PNO-III-86-052A, on 860601,reactor Feedwater Pump Failed to Trip.Augmented Investigation Team Determined Problem Caused by Setpoint Variation.Reactor Placed in Cold Shutdown

PNO-III-86-052A:on 860601,reactor Feedwater Pump Failed to Trip.Augmented Investigation Team Determined Problem Caused by Setpoint Variation.Reactor Placed in Cold Shutdown

ML20205T487
Person / Time
Site:	LaSalle
Issue date:	06/05/1986
From:	Guldemond W, Wright G NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To:
References
PNO-III-86-052A, PNO-III-86-52A, NUDOCS 8606130248
Download: ML20205T487 (4)

7 4977)b PRELIMINARY h0TIFICATION OF EVENT OR UNUSUAL OCCURRENCE--PNO-III-86-52A Data June 5, 1986 This preliminary notification constitutes EARLY notice of events of POSSIBL'E safety or public interest significance. The infomation is as initially received without [

verification or evaluation, and is basically all that is known by the Region III staff on this date.

Facility: Comorsealth Edison Co. Licensee Emergency Classification:

LaSalle Unit 2 Notification of an Unusual Event Marseilles, IL 61341 XX Alert Site Area Emergency Docket No: 50-374 General Emergency Not Applicable

Subject:

REACTOR TRIP PROBLEM-STATUS REPORT FROM NRC AUGMENTED INVESTIGATION TEAM (UPDATE)

The Augmented Investigation Team (AIT) arrived at the LaSalle site on June 2-3, 1986, begin gathering and evaluating data, interviewing plant personnel, and developing a preliminary sequence of events. (A copy of this sequence of events is attached.)

During the initial interviews, it was determined that information provided in the original Preliminary Notification was inaccurate: At the time of the event, reactor power was 93 per cent and at no time did either reactor feedwater pump trip. The "B" feedwater pump had a control system failure which caused it to coast down; the "A" pump was always available.

Initial review by the AIT indicates that the reactor water level did drop below the setpoint where a reactor trip should have occurred. (Normal operating level is about 36 inches above instrument zero and the scram setpoint is about 13.5 inches. This setpoint is the highest of three setpoints for water level protection. Actual water level dro to about six inches which still is about 13.7 feet above the top of the reactor core)pped .

The initial review of test data indicates that the problem rests with the Static-0-Ring differential pressure switches. There are four switches -two each in two channels.

Actuation of at least one switch in both channels is necessary for a reactor trip.

In this case, just one switch was actuated.

! It is the view of the AIT at this juncture that the problem is one of setpoint variation

! (i.e. the differential pressure switches would have responded, but not at the required l setpoint). Shortly after the event the licensee recalibrated each of the four switches to a nominal 13.5 inch trip setpoint. Six hours after all control rods were fully inserted i and the alert was terminated the setpoints were checked again by actually dropping the water level to determine where the switches would actuate. The measured actuation points were 10.2, 9.0, 6.9, and 3.9 inches. The team's judgment is that the reacter prntection system j would have tripped, had the water level continued to decrease.

i h61 48 860605 PNO-III-82-052 PDR l

DISTRIBUTION:

t H..St. ED0 NRR E/W Willste Mail: ADM:DMB Chairman Palladino PA IE NMSS DOT:Trans only Comm. Zech ELD OIA RES Cor:n. Bernthal AE0D l

lComm. Roberts Comm. Asselstine SPJ.N Regional Offices INFO .2.N NSAC d l/ hg CA RIII Residen Officp \

PDR Licensee: , .3 &

~

(Corp. Office - Reactor Lic. Only)

. PRELIMINARY NOTIFICATION OF EVENT OR UNUSUAL OCCURRENCE--PNO-III-86-52A Date June 5, 1986 Th] Static-0-Ring differential pressure switches were installed in mid 1985 in Unit 2 as part of an environmental qualification modification after initial plant operation.

Identical switches are also used in Unit 1. Unit I rcmains shut down for refueling and modifications.

The AIT is also investigating the handling of information related to the Jur.e 1st

v

nt by the licensee, personnel performance and assessing the timing of the report of th incident to the NRC. The team is also reviewing a prior "ay 9, 1986, trip to determine if the circumstances of that trip may have been related to similar erratic behavior of the differential pressure switches.

The licensee has also established a special team, headed by a corporate vice president, to coordinate the licensee's investigation of the incident.

The Unit 2 reactor is being placed in cold shutdown.

Thera has been significant local news media interest in this event. NRC Region III issu d a press release en June 3, 1986. The State of Illinois has been kept informed.

This information is current as of 10:00 a.m., June 5, 1986.

CONTACT-Maya' G. Wright W.

+

uldemand (

FTS 388-5695 FTS 388-5574 l

l l

l l

6/4/86 PRELIMINARY SEQUENCE OF EVENTS AUGMENTED INVESTIGATION TEAM ,

i LaSalle Unit 2 Failure to Trip Initial Conditions, June 1, 1986, 4:20 a.m. CDT

- Unit operating at approximately 90% power

- . Twc turbine driven feedwater pumps (TDFP) in operation, A in manual, B in 3 element automatic control

- Motor driven feedwater pump (MDFP) in standby

- Reactor recirculation pumps (RR) on high speed Surveillance test in progress on the A TDFP Sequence of Events (T=0 at approximately 4:20 a.m. CDT, June 1, 1986)

T Event 0 Reactor operator (RO) actuates the test switch for the stop valve on the A TDFP, to close the valve approximately 20% in accordance with an approved surveillance procedure.

5 sec. R0 notices the A TDFP governor valve opening and reactor water level rapidly increasing on narrow range indicators.

10 sec. R0 notes that the control systems for both TDFP are " locked up" and non-responsive to demand signals.

20 sec. R0 resets lockout on A TDFP and rapidly reduces its flow to restore reactor water level from 55 inches to a normal value of 35 inches.

30 sec. Level stabilizes at 55 inches and begins to fall at about 2 inches per second.

l 1 min 10 sec. Level stable at 33 inches. Lockout on the B TDFP reset. B TDFP ramps back to zero output for no apparent reason.

I min 10 sec4 Level starts droppir.g at approximately 2 inches per second.

I mit 20 sec R0 begins manual reduction of RR flow.

l 1 min 30 sec. Level reaches the nominal reactor trip setpoint of 13.5 inches.

l (Narrow range). No trip occurs. MDFP manually started.

l l 1 min 31 sec. Level reaches 10 inches, RPS train B trips. RR pumps auto transfer l from high speed and begin a coast down to slow speed. Channel B Automatic Depressurization confirmatory level annunciator received.

I 1 r.in 35 sec. Level begins a rapid rise to 20 inches. .

1 min 40 sec. Level reaches 35 inches end stabilizes.

2 min RR pumps start on slow speed.

2 min 5 sec. 1/2 scram on RPS train B reset. Power stable at approximately 40%. Actions begin to restore full power lineup including feedwater heater restoration, RR high speed restoration, and restoration of B TDFP.

0700 Power restored to approximately 60%. Shift change occurs.

0730 Shift engineer reviews level transient. Believes level fell below nominal reactor trip point without a scram based on narrow range recorder trace. I & C support requested for level switch calibrations. Operating engineer and production superintendent informed.

1000 Data received from an in-place data collection system (installed for startup testing) indicating level had fallen as low as 6 inches during the transient.

1400 Level switch setpoint data received.

Switch Nominal Setpoint Measured Setpoint A 13.5 inches 10 inches B 13.5 11 C 13.5 13.5 D 13.5 13.5 Licensee believes an RPS malfunction may have occurred. Controlled shutdown at 100 MWE per hour initiated. Corporate management inforn.ed.

1445 Production superintendent arrives on site.

• 1530 NRC Resident Inspector notified.

1730 NRC Operation Center notified via ENS in a " courtesy" call.

1745 Accuracy of narrow range recorder verified by test. Corporate i representative arrives on site.

1 i

1820 Accuracy of startup data acquisition system verified. Licensee

! concludes RPS train A failed to function. RPS train A tripped. Alert declared. ENS notification made. Plant at 40% power decreasing.

6/2/86 0030 Turbine off line.

l 0922 All rods fully inserted. Alert terminated.

t l

l I

-