Information Notice 1994-08, Potential for Surveillance Testing to Fail to Detect an Inoperable Main Steam Isolation Valve

Potential for Surveillance Testing to Fail to Detect an Inoperable Main Steam Isolation Valve
	ML031070034
Person / Time
Site:	Beaver Valley, Millstone, Hatch, Monticello, Calvert Cliffs, Dresden, Davis Besse, Peach Bottom, Browns Ferry, Salem, Oconee, Mcguire, Nine Mile Point, Palisades, Palo Verde, Perry, Indian Point, Fermi, Kewaunee, Catawba, Harris, Wolf Creek, Saint Lucie, Point Beach, Oyster Creek, Watts Bar, Hope Creek, Grand Gulf, Cooper, Sequoyah, Byron, Pilgrim, Arkansas Nuclear, Braidwood, Susquehanna, Summer, Prairie Island, Columbia, Seabrook, Brunswick, Surry, Limerick, North Anna, Turkey Point, River Bend, Vermont Yankee, Crystal River, Haddam Neck, Ginna, Diablo Canyon, Callaway, Vogtle, Waterford, Duane Arnold, Farley, Robinson, Clinton, South Texas, San Onofre, Cook, Comanche Peak, Yankee Rowe, Maine Yankee, Quad Cities, Humboldt Bay, La Crosse, Big Rock Point, Rancho Seco, Zion, Midland, Bellefonte, Fort Calhoun, FitzPatrick, McGuire, LaSalle, Fort Saint Vrain, Shoreham, Satsop, Trojan, Atlantic Nuclear Power Plant, Crane
Issue date:	02/01/1994
From:	Grimes B; Office of Nuclear Reactor Regulation
To:	;
References
	IN-94-008, NUDOCS 9401260242
	Download: ML031070034 (10)
	v • d • e

UNITED STATES

NUCLEAR REGULATORY COMMISSION

OFFICE OF NUCLEAR REACTOR REGULATION

WASHINGTON, D.C. 20555

February 1, 1994 NRC INFORMATION NOTICE 94-08:

POTENTIAL FOR SURVEILLANCE TESTING TO FAIL

TO DETECT AN INOPERABLE MAIN STEAM ISOLATION

VALVE

Addressees

All holders of operating licenses or construction permits for nuclear power

reactors.

PurDose

The U.S. Nuclear Regulatory Commission (NRC) is issuing this information

notice (IN)

to alert addressees to a potential for surveillance testing to

fail to detect that a main steam isolation valve is mechanically bound and

will not close. It is expected that recipients will review the information for

applicability to their facilities and consider actions, as appropriate, to

avoid similar problems. However, suggestions contained in this information

notice are not NRC requirements; therefore, no specific action or written

response is required.

DescriDtion of Circumstances

On April 19, 1993, while performing maintenance to repair a presumed faulty

limit switch on a main steam isolation valve, the licensee for the River Bend

Station (River Bend) found that the valve was mechanically bound and would not

close as required. The valve is a 24-inch-diameter (nominal), spring and

pneumatic closing, pneumatic-opening, internally balanced, poppet-type globe

valve manufactured by the Atwood & Morrill Company Inc.

Plant operators had

previously performed partial stroke surveillance testing of the valve on

February 27 and April 1, 1993, but did not detect that the valve would not

close. The licensee later determined that the testing failed to detect that

the valve was inoperable because the test did not adequately consider the

design of the valve and the positioning of the limit switch arm in relation to

the valve poppet travel.

The licensee determined that the valve would not close because improper

clearances between the valve poppet and the valve body had caused excessive

wear of the guide ribs and resulted in the valve poppet becoming mechanically

bound. The excessive wear may have been avoided had the licensee installed an

anti-rotation modification recommended by the manufacturer in 1989.

Subsequent to this event, the manufacturer reported the failure to close to

the NRC under Part 21 to Title 10 of the Code of Federal Regulations and

informed affected licensees of the failure mechanism and recommended actions

to prevent recurrence. NRC Inspection Report 50-458/93-18 and Licensee Event

Report.93-006 provide additional details on the valve failure.

(9401260242)

TJ'O

02 w>

44M'j

J7JD#V&// c/S

'(I

YJ

Kx/ IN 94-08

February 1, 1994 Discussion

The original design positioning of the limit switches was such that, during

partial stroke testing, the limit switches could be actuated and indicate

movement of the main valve poppet even though the valve poppet had not

actually moved. Thus, a failure of the valve to properly stroke may go

undetected by partial stroke testing.

Under normal operation, the valve stem travels 28 centimeters [11 inches] to

fully stroke in either direction. As the valve strokes open, the first

2.5 centimeters [1 inch) of stem travel moves an internal poppet which opens

an equalizing port allowing the pressure on both sides of the main poppet to

equalize. During the remainder of the open stroke, the internal poppet lifts

the main poppet and retracts it to the fully open position. During a partial

stroke test in the close direction, as the stem (and the internal poppet)

begins to close, the main poppet also begins to close because of gravity.

However, during the event, with the main poppet stuck in the open position, the stem travelled about 2.5 centimeters [I inch] and stopped when the

internal poppet seated in the equalizing port.

There are three limit switches on the valve that are of concern in this event.

The first two switches provide a safety-related signal to the reactor

protection system that the valve is 92 percent open. The third switch sends a

nonsafety-related signal to position indicating lights in the control room

indicating that the valve is 90 percent open. The licensee had set the

90-percent-open limit switch such that stem movement of about 2.8 centimeters

[1.1 inch] was required to actuate the switch and indicate that the main valve

poppet had moved to the 90-percent-open position. However, because the limit

switches are set with a + 2 percent tolerance, actual stem travel to actuate

the 90-percent-open limit switch may be only 2.25 centimeters [0.88 inch]. In

a worst-case scenario both the 92- and the 90-percent-open limit switches

could be actuated without the main valve poppet moving.

During the partial stroke testing conducted on February 7 and April 1, the

first two limit switches (92-percent-open indication) actuated, the third

limit switch (90-percent-open indication) did not actuate. Although the

procedural step called for receipt of the 90-percent-open indication, the

operators did not declare the valve inoperable because the first two limit

switches had actuated and they assumed that the third limit switch (nonsafety)

had failed.

Later, on April 17, during maintenance on the presumed faulty

limit switch, the licensee found that the main valve poppet was mechanically

bound and that the valve would not close.

The licensee for River Bend changed the third limit switch setting to actuate

at 85 percent of the open position to ensure that its actuation during partial

stroke testing would give positive indication of poppet movement. Pending

further evaluation of these valves during the next refueling outage, the

licensee is performing full stroke testing of the valves on a quarterly basis

and intends to install the anti-rotation modification recommended by the

vendor to prevent recurrence of the excessive wear of the valve guides. Also, operations personnel have been trained on the operation and function of the

limit switches.

--- IN 94-08 February 1, 1994 The valve described in this notice is used in safety-related applications at

nuclear facilities. One such application is as a main steam isolation valve.

At River Bend and most domestic boiling water reactors, there are two main

steam isolation valves for each main steam line; an inboard valve, located

inside the drywell, and an outboard valve, located Just outside the primary

containment. After a design-basis accident, these valves are required to

close and remain closed for 1 hour1.157407e-5 days 2.777778e-4 hours 1.653439e-6 weeks 3.805e-7 months . Should these valves fail to close, offsite dose limits could be exceeded. A similar failure of a main steam

isolation valve to close had occurred at a foreign boiling water reactor.

The potential for limit switch positioning to adversely affect surveillance

test accuracy may exist for valves other than that described in this notice.

This information notice requires no specific action or written response. If

you have any questions about the information in this notice, please contact

one of the persons listed below or the appropriate Office of Nuclear Reactor

Regulation (NRR) project manager.

by

Brian K. Grimes, Director

Division of Operating Reactor Support

Office of Nuclear Reactor Regulation

Technical contacts:

Thomas F. Westerman, RIV

(817) 860-8145

Patricia Campbell, NRR

(301) 504-1311

David P. Loveless, RIV

(512) 972-2507

William M. McNeill, RIV

(817) 860-8174 Attachment:

List of Recently Issued NRC Information Notices

v

-" IN 94-08 February 1, 1994 The valve described in this notice is used in safety-related applications at