05000331/LER-2003-006

LER-2003-006,

Docket Number
Event date:	2-5-2003
Report date:	01-26-2004
Reporting criterion:	10 CFR 50.73(a)(2)(iv)(A), System Actuation
Initial Reporting
ENS 40353	10 CFR 50.72(b)(2)(iv)(B), RPS System Actuation, 10 CFR 50.72(b)(3)(iv)(A), System Actuation
3312003006R00 - NRC Website

FACILITY NAME (1) Duane Arnold Energy Center DOCKET NUMBER (2) LER NUMBER (6

I. Description of Event:

On November 25, 2003. at about 1515, while operating at approximately 20% power, main condenser vacuum began degrading. Power was being held at 20% power so that helium testing could occur with plant systems in normal operation, yet at a low enough power level to minimize potential effects of a loss of condenser vacuum. Additionally, testing at this lower power level minimized dose accumulated during the testing. As a result of the degrading vacuum, operators entered Abnormal Operating Procedure_(A0p)_691 , NCondenser High Backpressure" and attempted to lower reactor power in order to avoid a main turl—iFiFtrip.

Vacuum continued to degrade to a point that a manual reactor scram was required. A manual reactor scram was inserted at about 1522, on November 25, 2003. All plant systems responded as expected to this event including Primary Containment Isolation System groups 2. 3. and 4, which isolated due to reactor water level dropping below 170". The reactor water level drop is expected following a scram from power due to void collapse in the reactor vessel. Reactor water level was restored to normal and the group Isolations were reset.

In the weeks prior to the reactor scram, the site had been investigating abnormally high main condenser air in-leakage. The cause of the degrading condenser vacuum was from this excessive air In-leakage. After extensive testing and investigation, the cause of the excessive air in-leakage was determined to be from a failed welded seam in the high-pressure condenser at the top of the loop seal.

II. Cause of Event:

The root cause for the manual reactor scram on degraded condenser in-leakage was a failed seam weld between the condenser loop seal top plate and the condenser shell. The seam weld failed because it was inadequate to provide structural support and long-term sealing. The weld was examined extensively in an attempt to determine the cause of its failure. A representative from the condenser manufacturer aided NMC in this investigation. NMC personnel, and the ----vendor representative concluded that the weld had originally been installed incorrectly. The — original design drawing required a full penetration weld, however, only a fillet weld had been installed.

Inspection of the actual weld indicated additional contributors to the welded seam failure in that some of the weld material showed evidence of erosion. This erosion was caused by exposure to high-energy steam from the discharge of main steam line bypass valve number one and the discharge from the last stage of the low-pressure turbine. The combination of weld material erosion and the flexing of the two plates resulted in the failure of the welded seam.

III. Assessment of Safety Consequences:

The condenser performs the function of collecting steam from the low-pressure turbines, condensing this steam Into water by removing energy, and providing a supply of water to the condensate pumps. It also provides a volume to collect bypass steam from the main steam system during start-ups, shut downs, and turbine trips. Additionally, under certain accident conditions, It provides a volume to collect leakage through the Main Steam Isolation Valves FACILITY NAME (1) Duane Arnold Energy Center DOCKET NUMBER (2) LER NUMBER (5)

SEOUENTIAL

� � PAGE 3 (MSIV). The identified leak from the seam weld would not have prevented the function to collect MSIV leakage, because the condenser is either at a slight negative pressure or atmospheric at the time the function would be necessary. The total volume available and the relative low pressure combine to assure that little or no main steam line leakage would escape from the condenser.

Additionally, the condenser remained available for heat removal throughout the event.

This event did not affect the availability of other systems needed to maintain safe shutdown —conditions,-remove residual heat, control the release of radioactive material, or mitigate.the � - - consequences of an accident.

Therefore, there were no actual safety consequences associated with this event. There was no effect on public health and safety as a result of this event.

IV. Corrective Actions:

The failed weld, and its companion weld in the low-pressure condenser, were both ground out and full penetration welds were installed in their places.

Other welds on the sides and bottom of the loop seal were evaluated and determined to be acceptable as-is.

Appropriate periodic condenser inspection plans, including welds and interior surfaces in the vicinity of high-energy steam flow, are being evaluated. CAP 30391 is tracking this action.

Additional focused inspections of both condensers during the next refuel outage (RFO 19) to inspect selected shell seam welds and other areas are being evaluated. CA 36995 is tracking this action.

V. Additional Information:

Previous Similar Occurrences:

A review of LERs at the DAEC over the last 3 years identified no LERs with similar causes.

EIIS System and Component Codes:

Condenser System: SG Reporting Requirements:

A 10CFR50.72(b)(3)(iv) notification was made on November 25, 2003, and is listed as event number EN 40353. This report is being submitted pursuant to 10CFR50.73(a)(2)(iv)(A).