ML20085G413

Informs That Implementation of Mod for Automatic FW Isolation Following MSLB Delayed Until EOC 16 for ONS Unit 3,in Lieu of Commitment Made in 930819 Supplemental Response to IE Bulletin 80-04

ML20085G413
Person / Time
Site:	Oconee
Issue date:	06/14/1995
From:	Hampton J DUKE POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
IEB-80-04, IEB-80-4, NUDOCS 9506200198
Download: ML20085G413 (6)

Text

. - . - , . . . . .. - - - . _ _ .

e l- ..

- DukeIbwerCompany 1 W.H m os n' 0coneeNuclear5te Vicehesident P.O. S u I439 . (80)88534990flice Seneco, SC29679 (&O)8&%W4 Fat

.- DUKE POWER June 14,1995 -

- U.S. Nuclear Regulatory Commission Attention: Document ControlDesk Washington,DC 20555 -

I-

Subject:

Oconee Nuclear Site.. ,

Docket Nos. 50-269, -270, -287 Supplemental Response to IE Bulletin 80 04

. Delay ofNRC Commitment Item l

l Ce*lemen L By letter dated August 19,1993, Oconee provided a supplemental response 'to IE Bulletin 80-04. On' i page 4 of the Anachment to the August 19th letter, under " Schedule for. Corrective Action"it is stated that "the schedule for implementation of the modifications for the automatic feedwater isolation following a Main Steam Line Break" is "EOC 15" for Unit 3. The stated corrective action constitutes a commitment to the NRC.-

Duke has initiated % corrective action but has determined that due to the complexity of the modification, a delay in the completion date for Unit 3 is necessary. Duke expects to invier,est &

moddication during the Unit 1 EOC 16 and Unit 2 EOC 15 Refueling outages, respectively However,

. the moddication will be delayed until the EOC16 outage for Unit 3. Due to the low probabdity of this postulated accident, Duke maintains that this delay will not reduce the level of safety to the public.

Pa8es 3-4 of the anachment to the August .19,1993_ letter provide a conceptual designifor the proposed moddication. Some changes to this cuiwioal design are provided in the w"'+M to &

letter. The attached moddication description is subject to change as the design details are fm=134 but p . this document provides a general description of the scope, intent, and limitations of the moddication.1 After reviewt of the' Stand.idiud Technical Specifications and :other guidance for Technical Specification requirements, Duke has determined that a Technical Spacifie=+ ion is needed to address

. the equipment installed in this modification.

9506200198 950614 PDR ADOCK 05000269 Q PDR -

- ammarstra l

1

- If there are any further questions about this item, please contact David Nix at (803) 885-3634.

Very trulyyours,

- / -

. J. W. Hampton Attachment-t

.i I.

l l

I I

l

-l

l

?

l

Document Control Desk June 14,1995 Page 3 cc: Mr. S. D. Ebneter, Regional Administrator U. S. Nuclear Regulatory Commission, Region II Mr. L. A. Wiens, Project Manager Office ofNuclear Reactor Regulation Mr. P. E. Harmon . ,

Senior Resident Inspector Oconee Nuclear Site l

l

~

N 3- y. ,

I DUKE POWER COMPANY.

Oconee Nuclear Station 1

NSM 12873 "Feedwater Isolation during MSLB" Modification Description 1.0 Executive Summary When Duke Power re-analyzed the Reactor Building pressure transient after a Main Steam

. line break (MSLB), it was discovemd that the containment pressure response was more _

severe than originally analyzed >in the FSAR.' ;The revised analyses'. required prompt.

operator action to isolate feedwater flow to prevent the Reactor' Building (RB) pressure from exceeding its design limit of 59 psig. To prevent RB pressure from exceeding 59 psig, :.

this nuclear station modification (NSM) is designed to trip both main feedwater pumps (MFDWPs), isolate all normal feedwater from both steam generators (S/G)'and prevent the turbine driven emergency feedwater pump (TDEFWP) from starting. Additionally, to decrease operator burden, this NSM will initiate the same MFDW/EFW circuitry during a MSLB outside the reactor building.

2.0 Modi 6 cation Description Problem: Because operator response cannot be assured within the necessary timeframe, a MSLB inside the RB could pressurize the RB aluve its design

, pressure.

Solution: This NSM is designed to isolate feedwater to both' steam generators (S/G) during a steam line break. S/G outlet pressure and turbine header pressure are used as input signals to the circuitry that will isolate feedwater during a MSLB inside or outside the RB. When a MSLB is sensed, both MFDWPs.

will be tripped, both main feedwater (MFDW) control valves, MFDW block valves,~ startup feedwater.(SUFDW) control valves, and SUFDW block valves will be closed, and the TDEFWP will be prevented from auto start.

A 3.0 Design Basis ,

This circuitry is required to respond to a MSLB. The MSLB is a design basis event which is currently described in the FSAR. Therefore, the associated pressure transmitters, logic, and actuation circuitry will be QA condition 1.

.2 _ . - . .a . . .a _ . . .

.s .- j

.. 1 Only MS pressure is used to actuate feedwater isolation. A Reactor Protection System (RPS) reactor trip signal ~was not used because it might not always initiate during a "small" l MSLB outside the RB. The Main Steam pressure setpoint (~ 650 psig) will be_ chosen based u;nn safety analysis modeling; it will be within the range of pressures allowed based upon the MFDW/EFW flow rates, the valve stroke times, the circuitry response time, and

the instrument uncertainty. The setpoint will be chosen by balancing the attributes of a low .

setpoint which would minimize spurious actuations and a high setpoint which.would minimize the size of the break required to actuate the circuit.L "Small" breaks may not ,

actuate the circuitry because MS pressure may not decrease below the'setpoint, but the transient time for those "small" breaks should not require an automatic feedwater isolation;  ;

because operator action should limit RB pressure below 59 psig. Another mason to use a-setpoint in the 650 psig range is that its saturation temperature is significantly above the Low Temperature Overpressure Protection (LTOP) temperature of about 325 degrees F. 3 The circuitry will be enabled and disabled administratively rather than automatically. 'An Operating procedure will contain a step to enable / disable the circuit. The unit status which will require .the ' circuitry to be enabled will be- determined' by. safety analysis. l

' Overpressurizing the RB is not a concern prior to the turbine being on line because there is 'l no extraction steam to heat the feedwater and as a result not enough energy to exceed RB design pressure.

The circuitry will have an enable / disable control switch (es) and an initiate push-button (s).

The enable position will allow a low steam pressure to actuate the circuitry while the

~ disable position ~ will prevent the circuitry from actuating.' If the switch is set to enable, depressing the push-button will simulate a low steam pressure and actuate the circuitry. The push-button switch will be used primarily for testing but may also be used by operations to

, initiate the Main Steamline Break circuitry if detected prior to auto actuation. The circuitry will be energized from QA-1 power panelboards. '{

Because solenoid valves will be added to the air supply for the MFDW control valves and SUFDW control valves, each control valve will close whether its ICS control is set to manual or auto unless there is a loss ofinstrument air to the valve operator _ Ifinstrument air is lost, the MFDW and SUFDW control valves will fail as-is. This is acceptable for two reasons: 1) restoration of the control valves would be necessary to continue unit operation

~

at power and 2) the MFDW.and SUFDW block valves can be closed to isolate feedwater from the S/G. .In addition,~ inhi_ biting startup of the EFW pump turbine ensures that RB design pressure will not be exceeded even if the block valves are the only valves to close in .

, the MFDW flowpath.' Each block valve will only close ifits control switch is set to auto.-

Operations procedures must have the MFDW and SUFDW block valves in - AUTO -

whenever the MSLB circuitry is required to be in service.

After detecting a MSLB, all MFDW and SUFDW to both S/Gs will be isolated to minimize the RD pressure and to minimize RCS overcooling. Additionally, because feedwater'is

- . .- ~ . _- - _ . .

isolated from both S/Gs, both MFDWPs will be tripped. Tripping both MFDWPs will trip -

the reactor and trip the turbine. Tripping both MFDWPs will initiate start signals to all three EFWPs due to the low MFDWP discharge pressure, but a portion of this NSMs circuitry will prevent the TDEFWP auto start in order to reduce RB pressure. Operators -

will have to take manual control of EFW to isolate flow to the S/G with the steam line break.

-If this circuitry actuates and EFW must be controlled by either SUFDW control valve, this.

circuitry would not immediately allow the SUFDW control valve to open. Manual operator action would have to be taken to disable the MSLB circuitry. Two reasons make this an'.

- acceptable design: the SUFDW control valve to control EFW is a backup flow path, and operator action can enable controlling the SUFDW control valve.

A feedvater pipe rupture m'a y be sensed by this MSLB circuitry. If this circoit actuates -

during n feedwater pipe rupture, then the circuit will help mitigate the leakage by closing the valves and tripping the MFDWPs. If this circuit does not actuate during a feedwater pipe rupture, then this circuit will have no affect during the scenario.

4 The control 1:ircuitry will have a timer circuit in order to prevent a spurious signal from :

unnecessarily isolating feedwater. A 1-2 second timeframe will maintain RB pressure below 59 psig yet is long enough to prevent most erroneous signals.-

The circuitry will " seal-in" once it is initiated until the control switch is set to disable. -This is designed to prevent possible valve oscillations between closed and auto during pressure

, swings. This will prevent overcooling and undercooling events due to valve swings and corresponding feedwater flow swings. Also, the " seal-in" ' assures Operators know the valves are closed until manually reset. The annunciator is actuated so that the operators know the feedwater isolation circuitry has actuated. The computer points are intended to provide useful post trip information.

Waterhammer phenomena during feedwater isolation is currently being evaluated. To reduce the amplitude of the waterhammer, a time delay is being considered for the control logic between MFWPs tripping and control ~ valve closure initiation.

The MFDW equipment being controlled by the new MSLB circuitry is non safety-related and was never intended to be safety-related. Therefore, this equipment is not single failure proof. However, the associated pressure transmitters, logic, and control circuitry installed by this modification for mitigation of a MSLB will be safety-related (QA-1). Therefore, these components will be redundant and single failure proof.

A technical specification will be required to address the equipment added by this i I

modification.

I l

. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ __. -. , ~ . . , . , , . . . ,_. ,, , _ _ ,