all reactor coolant pumps by closing valves 3HP-138 and 3HP-140 and stopping the high pressure injection pump.

An inspection of the veld showed a small pinhole on the surface of the weld.

Designation of Apparent Cause of Occurrence:

The apparent cause of the failure of this weld is diagnosed as lack of fusion (cold lap) between the weld filler material and the elbow fitting.

The lack of fusion was covered near the top of the weld by the weld pool.

Prolonged pressure on the small surface forced a pinhole to blow out, resulting in the leak. .

Analysis of Occurrence:

The reactor coolant pumps are designed such that either the Component Cooling System or the High Pressure Injection System will provide sufficient cooling for the pump seals. Only a small amount of reactor coolant outleakage will occur with the seals operating properly. The actions taken to isolate tne leak did not affect the integrity of the reactor coolant pump seals.

The system had previously passed the initial dye penetrant test and had been successfully hydrostatically tested to 4575 psig.

This event occurred prior to initial criticality for Unit 3. Therefore, any leakage would not have resulted in any contamination. The seal water is 8001080 9 g

. . m ,,*

?

  • supplied from the High Pressure Injection System and is normally at 110 F and 3000 psi. This water would also be relatively free of contaminants even af ter criticality and would not have adverse affects upon the environ-ment of the Auxiliary Building. Since the integrity of the reactor coolant pump seals were not affected and there was no contamination associated with  :

the weld failure, it is concluded that the health and safety of the public was not affected.

Corrective Action:

Repair of the defective weld was performed in accordance with applicable maintenance and quality assurance procedures. This included grinding out the defective portion of the weld to approximately one inch on either side of the pinhole. The full defect was verified to have been completely removed.

Six welds, similar to the defective weld, were also dye penetrant inspected.

Five of these welds were made by the same welder and inspected by the same inspector as the defective weld. These welds were chosen for testing because of the proximity to the defective weld and since all the welds were performed during the same period of time and probably with the same equipment. No defects were discovered in these welds.

1

. g S

- .- -.-.v .,