RS-17-126, Quad Cities Nuclear Power Station, Units 1 & 2, Revision 14 to Updated Final Safety Analysis Report, Chapter 7, Instrumentation and Controls

From kanterella
(Redirected from ML17298A342)
Jump to navigation Jump to search
Quad Cities Nuclear Power Station, Units 1 & 2, Revision 14 to Updated Final Safety Analysis Report, Chapter 7, Instrumentation and Controls
ML17298A342
Person / Time
Site: Quad Cities  Constellation icon.png
Issue date: 10/19/2017
From:
Exelon Generation Co
To:
Office of Nuclear Reactor Regulation
Shared Package
ML17298A290 List:
References
RS-17-126
Download: ML17298A342 (217)


Text

QUAD CITIES -UFSAR7-i 7.0 INSTRUMENTATION AND CONTROLS TABLE OF CONTENTS Page7.0 INSTRUMENTATION AND CONTROLS................................................................. 7.1-

17.1 INTRODUCTION

............................................................................................ 7.1-17.1.1Identification of Systems.................................................... 7.1-17.1.1.1Protective Systems....................................... 7.1-27.1.1.2Safe Shutdown............................................. 7.1-27.1.1.3Display Instrumentation............................. 7.1-27.1.1.4Core and Vessel I nstrumentation................ 7.1-37.1.1.5Other Instrum entation................................ 7.1-37.1.2Identification of Safe ty Criteria.......................................... 7.1-37.1.2.1Instrumentation Setpoints........................... 7.1-37.1.2.2Single Failure Criteria...............................

7.1-3a7.1.2.3Instrument Line Design............................... 7.1-47.1.2.4Qualification................................................. 7.1-47.1.3Other Control and I nstrumentation................................... 7.1-47.2REACTORPROTECTION(TRI P) SYSTEM............................................. 7.2-17.2.1Design Bases....................................................................... 7.2-17.2.2System Description............................................................. 7.2-17.2.2.1General......................................................... 7.2-17.2.2.2Power Sources.............................................. 7.2-27.2.2.3Instrumentation........................................... 7.2-27.2.2.4Logic.............................................................. 7.2-6 7.2.2.5Initiating Signals and Circuits...................7.2-107.2.2.6Scram Bypasses...........................................7.2-157.2.2.7Redundancy, Diversity, and Separation.....7.2-21 7.2.2.8Testability....................................................7.2-227.2.2.9Environmental Considerations...................7.2-277.2.2.10Operational Co nsiderations........................7.2-277.2.2.11Anticipated Trans ient Without Scram.......7.2-317.2.3Analysis of Design Requ irements Conf ormance...............7.2-317.2.3.1Single FailureCriterion..............................7.2-347.2.3.2Quality of Componen ts and Modules.........7.2-407.2.3.3Channel Integrity........................................7.2-41 7.2.3.4Channel Separation....................................7.2-427.2.3.5Control and Protection System Interaction...................................................7.2-447.2.3.6Capability for Test and Calibration............7.

2-467.2.3.7Establishment of Tr ip Setpoints.................7.

2-497.2.3.8Access to Setpoint Adjustments, Calibration, and Test Points.......................7.

2-517.2.3.9Identification of Pr otection Systems...........7.2-527.2.3.10System Repair.............................................7.2-527.2.4References...........................................................................7.2-55 QUAD CITIES -UFSAR TABLE OF CONTENTS (Continued)

Revision 11, October 20117-ii Page7.3ENGINEERED SAFETY FEATURE SYSTEMS INSTRUMENTATION AND CONTROL................................................... 7.3-1 7.3.1Emergency Core Cooling Systems Instrumentation and Control................................................................................. 7.3-1 7.3.1.1Core Spray System Instrumentation and Control.......................................................... 7.3-17.3.1.2RHR System LPCI Mode Instrumentation and Controls................................................. 7.3-77.3.1.3High Pressure Coolant Injection System Instrumentation and Control......................7.3-137.3.1.4Automatic Depressurization System Instrumentation and Controls....................7.3-197.3.2Primary Containment Is olation Systems..........................7.3-257.3.2.1Design Basis................................................7.3-257.3.2.2Isolation Logi c Description.........................7.3-257.3.2.3Primary Containment Isolation System Instrumentation..........................................7.3-337.3.2.4Design Eval uation.......................................7.

3-367.3.2.5InspectionandTesting ...............................7.3-387.3.2.6Conformanceto IEEE-279..........................7.3-387.3.3Secondary Containment Isolation System........................7.3-437.3.4References...........................................................................7.3-447.4SAFE SHUTDOWN................................................................................... 7.4-17.4.1Containment Cooling Mode of the Residual Heat Removal System.................................................................. 7.4-17.4.2Shutdown Outside the Co ntrol Room................................. 7.4-17.5DISPLAY INSTRUMENTATION ............................................................. 7.5-17.5.1Post-Accident Monitors....................................................... 7.5-17.5.1.1Description................................................... 7.5-17.5.1.2Analysis........................................................ 7.5-27.5.2Process Computer................................................................ 7.5-47.5.2.1Description................................................... 7.5-47.5.2.2Operator Functions......................................

7.5-57.5.3Safety Parameter Disp lay System...................................... 7.5-67.5.3.1Description................................................... 7.5-77.5.3.2Analysis........................................................ 7.5-87.5.4Detailed Control Room Design Review................................

7.5-97.5.5References...........................................................................7.5-10 QUAD CITIES -UFSAR TABLE OF CONTENTS (Continued)7-iiiRevision11, October 20117.6CORE AND VESSEL INSTRUMENTATION........................................... 7.6-17.6.1NuclearInstrumentation ...................................................

7.6-17.6.1.1DesignBases ............................................... 7.6-17.6.1.2GeneralDescription .................................... 7.6-17.6.1.3SourceRangeMonitoringSubsystem

......... 7.6-27.6.1.4IntermediateRangeMonitoring Subsystem.................................................... 7.6-57.6.1.5PowerRangeMonitoring Subsystem .......... 7.6-77.6.2Reactor Vessel Instrumentation .....................................7.

6-15e7.6.2.1Design Bases and Design Features..........7.6-15e7.6.2.2Description..................................................7.6-167.6.2.3Design Eval uation.......................................7.

6-197.6.2.4Surveillance and Testing............................7.

6-207.6.2.5Analog Trip Instrumentation......................7.6-217.6.3References...........................................................................7.6-227.7OTHER INSTRUMENTATION................................................................. 7.7-17.7.1Reactor Control Rod Control Systems................................ 7.7-17.7.1.1 Design Bases................................................ 7.7-17.7.1.2ControlRodAdjustmentControl (Reactor Manual Control Syst em) ............................. 7.7-27.7.1.3DesignEvalu ation........................................ 7.7-77.7.1.4Inspectionand Testing................................. 7.7-87.7.2Rod Worth Minimizer.......................................................... 7.7-87.7.2.1DesignBasis ................................................ 7.7-87.7.2.2DescriptionandD efinitions

........................ 7.7-87.7.2.3DesignEvaluation .....................................7.7-157.7.2.4SurveillanceandTesting ...........................7.

7-157.7.3 Load Control Design...........................................................7.7-167.7.3.1RecirculationFlowControlSystem............7.7-177.7.3.2EconomicGenerationControlSystem -

Abandoned....................................................7.7-177.7.3.3FailureModeandEffectsAnalyses............7.

7-187.7.3.4DesignEval uation.......................................7.

7-197.7.3.5Other Reactivity Control Systems..............7.7-197.7.4PressureRegulatorandTurbine-GeneratorControls ......7.7-207.7.4.1DesignBasis................................................7.7-207.7.4.2SystemDescription.....................................7.7-207.7.4.3DesignEval uation.......................................7.

7-217.7.5FeedwaterLevel ControlSystem ......................................7.

7-227.7.5.1DesignBasis................................................7.7-227.7.5.2System Description.....................................7.7-227.7.5.3DesignEval uation.......................................7.

7-247.7.6MainCondenser,Condensate,andCondensate Deminera lizer.....................................................................7.7-257.7.6.1DesignBases...............................................7.7-257.7.6.2System Description.....................................7.7-257.7.6.3DesignEval uation.......................................7.

7-25 QUAD CITIES -UFSAR TABLE OF CONTENTS (Continued)7-iv Page7.8ANTICIPATED TRANSIENT WITHOUT SCRAM MITIGATION SYSTEM..................................................................................................... 7.8-1 7.8.1Introduction......................................................................... 7.8-17.8.2Design Requirements.......................................................... 7.8-17.8.3Mitigation System Description

........................................... 7.8-27.8.3.1Recirculation Pu mp Trip.............................. 7.8-37.8.3.2 Alternate Rod Insertion .............................. 7.8-37.8.3.3Alternate Rod Insertion Valves................... 7.8-47.8.4Design Evaluation............................................................... 7.8-47.8.5Refere nces............................................................................ 7.8-6 QUAD CITIES -UFSAR TABLE OF CONTENTS (Continued)7-vRevision 9, October 2007 7.0 INSTRUMENTATION AND CONTROLS LIST OF TABLES Table7.2-1Analytical Limits for Reactor Protection Setpoints 7.3-1Analytical Limits for Group Isolation Signals 7.4-1Reactor Vessel Pressure and Level Indicators Available Outside the Control Room 7.6-1OPRM System Trips7.7-1EGC Console Top Plate Functions -Abandoned Equipment 7.7-2EGC Status Indicators (Annunciators) -Abandoned Equipment QUAD CITIES -UFSAR TABLE OF CONTENTS (Continued)7-viRevision 14, October 2017 7.0 INSTRUMENTATION AND CONTROLS LIST OF FIGURES Figure7.2-1Reactor Protection System Power Supply7.2-2Use of Control and Instrumentation Definitions 7.2-3Typical Logic Arrangement 7.2-4Typical Logic Arrangement 7.2-5Typical Logic Arrangement7.3-1Block Diagram: Primary Containment Isolation 7.6-1Nuclear Instrumentation System Ranges and Overlaps7.6-2Block Diagram Nuclear Instrumentation System 7.6-3SRM -Detector and Source Locations 7.6-4IRM -Detector Locations 7.6-5IRM -Response to Rod Withdrawal Error 7.6-6IRM -Power Distribution During Rod Withdrawal Error 7.6-7LPRM -Detector Locations 7.6-8LPRM -Local Detector Locations 7.6-9LPRM -Quadrant Symmetry 7.6-10APRM -LPRM Assignments, Channels 1, 2, and 4 7.6-11APRM -LPRM Assignments, Channels 3, 5, and 6 7.6-12Illustrative APRM Scram and Rod Block Trip vs. Recirculation Flow 7.6-13APRM Response During Flow-Induced Power Level Maneuvering 7.6-14APRM Response During Control Rod -Induced Power Level Maneuvering 7.6-15RBM -LPRM Input Assignment 7.6-16Deleted7.6-17Block Diagram -OPRM Subsystem7.7-1Conditions which Prevent Control Rod Withdrawal7.7-2Deleted 7.7-2ABlock Diagram -Rod Worth Minimizer 7.7-3Deleted 7.7-3ADeleted 7.7-3BReactor Pressure, Turbine Speed, and Recirculation Flow Control Systems 7.7-4Deleted 7.7-5Deleted 7.7-6Deleted 7.7-7Deleted QUAD CITIES -UFSAR TABLE OF CONTENTS (Continued)7-viiRevision 13, October 2015 7.0 INSTRUMENTATION AND CONTROLS DRAWINGS CITED IN THIS CHAPTER**The listed drawings are included as "General References" only; i.e., refer to the drawings to obtain additional detail or to obtain background information. These drawings are not part of the UFSAR. They are controlled by the Controlled Documents Program.DRAWING*SUBJECTM-35Diagram of Nuclear Boiler & Reactor Recirculating Piping M-41Diagram of Control Rod Drive Hydraulic Piping M-77Diagram of Nuclear Boiler & Reactor Recirculating Piping QUAD CITIES - UFSAR Revision 7, January 2003 7.1-1 7.0 INSTRUMENTATION AND CONTROLS

This chapter presents various plant instrumentation and control systems including

functions, design bases, system descriptions, design evaluations, and tests and inspections.

The information provided in this chapter emphasizes instruments and associated

equipment which constitute reactor protection and regulation systems. Particular attention

is given to the instrumentation aspects of process systems, with the mechanical and nuclear

design bases presented in the chapter/section which addresses the process system. Chapter

7 includes a discussion of the instrumentation and controls for systems of major safety

significance and those that provide reactor and turbine control. Discussions of

instrumentation and controls for other systems are contained within the sections that

address those systems.

7.1 INTRODUCTION

The equipment and evaluations presented in this chapter are applicable to either unit.

Instrumentation and controls are provided to perform protective and regulating functions.

Protective systems, consisting of the reactor protective circuitry and the instrumentation

and controls for engineered safety features (ESFs) , normally perform the most important of

the instrumentation and control safety functions.

[7.1-1]

The regulating instrumentation and controls provide the ability to regulate the unit from

shutdown to full power and to monitor and maintain key unit variables, such as reactor

power, flow, pressure, level, temperature, and radioactivity levels within predetermined

limits both at steady-state and during normal unit transients.

The inputs to the protective and regulating controls are provided by a diversity of

instruments. The following sections in this chapter provide descriptions of instrumentation

and major components, evaluations of the instrumentation input adequacy, and analyses

from both functional and reliability viewpoints.

7.1.1 Identification of Systems

Section 3.2 discusses the identification of safety-related instrumentation and control

systems and equipment. The station's work control system data base also contains information on classifications of components.

[7.1-2]

The reactor protection and ESF systems supplied by GE as the nuclear steam supply

system (NSSS) supplier are:

[7.1-3]

A. Reactor protection system,

B. Primary containment isolation system,

C. Emergency core cooling system, QUAD CITIES - UFSAR Revision 6, October 2001 7.1-2 7.1.1.1 Protective Systems

Protective systems include electrical and mechanical devices and circuitry required to initiate shutdown of the reactor and mitigate the consequences of an accident when

required. These include:

A. The reactor protection system (RPS) which acts to trip the reactor when parameters exceed preset limits (RPS is described in Section 7.2);

B. The anticipated transient without scram (ATWS) system which trips the recirculation pumps and provides an alternate method to scram the reactor in

the unlikely event that the RPS fails to do so (ATWS mitigation is described in

Section 7.8); and

Engineered safety feature (ESF) instrumentation and controls for emergency core cooling

and containment isolation functions which are addressed in Section 7.3 (other ESF systems

are discussed in Section 6.0):

[7.1-4]

1. Emergency core cooling systems:
a. Core spray,
b. Low pressure coolant injection (LPCI),
c. High pressure coolant injection (HPCI), and
d. Automatic depressurization system (ADS).
2. Containment isolation systems:
a. Primary containment isolation system (PCIS), and
b. Secondary containment isolation.

7.1.1.2 Safe Shutdown

Section 7.4 includes a discussion of reactor shutdown from outside the control room.

7.1.1.3 Display Instrumentation

Display instrumentation provides information used by the operator for normal operation

and safe shutdown of the unit, including monitoring of post accident conditions.

Compliance with Regulatory Guide 1.97, Rev. 02, the safety parameters display system QUAD CITIES - UFSAR 7.1-3 Revision 8, October 2005 (SPDS), and the process computer are discussed in Section 7.5. A summary of the detailed control room design review (DCRDR) is also provided.

7.1.1.4 Core and Vessel Instrumentation

Section 7.6 describes additional instrumentation which provide both safety and non-safety

functions, and which includes nuclear instrumentation and reactor vessel instrumentation.

7.1.1.5 Other Instrumentation

Reactor and turbine generator instrumentation and controls not essential for the safety of the plant are discussed in Section 7.7.

7.1.2 Identification of Safety Criteria

The design bases for the instrumentation and control systems include the safety criteria

pertinent to each of the systems described. The design basis for each of the systems is presented in the respective section which discusses the system. The technical basis for the

various protective functions is provided with the description of the protective system. A

general discussion of Regulatory Guide compliance is provided in Section 1.8. Specific topics

relevant to more than a single system are addressed in the following sections.

[7.1-5]

7.1.2.1 Instrumentation Setpoints

In the selection of the appropriate safety system setpoints, instrument error and accuracy are

considered

[7.1-6]

The Technical Specification allowable values and the associated instrument setpoints have

been established consistent with the methods described in Exelon's Instrument Setpoint Methodology (Nuclear Engineering Standard NES-EIC-20.04, "Analysis of Instrument

Channel Setpoint Error and Instrument Loop Accuracy") or NEDC-31336P-A, "General

Electric Instrument Setpoint Methodology," dated September 1996 (for Nuclear

Instrumentation System Functions only).

The allowable values associated with reactor vessel water level Functions in the Technical

Specifications are referenced with respect to instrument zero. The top of active fuel is 360

inches above vessel zero and instrument zero is 503 inches above vessel zero. The allowable values associated with suppression chamber water level Functions in the Technical

Specifications are referenced to the bottom of the chamber.

QUAD CITIES - UFSAR 7.1-3a Revision 8, October 2005 7.1.2.2 Single Failure Criteria

The compliance of the reactor protection and emergency core cooling systems with, and the

justification for all exceptions to IEEE 279-1968, Proposed Criteria for Nuclear Power Plant

Protection Systems, are contained in GE Topical Report NEDO-10139, Compliance of Protection Systems to Industry Criteria: General Electric BWR Nuclear Steam Supply System.

Compliance of the protection systems is presented in the sections providing the system details.

These systems typically employ one-out-of-two-twice logic to allow the systems to

accommodate single failures without jeopardizing functionality. The GE topical report is a generic report for an entire product line and these statements should be used concurrently with design requirements described in other sections of the UFSAR (such as Section 3.5 for Missile Protection, 3.6 for Pipe Break, etc.) that represent Quad Cities specific design requirements.

[7.1-7]

QUAD CITIES - UFSAR 7.1-4 7.1.2.3 Instrument Line Design

The normal design practice for static instrument piping is to provide high point vents and

low point drains.

[7.1-8]

Instrument and cable separation are described in Section 8.3.1.7

7.1.2.4 Qualification

The qualification of instrumentation and controls is described in Sections 3.10 and 3.11.

Additional discussion of display instrumentation qualification and separation for

Regulatory Guide 1.97 Category 1 variables is in Section 7.5.

[7.1-9]

7.1.3 Other Control and Instrumentation

Controls and instrumentation for the following auxiliary and emergency systems are

described in the sections that describe the systems:

[7.1-10]

System Section Reactor building heating and ventilation system 9.4.7 Reactor water cleanup system 5.4.8 Reactor core isolation cooling system 5.4.6 Fire protection system 9.5.1 Station service water system 9.2.2 Demineralized water makeup system 9.2.4 Service and instrument air systems 9.3.1 Communication systems 9.5.2 Spent fuel pool cooling and cleanup system 9.1.3 Fuel handling system 9.1.4 High radiation sampling system 9.3.2

QUAD CITIES - UF SAR Revision 9, October 2007 7.2-1 7.2 REACTOR PROTECTION (TRIP) SYSTEM

The reactor protection system (RPS) monitors reactor operation and initiates protective

action in the event of a potentially unsafe condition that might cause reactor damage or

subject plant personnel to a potentially hazardous environment. Monitoring is performed

by two separately powered RPS trip systems, both of whose outputs are needed to initiate protective action. Outputs from these systems initiate reactor scram (simultaneous rapid

insertion of control rods into the reactor core).

[7.2-1]

Topics within this section include how RPS functions relate to IEEE-279-1968, Proposed IEEE Criteria for Nuclear Power Plant Protection Systems , as summarized from GE Topical Report NEDO-10139. The applicable IEEE-279-1968 paragraphs have been noted where

the discussion concerns this standard, although conformance was not required. For more

detailed information refer to the topical report.

7.2.1 Design Bases

The reactor protection system is designed to:

[7.2-2]

A. Prevent, in conjunction with the containment and containment isolation system, the release of radioactive materials in excess of the limits of 10 CFR 100 (or 10 CFR 50.67 as applicable) as a consequence of any of the design basis accidents (Chapter 15);

B. Prevent fuel damage following any single equipment malfunction or single operator error;

C. Function independently of other plant controls and instrumentation;

D. Function safely following any single component malfunction; and

E. Meet the requirements of IEEE-279, "Standard for Nuclear Power Plant Protection Systems," Sept. 13, 1966.

In order to meet its design requirements, the reactor protection system, under various

conditions, initiates a reactor scram.

7.2.2 System Description

7.2.2.1 General

The RPS is classified as a safety-related system. It includes the motor-generator (M-G)

power supplies with associated control and indicating equipment, certain sensors, relays, bypass circuitry, and switches that cause rapid insertion of control rods (scram) to shut

down the reactor. The process computer system and annunciators are not part of the RPS.

Scram signals received from the neutron monitoring system and the analog trip cabinets

are discussed in Section 7.6.

[7.2-3]

QUAD CITIES - UFSAR Revision 6, October 2001 7.2-2 7.2.2.2 Power Sources

A simplified diagram of the RPS power distribution and sources is shown on Figure 7.2-1.

The reactor protection system consists of two independent trip systems powered by

independent electrical buses.

[7.2-4]

Power to each of the two reactor protection trip system buses (A and B) is supplied by its

own high-inertia (flywheel-equipped) ac M-G set (A and B). The station 125-V batteries

supply dc power to the backup scram valve solenoids.

[7.2-5]

The RPS bus breakers are equipped with mechanical interlocks to prevent both an M-G set

and the reserve power source from simultaneously supplying power to a RPS bus. The

normal feed for RPS bus A (M-G set A) is MCC 18-2(28-2). The normal feed for RPS bus B (M-G set B) is MCC 19-2(29-2). Either bus may be fed from the reserve feed from MCC 15-

2(25-2).

A key interlock system, consisting of two locking devices on the reserve power supply

breakers that require the same key, prevents reserve power from supplying more than one

RPS bus at a time. It prevents cross-connecting the independent buses and overloading the

reserve power instrument transformer.

During a power loss to the M-G set, the high-inertia flywheel is designed to maintain

generator output within 5% of rated values for at least one second to keep the RPS bus

energized. The non-Class 1E RPS M-G sets are provided with relaying to trip on

undervoltage and underfrequency conditions.

[7.2-6]

In addition, two Class 1E electrical protection assemblies (EPAs) are in series between each RPS power supply and its RPS bus breaker (see Figure 7.2-1). The EPAs protect the Class

1E components powered by the RPS buses from abnormal voltage and frequency conditions

resulting from failures of the non-Class 1E power supplies (RPS M-G sets or reserve power

supply). Each EPA includes a breaker and associated monitoring module consisting of

overvoltage, undervoltage, and underfrequency relays which trip the EPA breaker.

[7.2-7]

7.2.2.3 Instrumentation

A. Sensors

The reactor protection system receives the following inputs. Table 7.2-1 contains the

analytical limits utilized in determining the RPS setpoints.

[7.2-8]

1. The purpose of the neutron monitoring system scram trip as it applies to IEEE-279-1968 General Functional Requirements (paragraph 4.1) is to

protect the fuel against high heat generation rates.

Those portions of the neutron monitoring systems that provide a gross power protective function are:

A. Average power range monitor (APRM) with either fixed scram or flow reference scram QUAD CITIES - UFSAR Revision 12, October 2013 7.2-3 B. Intermediate range monitor (IRM)

The portion of the neutron monitoring system that provides a power oscillation protective function is the Oscillation Power Range Monitor (OPRM).

Eight channels of IRM with retractable detectors, six channels of APRM, and four channels of OPRM are provided. The APRM and OPRM receive

input signals from local power range monitor (LPRM) detector assemblies

containing detectors located at fixed geometric coordinates and at four

vertical elevations within the reactor core.

The neutron monitoring system instrumentation is described in Section 7.6.

2. The purpose of the reactor high pressure scram trip is to limit the positive pressure effect on reactor power. This reactor scram trip is established to

reduce the heat generation within the reactor whenever the high-pressure

setpoint is reached. In this way, the high pressure scram trip meets the

IEEE-279-1968 General Functional Requirements (paragraph 4.1).

[7.2-9]

The reactor high pressure scram works in conjunction with the pressure relief system in preventing reactor pressure from exceeding the pressure

safety limit. This high pressure scram setting also protects the core from

exceeding the thermal hydraulic safety limit as a result of pressure

increases for some events that occur when the reactor is operating at less

than rated power and flow. The reactor high pressure scram also provides

backup protection to the high neutron flux scram.

Two locally mounted pressure transmitters monitor the pressure and are arranged so that each pair provides input into the A & B trip systems.

The transmitter signal serves as an input to an analog trip unit for each

channel, the contacts of which are used in the RPS trip logic. The analog

trip unit supplies a signal to the analog channel trip relays. The logic for

these contacts is one-out-of-two-twice.

[7.2-10]

When the signal from the transmitter exceeds a preset value, the analog

trip unit monitoring this signal trips to send a reactor vessel high pressure trip signal to the RPS. Additional information on reactor vessel

instrumentation can be found in section 7.6.

3. The purpose of the reactor vessel low water scram trip as it applies to the IEEE-279-1968 General Functional Requirements (paragraph 4.1), is to

protect the reactor core by reducing fission heat generation in the core.

[7.2-11]

To meet this requirement, the reactor vessel low water level is monitored by four differential pressure transmitters which sense the difference between the pressure due to a constant reference column of water and the

pressure due to the actual water level in the vessel.

QUAD CITIES - UFSAR Revision 12, October 2013 7.2-4 The transmitter signal serves as an input to an analog trip unit for each channel, the contacts of which are used in the RPS trip logic. The analog

trip unit supplies a signal to the analog trip relays. The logic for these

contacts is one-out-of-two-twice.

When the signal from the transmitter deviates from a preset value, the analog trip unit monitoring this signal trips to send a reactor vessel low

water level signal to the respective RPS trip channel. Additional analog trip units and trip relays are provided for PCIS and HPCI. Additional

information on reactor vessel instrumentation can be found in Section 7.6.

[7.2-12] 4. The purpose of the turbine stop valve closure scram trip, as summarized by NEDO-10139 for the IEEE-279-1968 General Functional Requirements (paragraph 4.1), is to protect the reactor whenever it is sensed that its link to the heat sink is in the process of being removed.

[7.2-13]

To meet these requirements, the valve stem position of each turbine stop valve is monitored by limit switches. The limit switch allowable value is

less than or equal to 9.7% from the full-open position. In this way the trip

channel signals to the reactor protection system anticipate imminent

closure of the stop valves. Each RPS trip logic receives inputs from two

stop valves. The logic arrangement is established to enhance frequent

testing of these valves without causing a trip of one RPS trip system for

each valve test. The logic arrangement to produce a reactor scram is

three-out-of-four stop valve closures rather than one-out-of-two twice.

5. The purpose of the turbine control valve fast-closure scram trip, as summarized by NEDO-10139 for the IEEE-279-1968 General Functional

Requirements (paragraph 4.1), is to protect the reactor whenever it senses

that its link to the heat sink is in the process of being removed.

[7.2-14]

To meet the general functional requirements, the turbine control valve fast closure is monitored by pressure switches connected between each

fast-closure solenoid valve and its associated control valve disk dump port.

The electrohydraulic control (EHC) system compares generator stator current to the high pressure turbine exhaust (crossaround) pressure

and operates these valves upon a mismatch indicative of a turbine

generator load rejection (see Section 10.4). These pressure switches

on each fast-acting solenoid provide signals to both RPS trip systems.

The logic is a one-out-of-two-twice arrangement so that operation of

any solenoid causes a single system trip, and the operation of one or

more solenoids in each trip system initiates a scram.

[7.2-15]

6. The purpose of the main steam line isolation valve closure scram trip, as it applies to IEEE-279-1968 General Functional Requirements (paragraph 4.1), is to protect the reactor whenever its lines to the heat

sink (turbine or condenser) is in the process of being removed.

[7.2-16]

QUAD CITIES - UFSAR Revision 12, January 2013 7.2-5 The valve stem position of each of the eight main steam line isolation valves is monitored by limit switches. The limit switch allowable value is

less than or equal to 9.8% from the full open position.

Each RPS trip logic receives input from both valves in two main steam lines. The logic arrangement is established to enhance frequent testing of

these valves without causing a trip of one RPS trip system for each valve

test. The logic arrangement to produce reactor scram is three-out-of-four

steam lines isolated rather than a one-out-of-two twice arrangement.

7. The purpose of the scram discharge volume high water level scram trip, as it applies to IEEE-279-1968 General Functional Requirements (paragraph 4.1), is to assure that adequate volume remains to

accommodate the water discharged from the withdrawn control rod drives

in the event that a reactor scram occurs.

[7.2-17]

Scram discharge volume (SDV) high water level inputs to the RPS are from two float-type and two differential pressure-type level sensors on each of the SDVs. They are arranged such that a float-type and a

differential pressure-type level sensor for each channel are connected

to each SDV. An actuation of any level switch causes a channel trip; an

actuation of two level switches, one in each trip system, causes a

scram. A scram is initiated when sufficient capacity remains in the

SDV to accommodate the displacement of water for one scram.

[7.2-18]

8. The purpose of the primary containment (drywell) high pressure scram trip, as it applies to the IEEE-279-1968 General Functional Requirements (paragraph 4.1), is to detect an increase in the primary containment gauge pressure and produce protective action.

[7.2-19]

Primary containment pressure is monitored by four non-indicating pressure switches which are mounted on instrument racks outside the

drywell in the reactor building. Each switch provides an input to one trip

channel. Pipes that terminate in the secondary containment (reactor

building) connect the switches with the drywell interior. The switches are

grouped in pairs, physically separated, and electrically connected to the

RPS so that no single event will prevent a scram due to drywell high

pressure.

[7.2-20]

9. Deleted.
10. The turbine-generator condenser vacuum is monitored by four nonindicating pressure switches which are mounted on instrument

racks in the turbine building. Cables are routed from each switch

to the control room. Each switch provides an input to QUAD CITIES - UFSAR Revision 7, January 2003 7.2-6 one of the trip channels. The physical location of each switch is such that no single failure can prevent a scram due to a low

vacuum signal from the turbine-generator condenser.

11. Deleted.

[7.2-22]

B. Relays

Sensor trip channel and trip logic relays are fast-response, high-reliability relays. Power relays for interrupting the scram pilot valve solenoids are type

CR105 magnetic contactors, made by GE. The contactor has three main poles

which are operated directly by the armature. Several auxiliary poles are also provided. The auxiliary poles are used for nonessential functions. Two main

poles are used to break power to the scram solenoids and the third main pole is

used to seal-in the scram. The seal-in contact operates at the same time as the

scram contacts which operate the scram solenoids, since both are directly

operated by one mechanical unit (armature). Therefore, seal-in occurs

simultaneously with scram actuation. All RPS relays are selected so that the

continuous load will not exceed 50% of their continuous duty ratings.

Component electrical characteristics are selected so that the system response

time, from the opening of a sensor contact up to and including the opening of the

trip actuator contacts is less than 50 milliseconds. The time from the opening of

the trip actuator contacts until the control rods have inserted by 10% of their full

stroke is no more than 700 milliseconds.

[7.2-23]

7.2.2.4 Logic

The complexity of the control and instrumentation systems necessitates the use of the

definitions below. These definitions are most appropriate to safety-related systems. Figure

7.2-2 illustrates the use of the defined terms.

[7.2-24]

A. Trip System

A trip system is an interconnected arrangement of components making use of instrument channel outputs, trip logics, and trip actuators to accomplish a

trip function when appropriate logic is satisfied.

B. Trip

A trip is the change of state of a bistable device from one state to another.

A trip is generated by a trip channel, trip logic, or trip system, and

represents recognition of an abnormal condition.

C. Trip Channel

A trip channel is an arrangement of components required to originate a single signal. The channel includes the sensor and wiring up to the point where the QUAD CITIES - UFSAR 7.2-7 Revision 8, October 2005 trip signal is generated. A channel loses its identity where channel trip signals are combined.

D. Trip Logic

A trip logic is an arrangement of components designed to recognize specific combinations of signals from trip channels. A trip logic generates a trip signal

by actuating a trip actuator.

E. Trip Actuator

A trip actuator is the mechanism that carries out the final action of a trip logic.

F. Trip Actuator Logic

A trip actuator logic is an arrangement of components designed to recognize specific combinations of signals from trip logics. This term is needed to clearly

define portions of a complex trip system having more than one trip logic.

Because trip actuators are the mechanism by which trip logics generate trip

signals, the use of the term trip actuator logic is appropriate. When tripped, a

trip actuator logic carries out the function of the trip system.

A typical logic arrangement of the system is illustrated in Figures 7.2-3 through 7.2-5. The

reactor protection system is arranged as two separately powered trip systems. Each trip system has three trip logics, two of which are used to produce automatic trip signals. The

remaining trip logic is used for a manual trip signal. Each of the two trip logics used for

automatic trip signals receives input signals from at least one trip channel for each monitored

variable. Thus, at least four independent trip channels exist for each monitored variable.

The trip actuators associated with one trip logic provide inputs into each of the trip actuator

logics for the associated trip system. Thus, either of the two automatic trip logics associated

with one trip system can produce a trip system trip. The logic is a one-out-of-two arrangement. To produce a scram, the trip actuator logics of both trip systems must be

tripped. The overall logic of the RPS is therefore, one-out-of-two-twice, since at least one of the two automatic trip logics in each of the two trip systems must actuate in order to cause an

automatic RPS trip (scram).

The two RPS trip systems are called trip system A and trip system B. The automatic trip

logics of trip system A are A1 and A2; the manual trip logic of trip system A is A3. Similarly, the trip logics for trip system B are B1, B2, and B3. The trip actuators associated with any

particular trip logic are identified by the trip logic identity (such as trip actuators B2). The

trip actuator logics associated with a trip system are identified with the trip system identity (such as trip actuator logic A). Trip channels are identified by the name of the monitored

variable and the trip logic identity with which the channel is associated (such as reactor vessel

high pressure trip channel B1).

During operation, all sensor and trip contacts essential to safety are closed; trip channels, trip

logics, and trip actuators are normally energized.

QUAD CITIES - UFSAR 7.2-8 Revision 8, October 2005 Each control rod has two scram valves, and either two individual scram solenoid pilot valves (SSPVs) or one SSPV with two solenoid coils, arranged functionally as shown in Figure 7.2-3.

Each SSPV is solenoid operated, with both SSPV solenoids normally energized. The SSPVs control the air supply to both scram valves for the associated control rod. With either SSPV solenoid energized, air pressure holds the scram valves closed. The scram valves control the supply and discharge paths for control rod drive (CRD) water (refer to Section 4.6 for discussion of the CRD system).

One of the SSPV solenoids for each control rod is controlled by the reactor protection system (RPS) logic Channel A, the other valve by RPS logic Channel B. There are two DC solenoid-operated backup scram valves that provide a second means of controlling the air supply to the scram valves for all control rods. The DC solenoid for each backup scram valve is normally de-energized. The backup scram valves are energized to initiate a scram when both trip system A and trip system B are tripped.

[7.2-25] The functional arrangement of sensors and trip channels that make up a single trip logic is

shown in Figure 7.2-4. Whenever a trip channel sensor contact opens, its auxiliary relay de-

energizes, causing contacts in the trip logic to open. The opening of contacts in the trip logic

de-energizes its trip actuators. When de-energized, the trip actuators open contacts in all the

trip actuator logics for that trip system. This action results in de-energizing the scram pilot

valve solenoids associated with that trip system (one scram pilot valve solenoid for each

control rod). Unless the other scram pilot valve solenoid for each rod is de-energized, the rods

are not scrammed. If a trip then occurs in any of the trip logics of the other trip system, the remaining scram pilot valve solenoid for each rod is de-energized, blocking the air supply and

venting the air pressure from the scram valves. The scram valves then reposition allowing

accumulator water to act on the CRD piston. Thus, all control rods are scrammed. The water

displaced by the movement of each rod piston is vented into a scram discharge volume (SDV).

Figure 7.2-3 shows that when the solenoid for either backup scram valve is energized, the backup scram valve vents the air supply for the scram valves; this action initiates insertion of

every control rod regardless of the action of the scram pilot valves.

A scram can also be manually initiated. There are two scram buttons, one for trip logic A3

and one for trip logic B3. Depressing the scram button on trip logic A3 de-energizes trip

actuator A3 and opens corresponding contacts in trip actuator logics A. Only trip system A

will trip. To effect a manual scram, the buttons for both trip logic A3 and trip logic B3 must

be depressed. By operating the manual scram button for one trip logic at a time, followed by a

reset of that trip logic before actuating the other manual trip logic, each trip system can be

tested for manual scram capability.

The trip system requires manual reset by the operator; however, in the event of concurrent

trips of both trip systems A and B, manual reset is automatically inhibited for a minimum

time delay of 10 seconds. The time delay circuit prevents an incident such as has been

experienced at another BWR plant where during intermediate range monitoring (IRM)

calibration, a full scram signal was initiated and then inhibited by actuation of the scram

reset switch prior to the insertion of all control rods.

[7.2-26]

To restore the RPS to normal operation following any single trip system trip or scram, the trip

actuators must be manually reset. Reset is possible only if the conditions that caused the trip

or scram have been cleared and is accomplished by operating switches in the main control

room. To reset the air dump system, the scram must be reset and the SDV high level bypass

switch must be placed in the bypass position. The SDV is addressed in Section 4.6.

QUAD CITIES - UFSAR Revision 7, January 2003 7.2-9 The IEEE-279-1968 requirement for Completion of Protective Action Once It Is Initiated (paragraph 4.16) is addressed by the RPS in the following ways:

[7.2-27]

For the reactor protection system trip logic, actuators, and trip actuator logic, the interface of the RPS trip logic and the trip actuators assures that this design requirement is accomplished. The trip actuator is normally energized and is sealed-in by one of the power contacts to the trip logic string. Once the trip logic string has been open-circuited as a result of a process sensor trip channel becoming tripped, the

scram contactor seal-in contact opens. At this point in time, the completion of

protection action is directed regardless of the state of the initiating process sensor

trip channel.

The reactor protection system reset switch (when enabled) bypasses the seal-in contact to permit the RPS to be reset to its normally energized state when all

process sensor trip channels are within their normal (untripped) range of operation.

In the event of concurrent trips of both trip systems A and B, manual reset is

automatically inhibited for a minimum time delay of 10 seconds. The time delay

prevents reset prior to the insertion of all control rods.

This requirement applies to all of the following functions:

Neutron monitoring system scram trip

Reactor vessel high pressure scram trip

Reactor vessel low water level scram trip

Turbine stop valve closure scram trip

Turbine control valve fast closure scram trip

Main steam line isolation valve closure scram trip

Scram discharge volume high water level scram trip

Primary containment high pressure scram trip

Manual scram pushbuttons

The turbine stop valve closure and turbine control valve fast closure trip bypass function is

placed into effect only when the turbine first-stage pressure is at or below the setpoint

value. For plant operation above this setpoint, the trip channels will initiate protective

action once the scram contactors have de-energized and opened the seal-in contact

associated with the RPS trip logic.

The scram discharge volume high water level trip bypass function is only required after a reactor scram when the discharge volume has accumulated water and must be drained.

Consequently, this bypass function permits completion of protective action once it is

initiated and satisfies this design requirement.

The main steam line isolation valve closure trip bypass is in effect only when the reactor mode switch is in the SHUTDOWN, REFUEL, or STARTUP/HOT STANDBY position.

QUAD CITIES - UFSAR Revision 9, October 2007 7.2-10 Completion of protective action is not influenced by the reactor mode switch, trip logic test switch, or the Neutron monitoring system trip bypass.

This design requirement is not applicable for the reactor protection system motor-generator

sets and power distribution and reactor protection systems outputs to other systems.

7.2.2.5 Initiating Signals and Circuits

Table 7.2-1 lists the analytical limits utilized in determining the scram setpoints of the

protection system. Figure 7.2-4 shows the scram functions in block form.

[7.2-28]

A. Neutron Monitoring System High Flux and Core Power Oscillations

Four IRM channels and three APRM channels are connected to each of the two RPS trip systems. IRM and APRM trip logic is modified by the position of the mode

switch as indicated in Table 7.2-1.

Two OPRM channels are connected to each of the two RPS trip systems. The OPRM trip logic is enabled (armed) manually by operator action or automatically during certain reactor core power and reactor recirculation flow conditions.

Under certain circumstances, such as initial startup or refueling, shorting links in the manual scram circuits may be removed to provide either coincident or non-

coincident source range monitoring (SRM) trip capability. Shorting links will be removed from the RPS circuitry whenever more than one control rod will be

removed from fueled cells with the vessel head less than fully tensioned. (For

Example. During shutdown margin demonstrations.) This requirement is not

applicable during withdrawal of control rods controlled by the control rod removal

Technical Specifications. Single rod withdrawal with the shorting links installed

and the head not tensioned is allowed provided that the core loading has been

verified to match an analyzed shutdown margin configuration and the one-rod-out

refueling interlock has been demonstrated operable. Verification of the removal of the shorting links during these conditions will be performed within 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> of

withdrawal of control rods and once per 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> thereafter. Removing both

shorting links in both manual scram circuits enables the nuclear instrument non-

coincident trips, allowing a single trip from any of the nuclear instruments to cause

a scram. Coincident trips may be enabled by removing one shorting link in one scram channel and the shorting link for the opposite nuclear instrument channel in the other scram channel. Four SRM channels are provided with retractable

detectors.

The neutron monitoring system is discussed in detail in Section 7.6.

B. Reactor High Pressure

High pressure within the reactor system poses a direct threat of rupture to the reactor coolant system pressure boundary. A pressure increase while the reactor is

operating compresses the steam voids and results in a positive reactivity insertion

causing increased core heat generation that could lead to a violation of the core

thermal-hydraulic safety limit.

[7.2-29]

The reactor high pressure scram setting is chosen slightly above the reactor vessel maximum normal operating pressure to permit normal operation without

spurious scrams, yet provide a wide margin to the pressure safety limit.

QUAD CITIES - UFSAR Revision 5, June 1999 7.2-11 C. Reactor Vessel Low Water Level A low water level in the reactor vessel indicates that the reactor is in danger of being inadequately cooled. Should water level decrease too far, fuel

damage could result as steam forms around fuel rods.

[7.2-30]

The reactor vessel low water level scram setting prevents fuel damage following abnormal operational transients caused by single equipment

malfunctions or single operator errors that result in a decreasing reactor

vessel water level.

Specifically, the scram setting is chosen far enough below normal operational levels to avoid spurious scrams but high enough above the top of the active

fuel to assure that enough water is available to account for evaporation losses

and displacements of coolant following the most severe abnormal operational

transient involving a level decrease. (See Section 15.6.)

D. Turbine Stop Valve Closure

Closure of the turbine stop valves with the reactor at power can result in a significant addition of positive reactivity to the core as the nuclear

system pressure rise collapses steam voids.

[7.2-31]

The turbine stop valve closure scram, which initiates a scram earlier than either the neutron monitoring system or high reactor pressure, is required to provide a

satisfactory margin below the core thermal hydraulic safety limit for this

category of abnormal operational transients.

The scram counteracts the addition of positive reactivity due to pressure increases by inserting negative reactivity with the control rods. (See

Section 4.6.) Although the reactor high pressure scram, in conjunction with

the pressure relief system, is adequate to preclude overpressurizing the nuclear system, the turbine stop valve closure scram provides additional

margin to the pressure safety limit.

The turbine stop valve closure scram setting is selected to provide the earliest positive indication that the valves are closing. The trip logic was

chosen both to identify those situations in which a reactor scram is

required for fuel protection and to allow functional testing of this scram

function.

E. Turbine Control Valve Fast Closure (Turbine Generator Load Rejection)

With the reactor and turbine-generator at power, fast closure of the turbine control valves can result in a significant addition of positive reactivity to the core

as nuclear system pressure rises.

The turbine control valve fast closure scram, which initiates a scram earlier than either the neutron monitoring system or reactor high pressure, is required to provide a satisfactory margin to the core thermal-hydraulic safety limit for this

category of abnormal operational transients. The scram counteracts the addition of positive reactivity due to pressure by inserting negative reactivity with the

control rods. (See Section 4.6.) Although the reactor high pressure scram, in

conjunction with the pressure relief system, is adequate to preclude QUAD CITIES - UFSAR Revision 10, October 2009 7.2-12 overpressurizing the nuclear system, the turbine control valve fast closure scram provides additional margin to the pressure safety limit.

The turbine control valve fast closure scram setting is selected to provide timely indication of control valve fast closure. The trip logic was chosen

to identify those situations in which a reactor scram is required for fuel

protection.

F. Main Steam Line Isolation Valve Closure

The automatic isolation of the main steam lines on low pressure was provided to give protection against rapid reactor depressurization and the resulting rapid

cooldown of the vessel. Advantage was taken of the main steam line isolation valve closure scram feature in the RUN mode to ensure that high power

operation at low reactor pressures does not occur, thus providing protection for

the fuel cladding integrity safety limit.

[7.2-32]

In addition, the main steam line isolation valve closure scram in the RUN mode anticipates the pressure and flux transients which occur during normal or

inadvertent isolation valve closure.

The main steam line isolation valve closure scram setting is selected to give the earliest positive indication that the valves are closing. The trip logic allows

functional testing of valve closure trip channels with one steam line isolated.

G. Scram Discharge Volume High Water Level

During normal operation, the scram discharge volume will be empty due to natural draining via normally open drain and vent valves. However, upon

initiation of a reactor scram, these drain and vent valves are closed to retain the control rod drive discharge water and limit the loss of reactor water inventory.

Due to the hydraulic design of the piping and the volume, the rate of change of

water level is relatively slow and is assumed to be negligible in terms of its

transient response influence on the sensor.

[7.2-33]

Should the SDV fill to the point where not enough space remains for the water displaced during a scram, control rod movement would be hindered in the event

a scram were required.

[7.2-34]

The water level scram setpoint is set such that sufficient free volume remains to accommodate the water displaced during a scram.

H. Drywell High Pressure

A high pressure inside the drywell could indicate a loss of reactor coolant, requiring a scram of the reactor to minimize the possibility of fuel damage and to

reduce the addition of energy from the core to the coolant. The reactor vessel low

water level scram also acts to scram the reactor for loss-of-coolant accidents. The

drywell high pressure scram setting is selected to be as low as possible without

inducing spurious scrams or isolations.

[7.2-35]

QUAD CITIES - UFSAR Revision 7, January 2003 7.2-13 I. Deleted.

J. Turbine-Generator Condenser Low Vacuum

The reactor is protected from the effects of a complete loss of vacuum in the turbine-generator condenser by closing the turbine stop valves and, ultimately, the turbine bypass valves. Closure of the turbine stop and bypass

valves causes a pressure transient, a neutron flux rise, and an increase in surface heat flux similar to that caused by turbine stop valve closure. The

turbine stop valve closure scram function is adequate to prevent the cladding

safety limit from being exceeded in the event of a turbine trip transient with

bypass closure. The scram on condenser low vacuum reduces the severity by

anticipating the transient and scramming the reactor at a slightly higher

vacuum than the setpoints that close the turbine stop valves and bypass

valves.

K. Electro-Hydraulic Control Low Fluid Pressure

The EHC Low Fluid pressure scram function is provided by the pressure switches that sense turbine control valve fast closure.

L. Manual Scram

Manual Scram Pushbuttons:

[7.2-37]

To provide the operator with means to shutdown the reactor independent of the automatic functioning of the RPS, two pushbuttons located in the

control room initiate a scram when both are actuated by the operator.

The IEEE-279-1968 General Functional Requirements (paragraph 4.1) are not applicable to RPS functions requiring intervention by the control room operator, however, the manual scram pushbuttons do comply with the IEEE-279-1968

Manual Actuation (paragraph 4.17) design requirement. Failure of an automatic

RPS function affects the automatic portions of the system but the manual A3 and

B3 trip logics will still be able to initiate protective action. The manual scram

pushbuttons are implemented into the circuitry immediately QUAD CITIES - UFSAR Revision 5, June 1999 7.2-14 above the manual scram contactors in order to minimize the dependence of manual scram capability on other equipment.

[7.2-38]

Trip Logic Test Switch:

The General Functional Requirements of IEEE-279-1968 are not applicable to the trip logic test switch, however, the IEEE-279-1968 Manual Actuation

requirement is met as follows:

Operation of one test switch in the A trip system and one test switch in the B trip system will initiate a reactor scram. This provision serves as a backup to the

normal manual scram pushbuttons. Due to its electrical connection at the

beginning of the trip logic strings, it does not meet a strict interpretation in

requiring operation of a minimum of equipment. However, due to its backup role

to the more direct manual scram pushbuttons, it is not necessary that these

switches meet this requirement in a literal sense. Furthermore, failure of any

given test switch will not interfere with the automatic RPS functions in any

manner.

M. Reactor Mode Switch in SHUTDOWN

The General Functional Requirements of IEEE-279-1968 are addressed as follows for the reactor mode switch:

[7.2-39]

When the reactor mode switch has been placed in one of its four possible positions, it selects the particular sensors for the scram functions and the

appropriate bypasses for certain sensors.

In addition, the mode switch performs certain interlock functions that are not associated with the RPS. Among these interlock actions are restrictions on

control rod withdrawal and movement of refueling equipment.

The mode switch consists of a single manual actuator connected to distinct switch banks. Each bank is housed within a fire retardant cover. Contacts from

each bank are wired to individual terminal boards by separate cable routing.

When the mode switch is set to a given position, it enables those protective functions pertinent to that mode of operation to perform the necessary automatic

protective action.

As a backup function to the reactor scram pushbuttons, movement of the mode switch to the SHUTDOWN position de-energizes the manual A3 and B3 RPS trip

logic strings to initiate reactor shutdown (IEEE-279-1968 Manual Actuation, paragraph 4.17). An operating bypass is placed around the mode switch

contacts, after the scram time delay is complete, to permit manual reset of the

RPS when in the SHUTDOWN mode for an extended time. The RPS automatic

trip channels and trip logic are independent of the A3 and B3 manual trip logic strings to provide assurance that the manual actuation will not interfere with

the automatic protective channels.

This scram is not considered a protective function because it is not required to protect the fuel or nuclear system process barrier, and it does not act to

minimize the release of radioactive material from any barrier.

[7.2-40]

QUAD CITIES - UFSAR Revision 5, June 1999 7.2-15 N. IEEE-279-1968 General Functional Requirements for Other Signals and Circuits The RPS reset switch is under the administrative control of the control room operator. Since the reset switch, through auxiliary delay contacts, is introduced

in parallel with the trip actuator seal-in contact, failure of the reset switch

cannot prevent initiation of protective action when a sufficient number of trip channels are in the tripped condition. Hence, the automatic initiation

requirement for protective action is not invalidated by this reset switch.

[7.2-41]

The reactor protection system motor-generator sets and power distribution comply since the RPS is a normally energized system, and a loss of power from

both M-G sets will initiate reactor shutdown. Also, since the power source to the

RPS trip logic is introduced at the beginning of the series string of individual trip

channel outputs, the RPS power system does not interfere with the automatic

action requirements of the protection system.

The reactor protection system trip logic, actuators, and trip actuator logic is arranged with four trip logic strings in the reactor protection system in a one-

out-of-two-twice arrangement. Hence, the RPS trip logic and trip actuator

circuitry comply with the design requirement.

The RPS provides output signals from isolated relay contacts to initiate control room annunciation, to process computer logging of trips as they occur, to actuate

electrically operated valves to provide for backup scram capability, and to

actuate electrically operated valves to isolate the discharge volume drain and

vent isolation valve. These individual outputs are isolated from the relay

contacts used to accomplish the protective actions to assure that the latter

portions are capable of accomplishing the automatic protective action when

required.

O. IEEE-279-1968 Manual Actuation Requirements for Other Signals and Circuits

Since the reactor protection system reset switch reset function does not initiate protective action, the design complies with this design requirement.

For the reactor protection system trip logic, actuators, and trip actuator logic, the trip actuator logic may be placed in a tripped condition from either one of the two

automatic trip logics, A1 or A2, or the manual trip logic A3 associated with one

RPS trip system. This action can be accomplished with the trip logic test switch, manual scram pushbutton, reactor mode switch, or with removable fuses in the

RPS cabinets. As a result, the design meets this design requirement.

The IEEE-279-1968 Manual Actuation design requirement is not applicable to the RPS automatic trip functions, bypass functions, motor-generator sets and

power distribution, or outputs to other systems.

7.2.2.6 Scram Bypasses

A number of scram bypasses are provided to account for the varying protection

requirements depending on reactor conditions and to allow for instrument service during

reactor operations. Some bypasses are automatic, others are manual.

[7.2-42]

QUAD CITIES - UFSAR Revision 7, January 2003 7.2-16 Where automatic bypasses are employed, the bypass is automatically removed when the conditions for bypass no longer exist. Other operating bypasses are manually installed and are under the administrative control of the control room operator. These controls meet the

requirements of IEEE-279-1968 Operating Bypasses (paragraph 4.12) for the following

functions:

Neutron monitoring system scrams

Turbine stop valve closure scram

Turbine control valve fast closure scram

Main steam line isolation valve closure scram

Condenser low vacuum scram

Scram discharge volume high water level scram

Turbine stop valve closure and turbine control valve fast closure trip bypass

Main steam line isolation valve closure trip bypass

All manual bypass switches and the reactor mode switch are in the control room, under the

direct control of the control room operator. Manual bypasses are controlled by mechanical, electrical, or administrative controls to maintain trip function operability through other

channels when one channel is bypassed. Trip functions which use inputs from fluid sensors

may also have individual sensors valved out-of-service and returned to service under the administrative control of the operator. Trip functions which use limit switch or position

switch inputs cannot be manually bypassed. These administrative and design controls

meet the requirements of IEEE-279-1968 Channel Bypass or Removal from Operation (paragraph 4.11) and Access to Means for Bypassing (paragraph 4.14) for the applicable trip

functions:

Neutron monitoring system scrams

Turbine stop valve closure and control valve fast closure scrams

Main steam isolation valve closure and condenser low vacuum scrams

Scram discharge volume high water level scram

Reactor vessel high pressure scram (bypass by valve isolation)

Reactor vessel low water level scram (bypass by valve isolation)

Primary containment high pressure scram (bypass by valve isolation)

QUAD CITIES - UFSAR Revision 9, October 2007 7.2-17 If the ability to trip some part of the system has been bypassed, this fact is continuously indicated in the control room. The requirements of IEEE-279-1968 Indication of Bypass (paragraph 4.13) are met for bypasses involving these RPS trip functions:

Neutron monitoring system IRM, APRM, and OPRM scram Turbine stop valve closure and control valve fast closure, (Turbine Gen. Load Rejection) Main steam line isolation valve closure and condenser low vacuum scram Scram discharge volume high water level scram (if tripped)

Reactor vessel high pressure scram (if tripped, also provides computer record)

Reactor vessel low water level scram (if tripped, also provides computer record)

Primary containment high pressure scram (if tripped, also provides computer record)

Reactor mode switch (when conditions for bypass are satisfied)

Reactor protection functions that are not applicable to the IEEE-279-1968 requirements are

listed under the exceptions in Item I.

For short duration bypasses that are not a permanently installed, the requirements of IEEE-

279-1968 Indication of Bypass (paragraph 4.13) are met by Administrative Controls of the

bypass; i.e. Caution Card and/or Procedure.

[7.2-42a]

The scram bypasses are as follows:

A. Neutron Monitoring System

Bypasses for the neutron monitoring system channels are described in Section 7.6.

To meet the IEEE-279-1968 General Functional Requirements (paragraph 4.1) and Channel Bypass or Removal from Operation requirements (paragraph 4.11), a

sufficient APRM and IRM channels are provided in the design to permit continuous bypass of one APRM channel in each trip system and continuous bypass of one IRM

in each trip system. The remaining APRM and IRM channels in service are

adequate in number and in their spatial coverage of the reactor core to comply with

the requirements. Also, a sufficient number of OPRM channels (each channel consisting of two modules) have been provided to permit any one OPRM module in a given trip system to be manually bypassed, while still ensuring that the remaining operable OPRM channels comply with the IEEE 279 design requirements.

[7.2-43]

In addition, when the reactor mode switch is in RUN, an IRM trip will not cause a scram unless the corresponding APRM has a downscale trip. The OPRMs can be manually enabled but are automatically enabled only during reactor power/flow map conditions of high power and low flow.

[7.2-44]

B. Turbine Control Valve Fast Closure and Turbine Stop Valve Closure

To meet the IEEE-279-1968 General Functional Requirements (paragraph 4.1), the turbine control valve fast closure scram and turbine stop valve closure scram is

provided with a bypass to permit continued reactor operation at low power levels

when the turbine valves are closed.

[7.2-45]

QUAD CITIES - UFSAR Revision 6, October 2001 7.2-18 Closure of these valves from such a correspondingly low initial power level does not constitute a threat to the integrity of any barrier to the release of radioactive

material.

[7.2-46]

Removal of this bypass is automatically accomplished as the reactor power and turbine first-stage pressure become elevated to the setpoint value. The setpoint

for actuation of this bypass is determined from transient analysis considerations

taking into account the resultant consequences of a bypassed turbine RPS trip as

a function of reactor operating power.

[7.2-47]

Two turbine first-stage pressure switches are provided for each trip system to initiate the automatic bypass. The switches are arranged so that no single

failure can prevent a turbine stop valve closure or turbine control valve fast

closure scram.

[7.2-48]

C. Main Steam Line Isolation Valves Closure and Condenser Low Vacuum

The General Functional Requirements of IEEE-279-1968 (paragraph 4.1) for this function are addressed as follows:

[7.2-49]

The main steam line isolation valve closure trip bypass function is a manual bypass in that the reactor mode switch must be placed in SHUTDOWN, REFUEL, or STARTUP/HOT STANDBY position to obtain the trip bypass. This

bypass is provided to permit the RPS to be manually reset when the plant is

operating in one of the three aforementioned modes with the isolation valves

closed. These conditions exist during startups, maintenance and certain

reactivity tests during refueling.

D. Scram Discharge Volume High Water Level

A manual keylock switch located in the control room permits the operator to bypass the SDV high water level scram if the mode switch is in SHUTDOWN

or REFUEL. This bypass allows the operator to reset the RPS and air dump

system, so that the system is restored to operation while the operator drains

the SDV (IEEE-279-1968 Operating Bypasses, paragraph 4.12). In addition

to allowing the scram relays to be reset, actuating the bypass initiates a

control rod block. Resetting the trip actuators opens the SDV vent and drain

valves.

The IEEE-279-1968 General Functional Requirements (paragraph 4.1) for automatic response are not meaningful for the bypass channels, since the

discharge volume high water level trip is bypassed by manual operation of a

bypass switch and the reactor system mode switch. Administrative control must

be applied to remove the bypass once the water has been drained from the

instrument volume associated with the discharge piping.

E. Reactor Mode Switch

A reactor mode switch is provided to select the necessary scram functions for various plant conditions. In addition to selecting scram functions from the proper sensors, the mode switch provides appropriate bypasses. The mode switch also interlocks such functions as control rod blocks and refueling

equipment restrictions, which are not considered here as part of the RPS. The

switch itself is designed to provide separation between the two trip systems.

[7.2-50]

QUAD CITIES - UFSAR Revision 12, October 2013 7.2-19 The mode switch positions and their related scram/scram bypass functions are as follows:

1. SHUTDOWN - Initiates a reactor scram; selects neutron monitoring system scram for low neutron flux level operation (enables the IRM high-high flux

and selects the 15% power APRM high-high flux scram signals); bypasses

main steam line isolation valve closure and condenser low vacuum scrams.

2. REFUEL - Selects neutron monitoring system scram for low neutron flux level operation (enables the IRM high-high flux and selects the 15% power APRM

high-high flux scram signals), bypasses main steam line isolation valve closure and condenser low vacuum scrams.

3. STARTUP/HOT STANDBY - Selects neutron monitoring system scram for low neutron flux level operation (enables the IRM high-high flux and selects the

15% power APRM high-high flux scram signals); bypasses main steam line

isolation valve closure and condenser low vacuum scrams.

4. RUN - Selects neutron monitoring system scram for power range operation (bypasses the IRM high-high flux scram when the companion APRM is not

downscale or inoperative, and selects the APRM flow-biased high-high flux

setpoint).

The reactor mode switch complies with IEEE-279-1968 Channel Bypass or Removal from Operation (paragraph 4.11) in the following manner.

[7.2-51]

The use of four banks of contacts for the mode switch permits any RPS trip channel, which is connected into the mode switch, to be periodically tested in a

manner that is independent of the mode switch itself. Consequently, for any

stated position of the mode switch, a sufficient number of trip channels will remain operable during the periodic test to fulfill this design requirement.

Movement of the mode switch handle from one position to another will disconnect

all redundant channels associated with the former position and will connect all

redundant channels pertinent to the latter position. In this manner, the mode

switch complies with this design requirement.

There are no operating bypasses that are imposed upon the RPS trip channels or RPS trip logic as the result of the position of the mode switch itself (IEEE-279-

1968 Operating Bypasses, paragraph 4.12).

The mode switch is under the administrative control of plant personnel. Since other controls must be operated or other sensors must be in an appropriate state

to complete the operating bypass logic, the mode switch itself satisfies the

requirements of IEEE-279-1968 Access to Means for Bypassing (paragraph 4.14).

F. Manual Scram Pushbuttons

Since actuation of one manual scram pushbutton places its entire RPS trip system in a tripped condition, the automatic trip channels are ignored until such

time as the RPS is reset to its normally energized state. This particular result is

in compliance with IEEE-279-1968 Channel Bypass or Removal from Operation (paragraph 4.11).

QUAD CITIES - UFSAR Revision 7, January 2003 7.2-20 G. Trip Logic Test Switch The test switch is connected into the RPS trip logic preceding all individual trip channel outputs. Consequently, operation of the test switch causes the entire

trip logic string to become de-energized and places one RPS trip system in a

tripped state. Hence, the test switch meets the IEEE-279-1968 Channel Bypass

or Removal from Operation (paragraph 4.11) design requirement.

The test switch does not fit the bypass definition, but since it is capable of removing the trip logic from operation by placing it in a tripped state, it is

important that appropriate indication be given to the operator. In this situation, the operator would receive annunciation that one RPS trip system is in a tripped

state, but no trip channels would be annunciated if they remained within their

setpoint limit. This combination would provide the operator with an indication

that the test switch operation was proper. In this way, the trip logic test switch

meets the requirements of IEEE-279-1968 Indication of Bypass (paragraph 4.13).

H. Trip Bypass Features

The trip bypass features and their applicability to IEEE-279-1968 requirements are covered previously within the discussions for those specific trips.

I. Exceptions The requirements of IEEE-279-1968 Channel Bypass or Removal from Operation (paragraph 4.11) are not applicable for the following functions and equipment.

[7.2-51a]

Primary containment high pressure scram trip

Reactor protection system reset switch

Reactor protection system motor-generator sets and power distribution

Reactor protection system trip logic, actuators, and trip actuator logic

Reactor protection systems outputs to other systems

The requirements of IEEE-279-1968 Operating Bypasses (paragraph 4.12) are not applicable to the following functions and equipment.

Reactor vessel high pressure scram trip

Reactor vessel low water level scram trip

Primary containment high pressure scram trip

Manual scram pushbuttons

Trip logic test switch QUAD CITIES - UFSAR Revision 7, January 2003 7.2-21 Reactor protection system reset switch

Reactor protection system motor-generator sets and power distribution

Reactor protection system trip logic, actuators, and trip actuator logic

Reactor protection systems outputs to other systems

The requirements of IEEE-279-1968 Indication of Bypass (paragraph 4.13) are not applicable for the following functions and equipment.

Manual scram pushbuttons

Reactor protection system reset switch

Reactor protection system motor-generator sets and power distribution

Reactor protection system trip logic, actuators, and trip actuator logic

Reactor protection systems outputs to other systems

The requirements of IEEE-279-1968 Access to Means for Bypassing (paragraph 4.14) are not applicable to the following functions and equipment.

Reactor vessel high pressure scram trip

Reactor vessel low water level scram trip

Primary containment high pressure scram trip

Manual scram pushbuttons

Trip logic test switch

Reactor protection system reset switch

Reactor protection system motor-generator sets and power distribution

Reactor protection system trip logic, actuators, and trip actuator logic

Reactor protection systems outputs to other systems

7.2.2.7 Redundancy, Diversity, and Separation

Instrument piping that taps into the reactor vessel is routed through the drywell wall and

terminates inside the secondary containment (reactor building). Reactor vessel pressure

and water level information is sensed from this piping by instruments mounted on

instrument racks in the reactor building.

[7.2-52]

Valve position switches are mounted on valves from which position information is required.

The sensors for RPS signals from equipment in the turbine building are mounted locally in

the turbine building. The two M-G sets that supply power for the RPS QUAD CITIES - UFSAR Revision 5, June 1999 7.2-22 are located in the electrical equipment room in the service building in an area where they can be serviced during reactor operations. Power and sensor cables are routed to two RPS cabinets in the control room, where the logic circuitry of the system is formed. The trip

logics of each trip system are isolated in separate bays in each cabinet. The RPS, except for

the RPS power supplies upstream of the EPAs, was designed using Class I equipment to

assure a safe reactor shutdown during and after seismic disturbances.

The scram pilot valve solenoids are powered from eight trip actuator logic circuits: four circuits from trip system A, and four from trip system B. The four circuits associated with

any one trip system are run in separate conduits. One trip actuator logic circuit from each

trip system may run in the same conduit; wiring for the two solenoids associated with any

one control rod may run in the same conduit.

7.2.2.8 Testability

Provisions are made for timely verification that each active or passive component in the

RPS is capable of performing its intended function as an individual component and/or in

conjunction with other components. In fulfillment of this general objective, tests are

provided to verify that the following specific conditions exist:

[7.2-53]

A. Each instrument channel functions independent of all others;

B. Sensing devices will respond to process variables and provide channel trips at correct values;

C. Paralleled circuit elements can independently perform their intended function;

D. Series circuit elements are free from shorts that can nullify their function;

E. Redundant instrument or logic channels are free from interconnecting shorts that could violate independence in the event of a single malfunction;

F. No element of the system is omitted from the test if it can in any way impair operability of the system. If the test is done in parts, then the parts must be

overlapping to a sufficient degree to assure operability of the entire system;

and G. Each monitoring alarm or indication function is operable.

The reactor protection system can be tested during reactor operation by five separate tests.

The first of these is the manual trip actuator test. By depressing the manual scram button

for one trip system, the manual trip logic actuators are de-energized, opening contacts in

the trip actuator logics. After resetting the first trip system tested, the second trip system

is tripped with the other manual scram button. The total test verifies the ability to de-

energize all eight groups of scram pilot valve solenoids by using the manual scram pushbutton switches. Scram group indicator lights verify that the trip actuator contacts

have opened.

The second test is the automatic trip actuator test which is accomplished by operating the

keylocked test switches, one at a time, for each automatic trip logic. The switch de-

energizes the trip actuators for that trip logic, causing the associated trip actuator contacts

to open. The test verifies the ability of each trip logic to de-energize the trip QUAD CITIES - UFSAR Revision 7, January 2003 7.2-23 actuator logics associated with the parent trip system. The actuator and contact action can be verified by observing the physical position of these devices.

The third test includes calibration of the neutron monitoring system and analog trip system

by means of internal simulated inputs from calibration signal units. Section 7.6 describes

the calibration procedures. Likewise, the main steam line radiation monitoring system (Section 11.5) is calibrated using internal calibration signals.

[7.2-54]

The fourth test is the single rod scram test which verifies the capability of each rod to

scram. It is accomplished by operating the toggle switches on the protection system

operations panel. Timing traces can be made for each rod scrammed.

The fifth test involves applying a test signal to each RPS trip channel in turn and observing

that a trip logic trip results at the required trip point. This test also verifies the electrical independence of the trip channel circuitry. For trip channels which are initiated by position

switches, thermal switches, and radiation monitors, the appropriate method of applying a

test signal to the sensing instrument will be used. The test signals can be applied to the

process sensing instruments (pressure and differential pressure) through calibration taps.

The test is conducted as follows:

A. An instrument technician, following approved plant procedures, isolates specific instruments using the instrument valve (or instrument manifold valve) and a calibration set is attached to the instrument calibration taps which are arranged

to avoid spilling of water (if the instruments are normally filled).

B. A calibration signal sufficient to actuate the sensor contacts is applied while reading the value of applied pressure on calibrated test equipment.

C. The trip point and reset point are compared to the required setpoint and the trip values are logged.

D. Adjustments are made to the trip setting if necessary; and the adjustments are logged stating the measured "as-left" setpoint.

E. Communication with the control room is established during the test to verify the trip point as registered on control room instruments. The trip value is logged.

F. Proper protective relay operation is also verified by observation.

G. The calibration signal is then reduced to zero, the test set is removed, the calibration taps plugged, and the sensors are valved into service in their operating positions.

H. The test is logged as complete.

Reactor protection system response times are first verified during routine surveillance

testing. The elapsed times from sensor trip to each of the following events is measured:

A. Trip channel relay de-energized, and

B.Trip actuators de-energized.

QUAD CITIES - UFSAR Revision 9, October 2007 7.2-24The EPAs are routinely tested to ensure proper operation. The testing includes calibration as well as a verification that the breakers will trip during conditions of undervoltage, underfrequency, and overvoltage.

[7.2-55]

Reactor protection system safety-related HFA relays had their coils replaced with General

Electric Century Series coils. HFA relays are inspected on a sampling basis.

[7.2-56]

The following text discusses the applicability of the RPS functions to IEEE-279-1968

Capability of Sensor Checks (paragraph 4.9).

Neutron monitoring system scram trip

[7.2-57] During reactor operation in the RUN mode, the IRM detectors are stored below the reactor core in a low flux region. Movement of the detectors into the core

permits the operator to oversee the instrument response from the different IRM

channels and will confirm that the instrumentation is operable.

In the power range of operation, the individual LPRM detectors respond to local neutron flux and provide the operator with an indication that these instrument channels are responding properly. The six APRM channels may also be observed

to respond to changes in the gross power level of the reactor to confirm their

operation.

Each APRM instrument channel may also be calibrated with a simulated signal introduced into the amplifier input, and each IRM instrument channel may be

calibrated by introducing an external signal source into the amplifier input.

Each OPRM module may be calibrated with simulated signals introduced into the module utilizing the OPRM Maintenance Terminal.

During these tests, proper instrument response may be confirmed by observation of instrument lights in the control room and trip annunciators.

Reactor vessel high pressure scram trip One sensor may be valved out-of-service at a time to perform a periodic test of the trip channel. During this test, operation of the sensor, its contacts, and the

balance of the RPS trip channel may be confirmed.

Reactor vessel low water level scram trip Because of the one-out-of-two-twice configuration of the RPS trip logic for this protective function, one level sensor may be removed from service to perform the

periodic test on any trip channel.

Turbine stop valve closure scram trip The logic of the four RPS trips is as follows:

A1 (tripped) = Turbine stop valve 1 partially closed, and turbine stop valve 2 partially closed A2 (tripped) = Turbine stop valve 3 partially closed, and turbine stop valve 4 partially closed B1 (tripped) = Turbine stop valve 1 partially closed, and turbine stop valve 3 partially closed QUAD CITIES - UFSAR Revision 7, January 2003 7.2-25 B2 (tripped) = Turbine stop valve 2 partially closed, and turbine stop valve 4 partially closed

For any single stop valve closure test, two of the trip channels will be placed in a tripped condition, but none of the trip logics will be tripped, and no RPS

annunciation or computer trip channel logging will be evident. This

arrangement permits single valve testing without corresponding tripping of the

RPS, and the observation that no RPS trips result is a valid and necessary test

result.

At reduced power levels, two valves may be tested in sequence to produce RPS trips, annunciation of the trips, and computer printout of the trip channel

identification. These observations are another important test result that

confirms proper RPS operation.

In sequence, each combination of single valve closures and dual valve closures is performed to confirm proper operation of all trip channels.

Turbine control valve fast closure scram trip

During any control valve fast-closure test, one RPS trip channel will be tripped and will produce both control room annunciation and computer record of the trip

channel identification.

Main steam line isolation valve closure scram trip

[7.2-58]

For any single valve closure test, two of the trip channels will be placed in a tripped condition, but none of the trip logics will be tripped, and no RPS

annunciation or computer trip channel record will be evident. This arrangement

permits single valve testing without corresponding tripping of the RPS. The

observation that no RPS trips result is a valid and necessary test result.

[7.2-59]

At reduced power levels, two valves may be tested in sequence to produce RPS trips, annunciation of the trips, and computer printout of the trip channel

identification. These observations are another important test result that

confirms proper RPS operation.

In sequence, each combination of single valve closures in each of two main steam lines is performed to confirm proper operation of all eight trip channels.

These test results confirm that the valve limit switches operate as the valves are manually closed.

Scram discharge volume high water level scram trip

During reactor operation, the discharge volume level sensors may be tested by using the instrument isolation valves in proper sequence in conjunction with quantities of demineralized water.

Primary containment high pressure scram trip

During reactor operation one pressure switch may be valved out-of-service at a time to perform periodic testing.

QUAD CITIES - UFSAR Revision 7, January 2003 7.2-26 Reactor mode switch

Operation of the mode switch may be verified by the operator during plant operation by performing certain sensor tests to confirm proper RPS operation.

Movement of the mode switch from one position to another is not required for

these tests since the connection of appropriate sensors to the RPS logic, as well

as disconnection of inappropriate sensors, may be confirmed from the sensor

tests.

Turbine stop valve closure and turbine control valve fast closure trip bypass

Testing of individual pressure switches is permitted during plant operation by valving out-of-service one pressure switch at a time. A variable pressure source

may then be introduced to the switch to confirm the setpoint value and switch

operation.

Neutron monitoring system trip bypass

At any time, the operator may confirm proper operation of the neutron monitoring system bypass channels by placing the bypass switch for any given

trip system into specific positions and introducing trip conditions into one

neutron monitoring system trip channel at a time for that same trip system. A

sequential combination of these operations will provide for complete verification

of the neutron monitoring system bypass channels.

Scram discharge volume high water level trip bypass

During plant operation in the STARTUP/HOT STANDBY and RUN modes, imposition of this bypass function is inhibited by the reactor mode switch. Under

these circumstances, operation of the bypass switch should not produce a bypass

condition for any single trip channel. This fact can be determined from the

control room annunciator, a visual inspection of the bypass relays, and the process computer printout of any discharge volume high water level trip channel

placed in a tripped condition prior to the bypass switch test.

Main steam line isolation valve closure trip bypass

Testing of the bypass circuit is possible in the SHUTDOWN, REFUEL, or STARTUP/HOT STANDBY positions of the mode switch. Confirmation that the

bypass is not in effect in the RUN mode may be made at operating conditions.

Reactor protection systems outputs to other systems

Output signals from the RPS have not been derived at the process sensor interface due to a lack of adequate isolation at this point. Rather, the outputs

have been obtained from the trip channel relays and trip actuator relays which

do provide adequate isolation of the signal source.

QUAD CITIES - UFSAR Revision 4, April 1997 7.2-27 Exceptions

This design requirement is not applicable to the following equipment:

Manual scram pushbuttons

Trip logic test switch

Reactor protection system reset switch

Reactor protection system motor-generator sets and power distribution

Reactor protection system trip logic, actuators, and trip actuator logic

7.2.2.9 Environmental Considerations

The reactor protection system components which are located inside the primary

containment and which must function in an environment resulting from a break of the

nuclear system process barrier inside the primary containment, are the temperature

equalizing columns and condensing chambers. Special precautions are taken to ensure

their satisfactory operability after an accident. The condensing chambers are addressed in

the reactor vessel instrumentation portion of Section 7.6.

[7.2-60]

Sensing elements are equipped with enclosures so that they can withstand conditions that

may result from a steam or water line break long enough to perform satisfactorily.

[7.2-61]

Wiring and cables for RPS instrumentation were selected to avoid excessive deterioration

from temperature and humidity during the design life of the plant. Cables and connectors

used inside the primary containment were designed for continuous operation at an ambient temperature of 150°F and a relative humidity of 99%.

Cables required to carry low-level signals currents of less than 1 mA or voltages of less than 100 mV were designed and installed to eliminate, insofar as practical, electrostatic and electromagnetic pickup from power cables and other ac or dc fields. In these cases, ferromagnetic conduits or totally enclosed ferromagnetic trays are used.

7.2.2.10Operational Considerations

The operational considerations of the RPS are as follows:

[7.2-62]

A. Indicators

Indication or annunciation is available for all parameters used by the RPS.

Each of the eight scram groups (A1-A4 and B1-B4) is provided with a normally-energized indicator light at the RPS cabinets and on the main control panel. The

scram group indicators extinguish when an actuator logic opens.

The data presented to the operator for all of the RPS functions comply with the IEEE-279-1968 Information Readout (paragraph 4.20) design requirement.

QUAD CITIES - UFSAR Revision 11, October 2011 7.2-28 Indications provided for the specific RPS functions and conformance to IEEE-279-1968 Identification of Protective Actions are discussed in Item F.

B. Annunciators

Whenever an RPS sensor trips, it lights a white annunciator window for that variable on the reactor control panel in the control room. The first trip

system to trip also lights a red window to indicate which trip system tripped

first. [7.2-63]

An RPS trip channel trip also sounds a horn, which can be silenced by the operator. The annunciator window lights remain illuminated until all sensors

that tripped in a group of sensors monitoring the same variable are clear. When

all sensors in a group of sensors monitoring the same variable are clear, the alarm window slow flashes. The alarm window slow flashing is a visual indication to the operator that all sensors in that group of sensors are clear, and

the operator may reset the window with the reset pushbutton. The red window

is reset by a separate reset pushbutton. The individual sensors that tripped in a

group of sensors monitoring the same variable may be identified by the position

of the RPS relays (tripped or untripped). The location of the alarm windows on

the annunciator provides the operator with the means to quickly identify the

cause of RPS trips and to evaluate the threat to the fuel or nuclear system

process barrier.

The control room annunciations for the RPS functions and equipment comply with the requirements of IEEE-279-1968 Information Readout (paragraph 4.20).

[7.2-64]

Annunciators provided for the specific RPS functions and conformance to IEEE-279-1968 are discussed in Item F.

C.Computer Alarms To provide the operator with the ability to analyze an abnormal transient during which events occur too rapidly for direct operator comprehension, all

RPS trips are monitored by the process computer system and recorded in

historical archives that may be retrieved later for review. These archives are

described in detail in the process computer documentation.

[7.2-65] Computer inputs provided for the specific RPS functions and conformance to IEEE-279-1968 are discussed in Item F.

D. Operator Controls

The reactor mode switch, which selects the proper interlocking for the operating or shutdown condition of the plant including the scram/scram bypass functions, is addressed in Section 7.2.2.6.

QUAD CITIES - UFSAR Revision 10, October 2009 7.2-29 Whenever either manual scram pushbutton is depressed, a red indicating light in the pushbutton is illuminated and a trip system trip occurs. When the trip

pushbuttons for both trip systems are depressed, or the reactor mode switch is

placed to the shutdown position, a reactor scram occurs.

[7.2-66]

E. Operable Trip Channels

To ensure that the RPS remains functional, the number of operable trip channels for the essential monitored variables should be maintained at or above the

minimums given in the Technical Specifications. The minimums apply to any

untripped trip system; a tripped trip system may have any number of inoperative

trip channels. Because reactor protection requirements vary with the mode in

which the reactor operates, the tables in the Technical Specifications show

different functional requirements for the RUN, STARTUP/HOT STANDBY and

REFUEL modes.

[7.2-67]

F. IEEE-279-1968 Identification of Protective Actions (paragraph 4.19)

The reactor protection system trip logic, actuators, and trip actuator logic use four control room annunciators to identify the tripped portions of the RPS in

addition to the previously described trip channel annunciators:

[7.2-68]

A. A1 or A2 automatic trip logics tripped;

B. A3 manual trip logic tripped;

C. B1 or B2 automatic trip logics tripped; and

D. B3 manual trip logic tripped.

These same functions are connected through independent auxiliary contacts of the scram contactors to the process computer to provide a record of the relay operations. These methods may be used to identify the protective action for any

of the RPS functions listed below.

Reactor vessel high pressure scram trip

Reactor vessel low water level scram trip

Scram discharge volume high water level scram trip

Primary containment high pressure scram trip

Manual scram pushbuttons

Turbine control valve fast closure scram trip

Protective actions for the remaining RPS functions may be identified as described in the following text.

QUAD CITIES - UFSAR Revision 10, October 2009 7.2-30 Neutron monitoring system scram trip A common neutron monitoring system annunciator is provided in the control room to indicate the source of the RPS trip. The process computer provides

a record of the RPS A1, A2, B1, and B2 neutron monitoring system channel trips, as well as identification of individual IRM and APRM channel trips.

The Sequence of Events Recorder (SER) provides a record of the OPRM

system channel trips.

Each RPS trip system has one IRM upscale or inoperative annunciator and one APRM upscale or inoperative annunciator in the control room. Two

additional annunciators indicate any IRM downscale or any APRM

downscale. Each RPS trip system has an OPRM trip annunciator. Two

additional annunciators indicate any OPRM "alarm" or "trouble/inop" conditions.

Each instrument channel, whether IRM, APRM, or OPRM has control room panel lights indicating the status of the channel.

Turbine stop valve closure scram trip Partial or full closure of a particular set of two turbine stop valves will initiate a control room annunciator when the trip point has been exceeded.

This same condition will permit identification of the tripped channels in the form of a record from the process computer or by visual observation of the relay contacts in the RPS panels.

Main steam line isolation valve closure scram trip Partial or full closure of any main steam line valve is indicated by valve position indicator lights in the control room. These indications are not a

part of the reactor protection system but they do provide the operator with

valid information pertinent to the valve status.

Partial or full closure of two valves in a particular set of main steam lines will initiate a control room annunciator when the trip setpoint has been

exceeded. This same condition will permit identification of the tripped trip

channels in the form of a record from the process computer or visual inspection of the relay contacts at the RPS panels.

Reactor mode switch Identification of the mode switch in SHUTDOWN position scram trip is provided by the manual scram annunciators, their process computer trip

logic identification printout, and the mode switch in SHUTDOWN position

annunciator.

Reactor protection system reset switch Reset of the RPS is not a protective action; however, proper operation of the switch may be inferred from removal of annunciated conditions as the RPS

returns to its normally energized state.

Reactor protection systems outputs to other systems The design of the RPS output networks complies with this design requirement.

QUAD CITIES - UFSAR Revision 7, January 2003 7.2-31 The RPS trip bypasses provide no protective action; therefore, one control room annunciator is provided to indicate the bypass condition. The RPS bypasses are

discussed in Section 7.2.2.6.

This design requirement is not applicable to the following equipment.

Trip logic test switch

Reactor protection system motor-generator sets and power distribution

7.2.2.11 Anticipated Transient Without Scram

The alternate rod insertion (ARI) functions as an alternate means for reactor shutdown in the event that a required scram is not effected by the RPS. The anticipated transient

without scram system includes ARI and is addressed in Section 7.8.

[7.2-69]

7.2.3 Analysis of Design Requirements Conformance

The reactor protection system is designed to provide protection against the onset and

consequences of conditions that threaten the integrity of the fuel barrier and the nuclear

system process barrier. Chapter 15 identifies and evaluates events with respect to the fuel

barrier and reactor coolant pressure boundary (RCPB) integrity.

[7.2-70]

The scrams initiated by neutron monitoring system variables, turbine stop valve closure, turbine control valve fast closure, main steam isolation valve closure, and reactor vessel low

water level are sufficient to prevent fuel damage following abnormal operational transients.

Specifically, these scram functions initiate a scram in time to prevent the core from

exceeding the thermal-hydraulic safety limit during abnormal operational transients.

The scram initiated by reactor high pressure, in conjunction with the pressure relief

system, is sufficient to prevent damage to the nuclear system process barrier as a result of

internal pressure. For turbine-generator trips, the turbine stop valve closure scram and turbine control valve fast closure scram provide a greater margin to the nuclear system pressure safety limit than the high pressure scram. Chapter 15 identifies and evaluates

accidents and abnormal operational events that could result in reactor vessel pressure

increases.

The scrams initiated by the neutron monitoring system, main steam isolation valve closure, and reactor vessel low water level satisfactorily limit the radiological consequences of gross

failure of the fuel or nuclear system process barriers. Chapter 15 evaluates failures of the

fuel.

The scram discharge volume high water level scram, drywell high pressure scram, and manual scram provide protective functions not directly related to protecting the fuel or

process barriers.

[7.2-71]

The following text discusses the system variable inputs to the RPS functions as they apply

to IEEE-279-1968 Derivation of System Inputs (paragraph 4.8).

[7.2-72]

QUAD CITIES - UFSAR Revision 9, October 2007 7.2-32 Neutron monitoring system scram trip The measurement of neutron flux is an appropriate variable to determine the reactor power relative to the predetermined setpoint. Additional design details are

available in General Electric NED Topical Report APED-5706. The OPRMs are auto-enabled in the operating region of potential thermal hydraulic instability based on reactor flow and power inputs from the transmitters in the reactor coolant recirculation lines via flow units and from the APRMs respectively.

Turbine stop valve closure scram trip The measurement of turbine stop valve position is an appropriate variable for this RPS protective function. The desired variable is loss of the reactor heat sink.

However, stop valve closure is the logical variable to infer that the steam path has

been blocked between the reactor and the heat sink.

Turbine control valve fast closure scram trip Due to the normal throttling action of the turbine control valves with changes in the plant power level, measurement of control valve position is not an appropriate

variable for this protective function. The desired variable is "rapid loss of the

reactor heat sink"; consequently, some measurement of control valve closure rate is

indicated. Protection system design practice has discouraged use of rate-sensing devices for protective purposes, and in this instance, it was determined that detection of

hydraulic actuator operation or hydraulic fluid pressure would be a more positive

means of determining fast closure of the control valves.

These selected measurements are felt to be adequate and proper variables for the protective function taking into consideration the reliability of the chosen sensors

relative to other available sensors and the difficulty in making direct measurement

of control valve fast-closure rate.

Reactor vessel low water level scram trip Actual water level is the desired variable, and the selected sensors monitor this variable directly. Thus, the chosen variable is the proper one to provide the

necessary protective function.

Reactor vessel high pressure scram trip For this protective function, selection of reactor vessel pressure is an appropriate variable to provide the required protective function.

Main steam line isolation valve closure scram trip The measurement of the main steam line isolation valve position is an appropriate variable for the reactor protection system. The desired variable is loss of the

reactor heat sink, however, isolation valve closure is the logical variable to infer

that the steam path has been blocked between the reactor and the heat sink.

It should be noted that other valves in this steam path, such as turbine stop valves, etc., are also monitored by the reactor protection system to assure proper response

of the reactor to path blockages downstream of the main steam line isolation valves.

QUAD CITIES - UFSAR Revision 7, January 2003 7.2-33 Scram discharge volume high water level scram trip

The measurement of discharge volume water level is an appropriate variable for this protective function. The desired variable is "available volume" to accommodate a reactor scram. However, the measurement of consumed volume, by determining that the water level has risen to a fixed value, is sufficient to

infer the amount of remaining available volume, since the total volume is a fixed, predetermined value.

Primary containment high pressure scram trip

The measurement of primary containment high pressure is an appropriate variable to detect an abnormal condition within this boundary. High pressure

within the primary containment could indicate a break in the nuclear system

process barrier and these sensors would respond to limit the consequences of

such a break.

Reactor mode switch

Since the mode switch is used to connect appropriate sensors into the RPS logic depending upon the operating state of the reactor, the selection of particular

contacts to perform this logic operation is an appropriate means for obtaining the

desired function.

Turbine stop valve closure and turbine control valve fast closure trip bypass

Since the intent of this bypass is to permit continued reactor operation at low power levels when the turbine stop or control valves are closed, the selection of

turbine first-stage pressure is an appropriate variable for this bypass function.

In the power range of reactor operation, turbine first-stage pressure is essentially linear with increasing reactor power. Consequently, this variable

provides the desired measurement of power level.

Neutron monitoring system trip bypass

Due to the requirement for operator actuation of the bypass function, this design requirement is satisfied by the four control room bypass switches.

Scram discharge volume high water level trip bypass

Due to the manual action required for this bypass function, this design requirement is satisfied by operator interaction with a single bypass switch and

the mode switch.

Main steam line isolation valve closure trip bypass

The instrumentation furnished for this bypass function complies with the design requirement.

QUAD CITIES - UFSAR Revision 13, October 2015 7.2-34 The main steam line isolation valve closure trip will result from valve closure whenever the reactor is operating in the RUN mode. This constraint has been

selected to permit manual reset of the RPS under specified conditions whenever

the main steam line isolation valves are partially or fully closed.

Reactor protection systems outputs to other systems

Selection of specific outputs from the RPS to the annunciation and process computer systems has been based on the objective of monitoring the RPS

performance and providing meaningful information.

Exceptions

This design requirement is not applicable for the following equipment:

Manual scram pushbuttons

Trip logic test switch

Reactor protection system reset switch

Reactor protection system motor-generator sets and power distribution

Reactor protection system trip logic, actuators, and trip actuator logic

7.2.3.1 Single Failure Criterion

In terms of protection system nomenclature, the RPS is a one-out-of-two-twice logic system.

Theoretically, its reliability is slightly higher than a two-out-of-three system and slightly

lower than a one-out-of-two system. However, since the differences are slight, they can, in a practical sense, be neglected. The advantage of the dual trip system arrangement is that

it can be tested thoroughly during reactor operation without causing a scram. This capability for a thorough testing program, which contributes significantly to increased

reliability, is not possible for a one-out-of-two system.

[7.2-73]

The use of an independent trip channel for each trip logic allows the system to sustain any

trip channel failure without preventing other sensors monitoring the same variable from

initiating a scram. A single sensor or trip channel failure will cause a single trip system trip and actuate alarms that identify the trip. The failure of two or more sensors or trip

channels would cause either a single trip system trip, if the failures were confined to one

trip system, or a reactor scram, if the failures occurred in different trip systems. Any

intentional bypass, maintenance operation, calibration operation, or test - all of which result in a single trip system trip - leaves at least two trip channels per monitored variable capable of initiating a scram by causing a trip of the remaining trip system. The resistance

to spurious scrams contributes to plant safety, because unnecessary cycling of the reactor

through its operating modes would increase the probability of error or actual failure.

Each control rod is controlled as an individual unit. A failure of the controls for one rod

would not affect other rods. The backup scram valves provide a second method of venting the air pressure from the scram valves, even if either scram pilot valve solenoid for any

control rod fails to de-energize when a scram is required.

QUAD CITIES - UFSAR Revision 9, October 2007 7.2-35 Failure of either RPS M-G set would result, at worst, in a single trip system trip (the de-energization of one of the two scram valve pilot solenoids on each CRD). Alternate power is

available to the RPS buses. A complete, sustained loss of electrical power to both buses

would result in a scram, delayed by the motor-generator set flywheel inertia, in about three

seconds (see Section 7.2.2.2).

The following RPS functions meet the single-failure criterion of IEEE-279-1968 (paragraph

4.2). [7.2-74]

Neutron monitoring system scram trip

In order to simplify the description of the trip channel logic, the contact structure associated with IRM 11, APRM 1, and OPRM 1 for RPS relay A will be discussed as "IRM A", "APRM A", and "OPRM A" respectively, and shall be described in detail. This discussion may then be related to the other trip channel in a similar manner. A.With the reactor mode switch in the SHUTDOWN, REFUEL, or STARTUP/HOT STANDBY position, IRM A upscale or inoperative (unless it is bypassed), or APRM A upscale or inoperative (unless it is bypassed), or OPRM A automatic suppression function (ASF) trip (unless it is bypassed) will produce a channel trip of output relay A.

B.With the reactor mode switch in the RUN position, IRM A upscale or inoperative (unless it is bypassed) and APRM A downscale (unless it is bypassed), or APRM A upscale or inoperative (unless it is bypassed), or OPRM A ASF trip (unless it is bypassed) will produce a channel trip of output relay A. C.A trip of channel output relay A or a trip of channel output relay C (associated with IRM 13, APRM 3A, and OPRM 3) will produce a RPS A1 channel trip. Similarly, a trip of channel output relay E (IRM 12, APRM 2, OPRM 2) or relay G (IRM 14, APRM 3B, OPRM 7) will produce a RPS A2 channel trip. An A1 trip or an A2 trip will produce a trip for the "A" RPS trip system.

Cables from individual LPRM and IRM detectors are grouped under the reactor vessel to correspond with the RPS trip channel designations and are run in conduit from the vessel pedestal area to the neutron monitoring system cabinets.

Reactor vessel high pressure scram trip

Two pressure transmitters are connected to each of two physically separated taps. The two pairs of transmitters are physically separated and each provides a

high pressure analog signal to a separate analog trip cabinet in the Service

Building Cable Spreading Room. From the analog trip cabinet, a contact is wired

to the RPS cabinet in the Control Room. Wiring between the pressure

transmitters, analog trip cabinets, and RPS cabinets is run in metal conduits to

maintain both physical separation and electrical isolation of the redundant

channel. The physical separation and the signal arrangement assure that no

single physical event can prevent a reactor high pressure scram demand from

occurring.

QUAD CITIES - UFSAR 7.2-36 Revision 12, October 2013 Reactor vessel low water level scram trip

The transmitters and analog trip units are arranged in pairs, in the same way as the RPS system high pressure switches. Wiring from one level transmitter is run

separately from the wiring associated with the other level transmitter on the same instrument line, and the wiring associated with level transmitters on one instrument line is separate from the wiring associated with level transmitters on the other transmitter line. The physical separation and signal arrangement

assure that no single physical event can prevent a scram due to reactor vessel low

water level.

Turbine stop valve closure scram trip

Wiring from the limit switch junction box for each stop valve is run in two separate conduits, one for each contact of the limit switch, to maintain the necessary

electrical and physical separation.

Turbine control valve fast closure scram trip

The pressure switches are physically separated and one contact from each pressure switch is used in the RPS trip channels.

There is no single failure that will prevent proper operation of this protective function when it is required.

[7.2-75]

Main steam line isolation valve closure scram trip

Each main steam isolation valve has a limit switch junction box in close proximity to the valve. Wiring from the limit switch junction box on each valve to the control room

RPS relay panels is required to be run in two separate conduits, one for each contact of

the limit switch, to maintain the necessary electrical and physical separation. One

contact from each limit switch is used with the RPS A trip system; the other contact is

used with the RPS B trip system. Failure of any single limit switch will not prevent

proper protection system operation when it is required.

[7.2-76]

The two relays associated with any one trip logic are located in one panel that is physically and electrically separated from the panel containing the other trip logic

circuits.

Scram discharge volume high water level scram trip

Two of the four float-type switches are connected to the north bank and two are connected to the south bank, each with separate process taps. Two differential

pressure transmitters are also connected to each bank. Each of these has a

separate process tap.

[7.2-77]

Wiring from each sensor to the control room relay cabinets is run in a separate conduit to maintain the electrical and physical separation of the sensor trip

channels, and a separate trip channel relay is provided for each pair of sensors. A

pair consists of one float-type and one dp sensor from opposite banks.

QUAD CITIES - UFSAR Revision 6, October 2001 7.2-37 Primary containment high pressure scram trip

One pressure switch is mounted on each pressure tap, and the redundant taps are physically separated from one another by the reactor vessel. Wiring from

each pressure switch is run in separate rigid conduit to the RPS cabinets in the

control room to maintain both physical and electrical separation and isolation

among the trip channels.

[7.2-78]

A separate trip channel output relay is provided for each pressure switch and each relay is physically separated from the others in the RPS cabinets.

Reactor mode switch

The reactor mode switch complies with the single-failure criterion. The mode switch has two physically separated banks operated by a single geared handle.

The A channel and B channel of RPS are separated by these two banks. The

channels of RPS are electrically isolated from one another. SQUG reviews of the

panels these switches are located in, concluded panels are seismically adequate.

Consequently, mechanical damage to the Mode Switch is not a credible event.

Therefore, no credible failures of this switch can disable the protective functions

of RPS.

Trip logic test switch

One switch is placed in each of the four RPS trip logics with each switch consisting of a two-position keylock configuration. The four switches are

mounted in the RPS panels to achieve both physical separation and electrical

isolation from the redundant test switches.

Reactor protection system reset switch

Each contact of the reset switch is wired to an individual auxiliary relay coil when contacts are used in the RPS trip logic.

Proper operation of the reset switch and its auxiliary relays can be ascertained during periodic tests of the RPS or whenever any particular channel is returned

from a tripped state to the normal untripped condition.

Since opening of the process sensor trip channels is the initiating event for reactor scram, failure of the reset switch will not prevent de-energization of the

trip actuators during the time interval that the process actually exceeds the trip

setpoint.

Turbine stop valve closure and turbine control valve fast closure trip bypass

Two pressure switches are mounted on each of two turbine first-stage pressure taps. Contacts from the pressure switches are routed in conduit to the RPS cabinets in the control room. Each pressure switch contact is connected to a

single bypass channel output relay. The logic configuration for the bypass is

one-out-of-two-twice such that a single bypass channel is associated with a single

trip channel for stop valve closure and with a single trip channel for control valve

fast closure.

No single failure of this bypass circuitry will interfere with the normal protective action of the RPS trip channel.

QUAD CITIES - UFSAR Revision 9, October 2007 7.2-38 Neutron monitoring system trip bypass

For any given bypass switch, the following design provisions have been made to ensure that one and only one channel is bypassed at one time with a given bypass switch:

A. The switch operator is a joystick type with four positions located at the quadrant extremes (i.e., 90, 180, 270, and 360 degrees) with the vertical

center being the off position. This switch type makes selection of bypass for one channel mutually exclusive from selection of any other channel

associated with that same switch.

B. Contacts from the bypass switch are connected to auxiliary relays whose coils are energized when one and only one bypass is in effect.

C. Cabling associated with the bypass switch is run to separate terminal boards within the panel to achieve greater physical separation and electrical

isolation.

Hence, any single failure of this bypass will not remove the necessary OPRM, APRM or IRM protection trip channel.

Scram discharge volume high water level trip bypass

The design of the bypass function requires manual operation of a bypass switch and the mode switch to establish four bypass channels. For the bypass switch, a

single operator connects to two separate blocks of switch contacts within the

switch body, and wiring from contacts is routed to separate terminal strips.

One set of switch contacts, in conjunction with mode switch contacts, is used to energize two trip channel bypass relays when the bypass condition is desired. In

a similar fashion, the other set of bypass switch and mode switch contacts

energize two other trip channel bypass relays. Contacts from one relay are

connected in series with contacts from a relay in the other group to produce the

RPS A1 trip channel bypass function. The trip channel bypass function for the

redundant RPS A2 trip channel is produced from series-connected contacts of the

other two relays.

Consequently, it is necessary that four-out-of-four relays be energized in order to bypass the automatic RPS trip channels for this protective function. There is no

single failure of this bypass function that will satisfy the four-out-of-four

condition necessary to establish the bypass condition. Hence, this function

complies with the single-failure criterion.

Main steam line isolation valve closure trip bypass

Two contacts from each bank of the mode switch are each connected to individual bypass relays. Each contact energizes one of four bypass relays whose contacts

are connected into the RPS trip logic.

The relationship of these bypass relays to the RPS trip channels is on a one-to-one basis. Consequently, two particular bypass relays must be energized in

order to bypass the protective function and no single failure in the bypass circuitry will interfere with the protective action of the trip channels.

QUAD CITIES - UFSAR Revision 7, January 2003 7.2-39 The following text discusses the remaining RPS functions as they apply to IEEE-279-1968 Single-Failure Criterion (paragraph 4.2).

Manual scram pushbuttons

Two manual scram pushbuttons have been located on one panel with approximately 6 inches separation to permit the operator to initiate protective

action with one motion of one hand. To provide testability during plant

operation without initiating protective action, the logic of the switches is

two-out-of-two in that both switches must be depressed (not necessarily

simultaneously) to cause reactor scram.

The manual scram pushbuttons, with the reactor mode switch in the SHUTDOWN position, satisfies the single-failure criterion for manual scram.

These controls are backed up with the trip logic test switches and various power

supply circuit breakers.

On this basis, the reactor manual scram pushbuttons alone do not need to meet the single-failure criterion, but manual initiation of reactor scram in a aggregate

sense does comply with this design requirement.

Reactor protection system motor-generator sets and power distribution

The two RPS M-G sets, the auxiliary power source to permit M-G set maintenance, and the RPS power distribution panel need not comply with the

single-failure criterion since loss of power at the interface produces a safe condition for the reactor, and the presence of power does not interfere with

normal protection action of the trip channels.

Reactor protection system trip logic, actuators, and trip actuator logic

Those portions of the RPS downstream of the trip channels comply with the design requirement.

Any postulated single failure of a given trip logic will not affect the remaining three trip logics. Similarly, any single failure of a trip actuator will not affect the

remaining trip actuators, and any single failure of one trip actuator logic will not

affect the other trip actuator logic networks. The cabling associated with one

trip logic is routed in conduit that is physically separated from similar cabling

associated with the other trip logics. Cabling from the trip actuator logic to the

scram solenoid fuse panels is routed in individual conduits to comply with this

design requirement. Since many individual control rods are wired from any

given scram solenoid fuse panel, individual conduits are used to cable each control rod hydraulic control unit. Since any individual control rod may fail to

operate from either the A or B solenoid valves, wiring of these two solenoids for

one control rod are routed together within a single conduit.

QUAD CITIES - UFSAR Revision 7, January 2003 7.2-40 Reactor protection systems outputs to other systems

The designated outputs from the RPS are designed so that no single failure in any portion of the RPS, including these output networks, can prevent proper protection system operation when it is required.

It is not necessary that the output networks meet the single-failure criterion in terms of their purpose, but it is essential that the outputs not compromise the

single-failure performance of the RPS in terms of its protective function. This

latter objective has been accomplished in the design of these output functions.

7.2.3.2 Quality of Components and Modules

The RPS components and modules are specified to withstand the transient and steady state

conditions of the environment (e.g., temperatures, humidity, pressure and vibration). The station's work control system data base identifies the classification of components.

[7.2-79]

The equipment for the following functions apply to the requirements of IEEE-279-1968

Quality of Components and Modules (paragraph 4.3) in that they were chosen to meet the

requirements of their intended functions.

[7.2-80]

Neutron monitoring system scram trip

Reactor vessel high pressure scram trip

Reactor vessel low water level scram trip

Turbine stop valve closure scram trip

Turbine control valve fast closure scram trip

Main steam line isolation valve closure scram trip

Scram discharge volume high water level scram trip

Primary containment high pressure scram trip

Manual scram pushbuttons

Reactor mode switch

Reactor protection system reset switch

Turbine stop valve closure and turbine control valve fast closure trip bypass

Neutron monitoring system trip bypass

Scram discharge volume high water level trip bypass

Main steam line isolation valve closure trip bypass QUAD CITIES - UFSAR 7.2-41 The remaining RPS equipment apply to the IEEE-279-1968 standard as indicated.

Reactor protection system motor-generator sets and power distribution

Cabling used within the RPS panels has been selected to be appropriate for RPS use. The RPS M-G sets have been chosen to provide low maintenance.

Reactor protection system trip logic, actuators, and trip actuator logic

The RPS trip logic consists of series-connected relay contacts from the trip channel output relays. The RPS trip actuator logic consists of relay contacts

connected in a specific arrangement from the trip actuators. Within the RPS

panels in the control room, electrical circuits are fused. Individual control rod

drive scram solenoids are fused at the scram solenoid fuse panels.

Reactor protection systems outputs to other systems

At the RPS interface with the output networks, isolated contacts of various RPS relays have been used to provide the signal source. These contacts are classified

as being a portion of the RPS component. The load device driven by these

contact outputs is not included in the RPS scope. The use of isolated contact

outputs from the RPS provides a large measure of isolation and independence for this interface relative to the protective action portions of the RPS.

Trip logic test switch

This design requirement is not applicable to this RPS test function.

For each of the RPS functions, the original equipment was required to be certified by the

vendor to meet the requirements listed in the purchase order, and for the intended

application described for that function. These certifications, in conjunction with applicable

field experience for those components in their particular applications, qualified the

components. In this way, the functions meet the requirements of IEEE-279-1968

Equipment Qualification (paragraph 4.4)

In addition to the vendor qualification, qualification tests of the relay panels were

conducted to confirm their adequacy for this application.

For RPS outputs to other systems, the RPS contact outputs from the designated relays were

qualified during the relay and panel tests. Qualification testing beyond this interface was

not contemplated.

This design requirement is not applicable to the trip logic test switch function.

Refer to Section 3.11 for information on the current environmental qualification program.

7.2.3.3 Channel Integrity

Safe shutdown of the reactor during earthquake ground motion is assured by the design of the system as a Class 1 system and the fail-safe characteristics of the system. The system QUAD CITIES - UFSAR Revision 3, December 1995 7.2-42 fails only in a manner that causes a reactor scram when subjected to extremes of vibration and shock.

[7.2-81]

The following text discusses the RPS functions as they apply to the requirements of IEEE-

279-1968 Channel Integrity (paragraph 4.5).

Except as otherwise noted, vendor certification was required that the RPS components would perform in accordance with the requirements listed on the purchase specifications as

well as in the intended applications.

[7.2-82]

Trip logic test switch

The trip logic test switch is not a trip channel component; rather, it is an element in the individual RPS trip logic strings.

Reactor protection system reset switch

The RPS reset switch is not a trip channel component; rather, its auxiliary relays are elements in the individual RPS trip logic strings.

Reactor protection systems outputs to other systems

Selection of output signals from the RPS to other systems has been done in such a manner to ensure that the integrity of the protection system channels remains

intact and unchanged.

This design requirement is not applicable to the reactor protection system motor-generator

sets and power distribution.

7.2.3.4 Channel Separation

Wiring for the RPS outside of the enclosures in the control room is run in enclosed conduits

throughout the plant and used for no other wiring. The wires from duplicate sensors on a

common process tap are run in separate conduits. Wires for sensors of different variables

in the same RPS trip logic may run in the same conduit. The RPS cables have channel

separation requirements which are maintained by the conduit system.

[7.2-83]

Low level signal cables are routed separately from all power cables with a minimum

separation of 3 feet wherever practical. Where the low level signal cable runs at right

angles to a power cable, a separation distance of less than 3 feet may be used, based upon

the probable noise pickup relative to the allowable signal-to-noise ratio.

Except as otherwise noted in the following discussions, the RPS trip, reset, and bypass channels are physically separated and electrically isolated to meet the design requirements

of IEEE-279-1968 Channel Independence (paragraph 4.6). Sections 7.2.2.7 and 7.2.3.1

discuss the specific separation methods used for these functions.

[7.2-84]

Manual scram pushbuttons

The manual scram pushbutton is not a channel component; nevertheless, the channels are separated in that the contacts from one switch are wired into the QUAD CITIES - UFSAR Revision 9, October 2007 7.2-43 A3 trip logic and the contacts of the second switch are wired into the B3 trip logic.

Trip logic test switch

While the test switch is not a trip channel component, it is imperative that its use in the RPS trip logic maintain the existing channel independent of the

automatic protective trip channels. The application of four test switches, one per

trip logic, ensures that this design requirement is satisfied.

Neutron monitoring system trip bypass

The neutron monitoring bypass channels comply with this design requirement.

The bypass channel output to the individual OPRM, APRM or IRM trip channel is obtained from an isolated relay contact. This contact output is physically

separated and electrically connected with the other bypass channels in order to

provide for one and only one bypass within one RPS trip system at any given

time; however, this cross connection does not invalidate the isolated contact from

each relay to the neutron monitoring system trip channel.

Scram discharge volume high water level trip bypass

The bypass circuitry complies with this design requirement. For operator convenience, a single switch has been selected for the bypass function. Factors

considered in this selection were the number of bypass operations required in

any given operating period and the expected duration of each bypass. Since the

bypass switch is used only to permit manual reset of the RPS and permit the

operator to drain the discharge volume following reactor scram, the switch will

be used infrequently and for short time periods. These considerations suggest

that a single switch is a better choice than multiple switches when viewed from

the operator's standpoint.

Care has been taken to assure that sufficient physical separation and electrical isolation exists to assure that the bypass channels are satisfactorily independent.

Moreover, the conditions for bypass have been made quite stringent in order to

provide additional margin.

Reactor protection systems outputs to other systems

Use of isolated relay contacts from the RPS relays assures that the RPS trip channels are maintained independent of one another. The design has considered

the effect of the output devices representing a potential point of common failure

for all trip channels, and steps have been incorporated into the system to prevent

this situation.

This design requirement is not applicable for the following equipment:

Reactor protection system motor-generator sets and power distribution

Reactor protection system trip logic, actuators, and trip actuator logic QUAD CITIES - UFSAR Revision 9, October 2007 7.2-44 7.2.3.5 Control and Protection System Interaction

Trip channels providing inputs to the RPS are not used for automatic control of process

systems; thus, the operations of protection and process systems are separated. Sensors, trip channels, and trip logics of the RPS are not used directly for automatic control of

process systems. Therefore, failure in the controls and instrumentation of process systems

cannot induce failure of any portion of the protection system.

[7.2-85]

Reactor protection system inputs to annunciators, recorders, and the computer are

arranged so that no malfunction of the annunciating, recording, or computing equipment can functionally disable the system. Signals directly from the RPS sensors are not used as

inputs to annunciating or data logging equipment. RPS inputs are addressed in Section

7.2.2.5.

The following text discusses the RPS functions as they apply to the requirements of IEEE-

279-1968 Control and Protection System Interaction (paragraph 4.7).

[7.2-86]

For the neutron monitoring system trip function, the IRM, APRM, and OPRM trip channels comply with this design requirement. Within the IRM and APRM modules, prior to their

output trip unit driving the RPS, analog outputs are derived for use with control room

meters, recorders, and the process computer. Electrical isolation has been incorporated into the design at this interface to prevent any signal failure from influencing the protective

output from the trip unit.

The trip channels for each of the remaining RPS trip functions comply with this design

requirement. Each trip channel output relay uses two contacts within the RPS trip logic.

One additional contact from each relay is wired to a common control room annunciator.

Another contact from each relay is wired to the process computer. Sections 7.2.2.7 and

7.2.3.1 discuss the specific separation methods used for these functions. Other interactions

or interfaces with the RPS functions are described as follows:

Scram discharge volume high water level scram trip

Two additional level switches are connected to the process taps. One level switch produces a control rod withdrawal block in the reactor manual control circuitry.

The other level switch produces a control room alarm that the discharge volume

is starting to fill. The only connection between these level switches with the four

protection system level switches is through the process medium at the taps.

Primary containment high pressure scram trip

One contact from each relay is wired to the primary containment isolation system to initiate protective isolation functions.

QUAD CITIES - UFSAR Revision 2, December 1993 7.2-45 Manual scram pushbuttons

Since the manual scram pushbutton is used only in the A3 and B3 RPS trip logic strings, there is no interaction with the control systems.

Reactor mode switch

The reactor mode switch is used for protective functions and restrictive interlocks on control rod withdrawal and refueling equipment movement.

Additional contacts of the mode switch are used to disable certain computer

inputs when the alarms would represent incorrect information for the operator.

No control functions are associated with the mode switch.

Trip logic test switch

Since this test switch is used only for the RPS and is located on the RPS panels, this design requirement is satisfied.

Reactor protection system reset switch

Switch contacts of the RPS reset switch are used only to control auxiliary relays, and contacts from the relays are used only in the trip actuator coil circuit.

Consequently, this RPS function has no interaction with any other system in the

plant.

Reactor protection system motor-generator sets and power distribution

The RPS M-G sets, power distribution panel, and cabling for the power distribution throughout the RPS cabinets have no interaction with any of the

control systems of the plant.

Reactor protection system trip logic, actuators, and trip actuator logic

The four RPS trip logic strings are totally separate from any other plant system. The RPS trip actuators utilize the power contacts of the scram contactors to

provide the trip actuators logic and the seal-in contact of the trip actuator, and utilize auxiliary contacts for control room annunciation, the process computer

inputs and initiation of the backup scram valves.

The trip actuator logic has no interaction with any other plant system, and the scram solenoids are physically separate and electrically isolated from the other

portions of the control rod drive hydraulic control unit.

Turbine stop valve closure, turbine control valve fast closure, and main steam isolation valve closure trip bypass

Two output relay contacts are used in the RPS trip logic, and one additional contact from each relay is used to initiate a control room annunciator for this

bypass function.

Neutron monitoring system trip bypass

In practice, each bypass channel consists of multiple relay coils in parallel with contacts from these relays used for different functions. From one relay, contact QUAD CITIES - UFSAR Revision 9, October 2007 7.2-46 outputs are used to provide an input to the process computer; from a second relay, contact outputs are used to provide control room annunciation of the bypass

condition; and for a third relay, contact outputs are used to bypass the neutron

monitoring system trip channels outputs. A similar configuration exists for bypass of the OPRM subsystem trips. From one relay, a contact is used to bypass the OPRM trip input to the RPS logic; and from a second relay, contact outputs are used to provide bypass status input to the OPRM as well as to Main Control Room annunciator and indicating light logic.

Scram discharge volume high water level trip bypass For each trip channel bypass relay, four contacts are used in the bypass logic. One contact of each relay is also wired to a common annunciator in the control room

and one contact of the A and B relays is wired to the control rod block circuitry to

prevent rod withdrawal whenever the trip channel bypass is in effect.

Reactor protection systems outputs to other systems Each output network has been investigated to determine the effects of postulated failures and to verify that these failures will not produce a control action that will

lead to a need for protective action and, at the same time, will not remove the

protection system capability to produce the required protective action.

7.2.3.6 Capability for Test and Calibration

Calibration and test controls for the neutron monitoring system are located in the control

room and are, because of their physical location, under the direct control of the control room

operator. Calibration and test controls for pressure transmitters, pressure switches, level

switches, and valve position switches are located on the switches themselves. These switches

are located in the turbine building, reactor building, and primary containment. Calibration

and test controls for the analog trip units associated with the transmitters are located on the Master Trip Units. The Master Trip Units are located in the Service Building Cable

Spreading Room.

[7.2-87]

The following text discusses the RPS functions as they apply to the requirements of IEEE-

279-1968 Capability for Test and Calibration (paragraph 4.10).

Neutron monitoring system scram trip The LPRMs provide inputs to the APRMs and must be calibrated before the APRMs. The LPRM gains are set using gain-adjustment-factors determined by the process computer nuclear calculations involving the reactor heat balance and the relative local flux distributions provided by the traversing incore probe (TIP)

system. [7.2-88]

The APRM gain-adjustment-factors are then determined using the reactor heat balance, and the gain of the APRM amplifiers are adjusted such that the APRMs

will reflect the fraction of power as calculated by the heat balance.

Each OPRM module may be calibrated with simulated signals introduced into the module utilizing the OPRM Maintenance Terminal. The maintenance terminal can also be used to perform manual testing of the OPRM modules. The OPRM automatically performs self-health tests and reports any detected failures of the individual hardware modules.

QUAD CITIES - UFSAR Revision 12, October 2013 7.2-47 Reactor vessel high pressure scram trip

Once a pressure sensor has been taken out-of-service, confirmation of the pressure setpoint can be made by use of a variable source of pressure or an analog signal. As the setpoint is exceeded, the control room operator will obtain

annunciation of the trip and computer record of the trip channel identification.

[7.2-89]

Reactor vessel low water level scram trip

During this calibration procedure, operation of the level sensor contacts can be confirmed relative to the indicated level scale reading of the instrument. The

relationship between indicated level and reactor vessel actual water level is

established by calibration of the instrument and the specific plant installation

detail. As a result, periodic calibration is accomplished relative to the indicated

water level.

Turbine stop valve closure scram trip

During reactor shutdown, calibration of the setpoint of the turbine stop valve limit switch at a valve position of 10% closure is possible by physical observation

of the valve stem.

Turbine control valve fast closure scram trip

During plant operation above the automatic bypass setpoint, one control valve at a time may be slowly closed through the normal servo control loop. As the

control valve approaches the closed position, the fast-acting solenoid is tripped to

cause rapid closure of the control valve for the remainder of its stroke. This

action causes the pressure switch input to the RPS to change to its tripped state

and provides a means of periodic testing of this interface.

Main steam line isolation valve closure scram trip

The main steam line isolation valve limit switches are mounted such that they are not adjustable. Calibration is therefore not required.

[7.2-90]

During reactor shutdown, the main steam line isolation valve limit switch setpoint at a valve position of 10% closure, is verified by physical observation of

the valve stem.

During plant operation, the operator can confirm limit switch operability during the periodic scram functional test.

Scram discharge volume high water level scram trip

The logic of the RPS permits the sensors to be removed from service one at a time and tested or calibrated.

[7.2-91]

QUAD CITIES - UFSAR Revision 7, January 2003 7.2-48 Primary containment high pressure scram trip

Once a pressure switch has been properly valved out-of-service, testing of the pressure switch and its setpoint may be performed using a variable source of pressure. When the trip setpoint has been exceeded, the control room operator

will obtain an annunciation of the trip and a typed record of the trip channel

identification from the process computer.

Manual scram pushbuttons

During reactor operation, one manual pushbutton may be depressed to test the proper operation of this switch, and once the RPS has been reset, the other

switch may be depressed to test its operation. For each such operation, a control

room annunciation will be initiated and the process computer will print the

identification of the pertinent trip.

Reactor mode switch

Operation of the reactor mode switch from one position to another may be employed to confirm certain aspects of the RPS trip channels during periodic test

and calibration. During tests of the trip channels, proper operation of the mode

switch contacts may be easily verified by noting that certain sensors are

connected into the RPS logic and that other sensors are disconnected from the RPS logic in an appropriate manner for the given position of the mode switch.

Reactor protection system reset switch

Operation of the reset switch following a trip of one RPS trip system will confirm that the switch is performing its intended function. Operation of the reset switch

following trip of both RPS trip systems will confirm that all portions of the

switch and relay logic are functioning properly since half of the control rods are

returned to a normal state for one actuation of the switch.

Reactor protection system trip logic, actuators, and trip actuator logic

The trip logic test switch permits each individual trip logic, trip actuator, and trip actuator logic to be tested on a periodic basis. Testing of each process sensor

of the protection system also affords an opportunity to verify proper operation of

these components.

Turbine stop valve closure and turbine control valve fast closure trip bypass

Administrative control is exercised to valve one pressure switch out-of-service for the periodic test. During this test, a variable pressure source may be introduced

to operate the switch at the setpoint value. When the condition for bypass has

been achieved on an individual sensor under test, the control room annunciator

for this bypass function will be initiated. If the RPS trip channel associated with

this sensor had been in its tripped state, the process computer would log the

return to normal state for the RPS trip logic. When the plant is QUAD CITIES - UFSAR 7.2-49 operating above the setpoint, testing of the turbine stop valve and control valve closure trip channels will confirm that the bypass function is not in effect.

[7.2-92]

Neutron monitoring system trip bypass Due to the discrete nature of this bypass function, the term calibration is not meaningful. However, proper operation of the bypass switches and associated logic

is possible by periodic testing of the possible combinations of bypass switch position

and neutron monitoring system trip channel status.

[7.2-93]

Scram discharge volume high water level trip bypass In the STARTUP/HOT STANDBY and RUN modes of plant operation, the preceding procedure may be used to confirm the trip channels are not bypassed as a

result of operation of the bypass switch. In the SHUTDOWN and REFUEL modes

of plant operation, a similar procedure may be utilized to produce bypassing of all

four trip channels. Due to the discreet nature of the bypass function, calibration is

not meaningful.

Main steam line isolation valve closure trip bypass Testing of the bypass circuit can only be accomplished when the mode switch is not in the RUN position. Hence, this test may be performed in the startup operating

phase.

Since it can be confirmed that the bypass is not in effect when operating in the RUN mode, the suggested test is adequate to confirm proper bypass status during plant

operation.

Reactor protection systems outputs to other systems The output functions provided to the annunciator and process computer systems aid the operator in the RPS periodic testing process. There is no requirement for the

output functions themselves to be subject to periodic testing since they represent an

information source rather than a protective function.

Exceptions This design requirement is not applicable to the following equipment:

Trip logic test switch

Reactor protection system motor-generator sets and power distribution

7.2.3.7 Establishment of Trip Setpoints Initially, conservative trip settings were selected so that they were far enough above or below

normal operating levels that spurious scrams and operating inconvenience were avoided.

Analyses were performed using trip settings as preliminary inputs or conditions to verify that

the reactor fuel and nuclear system process barrier were protected in accordance with the system design intent. In all cases, the specific scram trip point was not selected solely on the

value of the trip point that results in no damage to the fuel or QUAD CITIES - UFSAR Revision 9, October 2007 7.2-50 nuclear system process barrier but was selected based on operating experience and safety design basis constraints. The current methodology used to established the Technical

Specification allowable values and the associated instrument trip setpoints is described in

Section 7.1.2.1.

[7.2-94]

Multiple setpoints are used where it is necessary to provide more restrictive reactor

protection limits due to the mode of operation or operating conditions. The following text

discusses the RPS functions as they apply to IEEE-279-1968 Multiple Setpoints (paragraph

4.15). [7.2-95]

Neutron monitoring system scram trip The trip setpoint of each IRM channel is established near the full scale mark for each range of IRM operation. As the operator switches an IRM from one range to the next, the trip setpoint tracks the operator's selection. With the reactor mode

switch in STARTUP/HOT STANDBY, the IRM trips are enabled and the APRM

trips are fixed at the low power setpoint.

In the transition from STARTUP/HOT STANDBY to RUN mode of operation, the reactor mode switch is used to convert from IRM protection to APRM protection.

In RUN, the APRM trip setpoint is raised to a flow-biased value and the IRM trips are essentially bypassed (i.e., the corresponding APRM must indicate

downscale for the IRM trip to be recognized).

The OPRM does not have multiple setpoints to accommodate different operating conditions. However, the OPRM trip function is disabled unless manual action is taken to enable it or the OPRM automatically enables itself upon detection of entry into the high power, low core flow region of the power/flow operating map where there is a potential for instabilities. Each of these multiple setpoint provisions is a portion of the reactor protection system and complies with the design requirements of IEEE-279.

Reactor mode switch Operation of the mode switch from one position to another imposes different RPS trip channels into the RPS logic. This action does not influence the established

setpoint of any given RPS trip channel, but merely connects one set of channels

as another set are disconnected. Consequently, the mode switch meets this

design requirement.

Neutron monitoring system trip bypass Due to the different ranges of operation of the IRM and APRM systems, the four neutron monitoring system bypass switches are designated so that they

correspond with those two different neutron monitoring system equipments. For

any given bypass switch, multiple setpoints are not provided in the design.

This design requirement is not applicable to the following functions and equipment: Reactor vessel high pressure scram trip Reactor vessel low water level scram trip Turbine stop valve closure scram trip Turbine control valve fast closure scram trip Main steam line isolation valve closure scram trip QUAD CITIES - UFSAR Revision 7, January 2003 7.2-51 Scram discharge volume high water level scram trip

Primary containment high pressure scram trip

Manual scram pushbuttons

Trip logic test switch

Reactor protection system reset switch

Reactor protection system motor-generator sets and power distribution

Reactor protection system trip logic, actuators, and trip actuator logic

Turbine stop valve closure and turbine control valve fast closure trip bypass

Scram discharge volume high water level trip bypass

Main steam line isolation valve closure trip bypass

Reactor protection systems outputs to other systems

7.2.3.8 Access to Setpoint Adjustments, Calibration, and Test Points

Administrative controls are used as the basis for assuring that access to Setpoint

Adjustments, Calibrations, and Test Points are limited to qualified, plant personnel and

that permission of Operations is obtained to gain access.

[7.2-96]

The following text covers the RPS functions as they apply to the requirements of IEEE-279-

1968 Access to Setpoint Adjustments, Calibration, and Test Points (paragraph 4.18).

[7.2-97]

Access to setpoints and calibration controls are under the administrative control of

operating personnel for the following RPS functions.

Neutron monitoring system scram trip

Reactor vessel high pressure scram trip

Reactor vessel low water level scram trip

Primary containment high pressure scram trip

Turbine stop valve closure and turbine control valve fast closure trip bypass

Neutron monitoring system trip bypass QUAD CITIES - UFSAR Revision 5, June 1999 7.2-52 Access to the turbine stop valve closure scram trip and the main steam line isolation valve closure scram trip process limit switch inputs is not anticipated during reactor operation

due to ambient environmental conditions. The reactor operator is permitted full access to the valve test controls for the turbine stop valve closure, main steam line isolation valve

closure, and turbine control valve fast closure scram trip functions since motion of the valve

during this test produces a valid process sensor response.

This design requirement is not applicable to the following functions.

Scram discharge volume high water level scram trip

Manual scram pushbuttons

Reactor mode switch

Trip logic test switch

Reactor protection system reset switch

Reactor protection system motor-generator sets and power distribution

Reactor protection system trip logic, actuators, and trip actuator logic

Scram discharge volume high water level trip bypass

Main steam line isolation valve closure trip bypass

Reactor protection system outputs to other systems

7.2.3.9 Identification of Protection Systems

The RPS and engineered safety equipment are physically identified per their function.

Identification of trays, conduits and junction boxes is by means of stencil or adhesive

markers. Control room panels, local panels, and racks are identified by engraved

nameplates. Electrical panels, junction boxes, and components of the RPS are prominently

identified by nameplates. Circuits entering junction or pull boxes are marked inside the

boxes. Wiring and cabling outside cabinets and panels are identified by color, tag or other

conspicuous means.

[7.2-98]

In addition, the operators and instrument mechanics that work with and maintain this

equipment are trained in its identification and use. Normal plant operating procedures require that the Shift Manager or Unit Supervisor on duty authorize the performance of all

work on these RPS components. The station out-of-service card procedure is used whenever

systems are taken out of service for maintenance.

7.2.3.10 System Repair

The design of the following components, functions, and systems complies with the

IEEE-279-1968 System Repair design requirement (paragraph 4.21).

[7.2-99]

QUAD CITIES - UFSAR Revision 6, October 2001 7.2-53 Reactor mode switch

Trip logic test switch

Reactor protection system reset switch

Reactor protection system motor-generator sets and power distribution

Reactor protection system trip logic, actuators, and trip actuator logic

Neutron monitoring system trip bypass

Scram discharge volume high water level trip bypass

Main steam line isolation valve closure trip bypass

Conformance of other RPS functions to IEEE-279-1968 System Repair requirements are as

follows:

Neutron monitoring system scram trip

Replacement of IRM and LPRM detectors must be accomplished during plant shutdown. Repair of the remaining portions of the neutron monitoring system

may be accomplished during plant operation by appropriate bypassing of the

defective trip channel output. The design of the system facilitates rapid

diagnosis and repair.

Reactor vessel high pressure scram trip

Due to the one-to-one relationship of pressure sensor and trip channel output relay, this design requirement is satisfied for this protective function.

[7.2-100]

Reactor vessel low water level scram trip

The one-to-one relationship between a level sensor and a trip channel output relay permits the plant personnel to identify any component failure during operation of the plant. Provisions have been made to facilitate repair of the

channel components during plant operation.

Turbine stop valve closure scram trip

Because of the inherent simplicity of the valve limit switch for the process sensor, and the relationship of one limit switch contact with one trip channel output relay, the design of the system facilitates maintenance of this protective

function.

During power operation, it may be necessary to reduce power in order to close more than one turbine stop valve in order to accomplish a specific RPS test. The sequence of tests should permit the operator to determine a defective limit switch

contact or trip channel output relay.

QUAD CITIES - UFSAR Revision 7, January 2003 7.2-54 Turbine control valve fast closure scram trip

Periodic tests of portions of this protective function during plant operation will likely require a temporary reduction in plant output and may be accomplished

with the provisions for testing of the turbine equipment.

Main steam line isolation valve closure scram trip

Due to the inherent simplicity of the valve limit switch for the process sensor, and the relationship of one limit switch contact with one trip channel output

relay, the design of the system facilitates maintenance of this protective function.

During power operation, it may be necessary to reduce power in order to close valves in more than one main steam line. With this arrangement, a sequence of valve tests will permit the operator to determine fully a defective component or

isolate the difficulty to one of two limit switches in a given main steam line.

Scram discharge volume high water level scram trip

Because the water level measurement and its one-to-one relationship between a given level sensor and its associated trip channel output relay are inherently

simple, the design facilitates maintenance of this protective function.

Primary containment high pressure scram trip

Due to the one-to-one relationship of pressure switch and trip channel output relay, this design requirement is satisfied by this protective function.

Manual scram pushbuttons

Due to the simplicity of the manual scram function, the design complies with this requirement.

Reactor protection systems outputs to other systems

The design of these networks facilitates repair of the RPS by providing timely information readout and identification of failures for the operating personnel.

QUAD CITIES - UFSAR Revision 7, January 2003 7.2-55 7.2.4 References

1. Appendix A of NRC Standard Review Plan (SRP) NUREG 800, Rev. 2, July 1981.

2.General Electric Topical Report NEDO-10139, "Compliance of Protection Systems to Industry Criteria: General Electric BWR Nuclear Steam Supply System," June, 1970.

3.General Electric Safety Evaluation Report NEDO-31400A "Safety Evaluation for Eliminating the BWR Main Steam Isolation Valve Closure and Scram Function of the

Main Steam Line Radiation Monitor," October 1992.

4.DCP 9900185, Unit 1 MSL Rad Monitor Scram and Group 1 Isolation Trip Function Removal.

5.EC 23949 (DCP 9900184), Unit 2 MSL Rad Monitor Scram and Group 1 Isolation Trip Function Removals.

(Sheet 1 of 2)

Revision 9, October 2007 QUAD CITIES - UFSAR

Table 7.2-1

ANALYTICAL LIMITS FOR REACTOR PROTECTION SETPOINTS

Initiating Conditions Analytical Limit [Note 1]

1. Reactor neutron
a. APRM high-high flux (flow biased) (RUN mode) < 0.56W D + 71% RTP

[Note 2] b. APRM fixed neutron flux-high < 125% c. APRM inoperative - d. APRM downscale with companion IRM high-high (RUN mode) APRM > 1% power IRM < 125/125 e. APRM high-high flux (bypassed in RUN mode)

< 20% power f. IRM high-high flux (bypassed in RUN mode with

APRM upscale) IRM < 125/125 APRM > 1% power g. IRM inoperative (bypassed in RUN mode with APRM

upscale) - APRM > 1% power h. SRM high flux (bypassed when shorting links

installed) 1 x 10 6 cps i. Flux oscillation

2. Reactor high pressure See UFSAR Table 7.6-1

< 1060 psig

3. Reactor low water level

> 0 inches

4. Turbine stop valve closure (RUN mode >38.5% RTP)

< 10% closure

5. Turbine control valve fast closure, valve trip system oil pressure low(RUN mode >38.5%

RTP) > 460 psig [Note 3]

6. MSIV closure (RUN mode)

< 10% closure (Sheet 2 of 2)

Revision 10, October 2009 QUAD CITIES - UFSAR

Table 7.2-1

ANALYTICAL LIMITS FOR REACTOR PROTECTION SETPOINTS

Initiating Conditions Analytical Limit [Note 1]

7. High scram discharge volume water level

< 40 gallons

8. Primary containment (drywell) high pressure

< 2.5 psig

9. Turbine condenser low vacuum

> 20 inches Hg

10. Mode switch in SHUTDOWN (auto reset after 10 seconds)

-

Note 1 Analytical Limit shown unless noted otherwise. Consult Technical Specifications for associated 'Allowable Value'.

Note 2 W D is the percent of drive flow required to produce a rated core flow of 98 million lb/hr.

Note 3 Trip is indicative of turbine control valve fast closure (due to low EHC fluid pressure) as a result of fast acting valve actuation.

QUAD CITIES - UFSAR 7.3-1 Revision 7, January 2003 7.3 ENGINEERED SAFETY FEATURE SYSTEMS INSTRUMENTATION AND CONTROL

The engineered safety feature (ESF) systems are provided to mitigate the consequences of

postulated accidents. The ESF systems described in this section are not used during

normal plant operations. These systems must, however, be operable as defined in the

Technical Specifications.

[7.3-1]

The ESF systems addressed in this section include the following:

A. Emergency core cooling systems (ECCS):

1. Core spray system;
2. Low pressure coolant injection (LPCI) mode of the residual heat removal (RHR) system;
3. High pressure coolant injection (HPCI) system; and
4. Automatic depressurization system (ADS).

B. Containment isolation systems:

1. Primary containment isolation system (PCIS); and
2. Secondary containment isolation.

7.3.1 Emergency Core Cooling Systems Instrumentation and Control

Refer to Section 6.3 for ECCS design bases and description.

7.3.1.1 Core Spray System Instrumentation and Control

The control system is arranged to provide two independent and separately isolated control

and power circuits for the operation of the two independent, 100% capacity core spray loops.

(Refer to Figure 6.3-5).

There are three primary initiation or permissive signals related to operation of the core

spray system. These signals are generated by the following sensors:

[7.3-2]

A. Four independent low-low reactor water level transmitters and trip units;

B. Four independent high drywell pressure switches; and

C. Two low reactor pressure switches using different operating principles.

QUAD CITIES - UFSAR Revision 8, October 2005 7.3-2 The core spray initiation signal requires any one of the following logic combinations:

A. Low-low reactor water level (one-out-of-two-twice) coincident with low reactor pressure (one-out-of-two);

B. High drywell pressure (one-out-of-two-twice); or

C. Low-low reactor water level (two-out-of-two in the corresponding division) continuously for 9 minutes (analytical limit). This signal is generated by the

ADS system logic.

The core spray initiation signal starts the core spray pumps, opens the suction valves (if

closed), and closes the test bypass valves (if open).

The permissive signal which opens the core spray injection (discharge) valves, requires a (one-out-of-two) low reactor pressure signal in addition to the core spray initiation signal.

With normal auxiliary ac power available the actions described above occur automatically

without delay. A diesel generator start signal is also generated by either a low-low reactor water level signal or high drywell pressure signal (both one-out-of-two-twice). If normal ac

power is not available the pumps are started sequentially as described in Section 6.3.

7.3.1.1.1 Conformance to IEEE-279

The following is a point-by-point comparison of the core spray system with the

requirements of proposed IEEE Std 279-1968 which has been summarized from GE Topical

Report, NEDO-10139.[1] For more detailed information refer to the topical report. The GE topical report is a generic report for an entire product line and these statements should be used concurrently with design requirements described in other sections of the UFSAR (such as Section 3.5 for Missile Protection, 3.6 for Pipe Break, etc.) that represent Quad Cities specific design requirements.

[7.3-3]

7.3.1.1.1.1 General Functional Requirement (IEEE-279, Paragraph 4.1)

The following summarizes the general functional requirements of IEEE-279 and the

provision of the core spray system in fulfillment of these requirements.

A. Auto-Initiation of Appropriate Action

Appropriate action for the core spray control system is defined as the activation of equipment for introducing low pressure water through the core spray sparger

when reactor vessel level drops below a predetermined point or the drywell

pressure increases above a predetermined value, and the vessel pressure is below

a predetermined value lower than the pump shutoff head. This action occurs

automatically.

B. Precision

The sensory equipment positively initiates action before process variables go beyond precisely established limits. In the case of vessel level sensors, high

drywell ambient temperature can introduce errors that will lower the trip point

for starting of the core spray pumps. Errors that result from drywell

temperatures less than the temperature that causes a high drywell pressure trip

are not large enough to be objectionable from a safety point of view.

QUAD CITIES - UFSAR 7.3-3 Revision 7, January 2003 C. Reliability

Reliability of the control system is commensurate with the controlled equipment so that the overall system reliability is not limited by the controls.

D. Action Over the Full Range of Environmental Conditions

Refer to Section 3.11 for information on the current environmental qualification program.

7.3.1.1.1.2 Single Failure Criterion (IEEE-279, paragraph 4.2)

The core spray system, comprised of two independent sets of controls for the two physically

separate pumping systems, meets all credible aspects of the single failure criterion.

7.3.1.1.1.3 Quality of Components (IEEE-279, paragraph 4.3)

Components used in the core spray control system have been carefully selected on the basis

of suitability for the specific application. All of the sensors and logic relays are of the same

types used in the reactor protection system (RPS) described in Section 7.2. Ratings have

been selected with sufficient conservatism to insure against significant deterioration during

anticipated duty over the lifetime of the plant.

7.3.1.1.1.4 Equipment Qualification (IEEE-279, paragraph 4.4)

No components of the core spray control system are required to operate in the drywell

environment with the exception of the temperature compensating columns for the vessel

level sensors. These columns are calibrated for a specific normal ambient temperature and can introduce nominal errors under steam leak (high drywell temperature) conditions (see paragraph 4.1). All other sensory equipment is located in the reactor building outside the drywell and is capable of accurate operation with wider swings in ambient temperature

than results from normal or abnormal (loss of ventilation and LOCA) conditions.

All components used in the core spray control system have demonstrated reliable operation

in similar nuclear power plant protection system or industrial applications.

Refer to Section 3.11 for information on the current environmental qualification program.

7.3.1.1.1.5 Channel Integrity (IEEE-279, paragraph 4.5)

The core spray control system is designed to tolerate the spectrum of failures listed under

the general requirements and the single failure criteria. Each of the two core spray systems sensors are backed up by sensors from the other so neither system alone loses its

integrity because of a failure or failures in its sensory equipment.

QUAD CITIES - UFSAR Revision 7, January 2003 7.3-4 The core spray system control backup has been achieved without compromising the integrity of the channel being backed up because it can be shown by analysis that complete destruction of a wireway (conduit) carrying wi res between the two relay cabinets cannot prevent operation of both core spray loops.

During a DBA, the control system environment does not differ significantly from normal.

7.3.1.1.1.6 Channel Independence (IEEE-279, paragraph 4.6)

Channel independence of the sensors for each variable is provided by electrical isolation and mechanical separation. The A and C sensor s for reactor vessel level are located on a stanchion adjacent to the Division I instrume nt rack, and the B and D sensors are located on a pair of stanchions adjacent to the Division II instrument rack. The A and C sensors have a common process tap, which is widely separated from the corresponding tap for sensors B and D. Disabling of one or all sensors at one location does not disable the control

for either of the two core spray loops, or two separate divisions of LPCI.

Relay cabinets for core spray system A are in a separate physical division from that for core spray system B, and each division is complete in itself, with its own station battery control and instrument bus, power distribution buses, and motor control centers. The divisional split is carried all the way from the process ta ps to the final control element, and includes both control and motive power supplies.

7.3.1.1.1.7 Control and Protection Interaction (IEEE-279, paragraph 4.7)

The core spray system is strictly an off-on syst em, and no signal whose failure could cause need of core spray can also prevent core sp ray from starting. Annunciator circuits using contacts of sensor relays and basic relays cannot impair the operability of the core spray system control because of the electrical separa tion between controls of the two systems.

7.3.1.1.1.8 Derivation of System Inputs (IEEE-279, paragraph 4.8)

The inputs which start the core spray system are direct measures of the variables that indicate the need for low pressure core coo ling; such as reactor vessel low water, high drywell pressure, and reactor lo w pressure. Reactor vessel leve l is sensed by vessel water level transmitters and trip units. Drywell high pressure is sensed by nonindicating pressure switches on four separate sensing lines connected to two se parate penetrations.

Each sensing line has its own root valve, and each pressure switch has its own instrument valve. Two reactor vessel pressure switches for the low pressure injection valve opening permissive are on two separate instrument lines going through the drywell at two different general locations. These switches operate re lays whose contacts are connected in A or B logic for the core spray va lve opening permissives.

QUAD CITIES - UFSAR 7.3-5 7.3.1.1.1.9 Capability for Sensor Checks (IEEE-279, paragraph 4.9)

All sensors are of the pressure sensing type and are installed with calibration taps and

instrument valves, to permit testing during normal plant operation or during shutdown.

7.3.1.1.1.10 Capability for Test and Calibration (IEEE-279, paragraph 4.10)

The core spray control system is capable of being completely tested during normal plant operation to verify that each element of the system, active or passive, is capable of

performing its intended function.

7.3.1.1.1.11 Channel Bypass or Removal from Operation (IEEE-279, paragraph 4.11)

Calibration of each sensor will introduce a single instrument channel trip. This does not

cause a protective function without coincident operation of a second channel. Removal of an

instrument channel from service during calibration is brief and in compliance with special

provision of IEEE-279, paragraph 4.11 for one-out-of-two-twice systems.

7.3.1.1.1.12 Operating Bypasses (IEEE-279, paragraph 4.12)

There are no operating bypasses for the core spray system.

7.3.1.1.1.13 Indication of Bypasses (IEEE-279, paragraph 4.13)

There are no automatic bypasses of any part of the core spray control system.

7.3.1.1.1.14 Access to Means for Bypassing (IEEE-279, paragraph 4.14)

Access to switchgear, motor control centers, and instrument valves is procedurally

controlled.

7.3.1.1.1.15 Multiple Trip Settings (IEEE-279, paragraph 4.15)

Paragraph 4.15 of IEEE-279 is not applicable because all setpoints are fixed.

7.3.1.1.1.16 Completion of Protection Action Once Initiated (IEEE-279, paragraph 4.16)

The final control elements for the core spray system are essentially bistable; that is, pump

breakers stay closed without control power, and motor operated valves stay open once they QUAD CITIES - UFSAR Revision 7, January 2003 7.3-6 have reached their open position, even though the motor starter may drop out (which will occur when the valve open limit switch is reached). In the event of an interruption in ac power, the control system will reset itself and recycle on restoration of power. Thus protective

action once initiated must go to completion or continue until terminated by deliberate operator action.

7.3.1.1.1.17 Manual Actuation (IEEE-279, paragraph 4.17)

Each piece of core spray actuation equipment (pump, valve, breaker, and starter) is capable of

individual manual initiation, electrically from the control panel in the main control room and

locally, if desired, by use of physical mechanisms. The valves have handwheels for manual

operation, and the switchgear is capable of having closing springs charged manually and the

breaker closed by mechanical linkages on the switchgear.

In no event can failure of an automatic control circuit for one core spray loop disable the

manual electrical control circuit for the other core spray loop. Single electrical failures cannot

disable manual electric control of the core spray function.

7.3.1.1.1.18 Access to Setpoint Adjustment (IEEE-279, paragraph 4.18)

Administrative controls are used as the basis for assuring that access to core spray Setpoint

Adjustments, Calibrations, and Test Points are limited to qualified, plant personnel and that

permission of Operations is obtained to gain access. The range of the drywell and reactor

vessel pressure switches is not adjustable. The reactor vessel level transmitters have zero and span adjustments that are external to the transmitters but require removal of the nameplate to gain access. Because of these restrictions, compliance with the access requirements of IEEE-279 is considered complete.

7.3.1.1.1.19 Identification of Protective Actions (IEEE-279, paragraph 4.19)

Protective actions (here interpreted to mean pickup of a single sensor relay) are directly

indicated and identified by action of the sensor relay, which has an identification tag and a

clear glass front window permitting convenient, visible verification of the relay position. Any

one of the sensor relays also actuates an annunciator, so that no single-channel trip (relay

pickup) will go unnoticed. Either of these indications should be adequate, so this combination

of annunciation and visible verification relay actuation fulfills the requirements of this

criterion. In addition, indicator lights are provided to show pickup of sensor relays.

7.3.1.1.1.20 Information Readout (IEEE-279, paragraph 4.20)

The core spray control system is designed to provide the operator with accurate and timely

information pertinent to its status. It does not introduce signals into other systems that could

cause anomalous indications confusing to the operator. There are many passive as well as

active elements of this energize-to-operate system which are not continuously QUAD CITIES - UFSAR Revision 7, January 2003 7.3-7 monitored for operability. Examples are circuits which are normally open and are not monitored for continuity on a continuous basis, pressure and level sensors, which, although

continuously active, are not continuously exercised and verified as operable. However, ATS alarms provide warning for loss of power or gross failure of electronic card circuits associated with reactor vessel level sensors. Verifying the operability of these components is accomplished by periodic testing and by proper selection of test periods to be compatible

with the historically established reliability of the components tested. Sufficient information

is provided on a continuous basis so that the operator can have a high degree of confidence

that the core spray function is available and operating properly.

7.3.1.1.1.21 System Repair (IEEE-279, paragraph 4.21)

The core spray control system is designed to avoid a need for repair rather than for fast

replacement of components. Thus, reliability is built-in rather than approached by rapid

return-to-service maintenance. All devices in the system are designed for a 40-year lifetime under the imposed duty cycles. Since this duty cycle is composed mainly of periodic testing

rather than operation, lifetime is more a matter of shelf life than active life. However, all

components are selected for continuous duty plus thousands of cycles of operation, far

beyond that anticipated in actual service. The pump breakers are an exception to this with

regard to the large number of operating cycles available. Nevertheless, even these breakers

should not require contact replacement within 40 years, assuming periodic pump starts

every 3 months.

7.3.1.1.2 Failure Mode and Effects-Analysis Summary

No single component cable, wireway, or cabinet failure can disable the core spray function.

Therefore, the core spray system is considered to have fully met the single failure criterion

of IEEE-279.

7.3.1.2 RHR System LPCI Mode Instrumentation and Controls

The residual heat removal (RHR) system can be operated in any one of three modes: Low

pressure coolant injection , containment cooling, and reactor shutdown cooling. Low

pressure coolant injection and containment cooling are primarily safety functions. The

LPCI mode instrumentation and control is described in this section. Containment cooling is addressed in Section 6.2, and reactor shutdown cooling in Section 5.4.

[7.3-4]

In general, LPCI operation involves restoring and maintaining the water level in the

reactor vessel at a sufficient level for adequate cooling after a loss-of-coolant accident (LOCA). The LPCI initiation logic system operates in conjunction with HPCI, ADS and

core spray logic.

[7.3-5]

Initiation of LPCI occurs on signals indicating low-low reactor water level coincident with reactor low pressure, high drywell pressure, or low-low reactor water level continuously for

9 minutes (analytical limit). Low-low reactor water level and high drywell pressure are

each detected by four independent level transmitters and pressure switches connected in a one-out-of-two-twice logic. Reactor low pressure is detected by two independent pressure

switches, each of a different design principle. The switches are connected in a one-out-of-

two logic. Upon receipt of an initiation signal with normal ac power available the:

[7.3-6]

QUAD CITIES - UFSAR Revision 8, October 2005 7.3-8 1. Permissive becomes available to activate pumps and valves,

2. All four RHR pumps start,
3. RHR service water pumps stop (if running).

If normal ac power is not available, pumps are started sequentially as described in Section

6.3. For a description of LPCI's interaction with shutdown cooling refer to Section 6.3.

Prior to opening of the admission valves, it is necessary that sufficient information be

available to determine if the break has occurred in a recirculation loop, and if so, which

loop. If neither loop is broken, a preselected loop will be used for injection. This selection is

necessary because LPCI injects through the recirculation loops.

The system makes the loop selection by comparing the pressure in the five riser pipes on

one recirculation loop with the pressure in the corresponding riser pipes on the other

recirculation loop. A schematic of the instrument arrangement is shown in Figure 6.3-12.

The unbroken recirculation loop will have a higher pressure than the broken loop. Two

differential pressure instruments indicating a higher pressure in one loop than in the other (in a one-out-of-two-twice arrangement) cause LPCI flow to be injected into the higher

pressure loop.

The break detection logic arrangement is shown in Figure 6.3-13. As shown, the logic is

actuated by high drywell pressure or low-low reactor water level.

7.3.1.2.1 Conformance with IEEE-279

The following is a point-by-point comparison of the LPCI system with the requirements of

proposed IEEE Std 279-1968 which has been summarized from GE Topical Report, NEDO-

10139.[1] For more detailed information, refer to the topical report. The GE topical report is a generic report for an entire product line and these statements should be used concurrently with design requirements described in other sections of the UFSAR (such as Section 3.5 for Missile Protection, 3.6 for Pipe Break, etc.) that represent Quad Cities specific design requirements.

[7.3-7]

The low pressure core cooling system consists of three loops: core spray system loop A, core

spray system loop B, and the LPCI system. Therefore, it should be made clear that the

LPCI system by itself is not required to meet all the requirements of IEEE-279 since it is

backed up by the two core spray systems. The following comparison is provided only to

show the adequacy of the LPCI system design.

7.3.1.2.1.1 General Functional Requirement (IEEE-279, paragraph 4.1)

A. Auto-Initiation of Appropriate Action

Appropriate action for the LPCI control system is defined as the activation of equipment for introducing low pressure water into the reactor via the

recirculation line when reactor vessel level drops below a predetermined point, or

the drywell pressure increases above a predetermined value and reactor vessel

pressure is below the pump shutoff head. This action occurs automatically.

QUAD CITIES - UFSAR 7.3-9 Revision 7, January 2003 B. Precision

See Section 7.3.1.1.1.1 which applies equally to the LPCI and core spray systems.

Sensors which initiate the core spray system are the same sensors as used to

initiate the LPCI system. However, reactor vessel low level initiation is provided by separate slave trip units and trip relays.

C. Reliability

Reliability of the control system is commensurate with the controlled equipment so that the overall system reliability is not limited by the controls.

D. Action Over the Full Range of Environmental Conditions

Refer to Section 3.11 for information on the current environmental qualification program.

7.3.1.2.1.2 Single-Failure Criterion (IEEE-279, paragraph 4.2)

The LPCI system is a single system in that water is injected into the reactor via a single

injection valve. Therefore, the LPCI system is not required (in itself) to meet the intent of the single-failure criterion. However, redundancy in equipment and control logic circuitry

is provided so that it is highly unlikely that the complete LPCI system can be rendered

inoperative.

Two control logic circuits are provided. Control logic A is provided to initiate loop A pumps

and valves and logic B is provided to initiate loop B equipment. This does not apply to the

initiation of the injection valves.

Tolerance to single failures or events is provided in the control logic initiation circuitry so

that these failures will be limited to the possible disabling of the initiation of only one loop (two of four pumps available).

The LPCI system is designed to detect the location of a recirculation line break and select

the unbroken loop for injection. The sensing circuit for break detection and valve selection

is arranged so that failure of a single device or circuit to function on demand will not

prevent selection of the correct loop for injection. Tolerance to the following single failures

or events has been incorporated into the loop selection control system design.

A. Single open circuit, B. Single relay failure to pickup, C. Single relay failure to dropout, D. Single instrument failure, and E. Single control power failure.

Reliability of the control system is compatible with and more reliable than the controlled

equipment (injection valve). It should be made clear that those single failures which could

cause improper loop selection (that is, selected short circuits which pickup specific relays)

will not disable the core spray function. Therefore it is concluded that failure of the loop

selection scheme to fully comply with the single-failure criterion of IEEE-279 paragraph QUAD CITIES - UFSAR 7.3-10 4.2 does not constitute a violation of IEEE-279 insofar as the low pressure cooling function is concerned.

7.3.1.2.1.3 Quality of Components (IEEE-279, paragraph 4.3)

See Section 7.3.1.1.1.3 which also applies generally to the LPCI system.

7.3.1.2.1.4 Equipment Qualification (IEEE-279, paragraph 4.4)

See Section 7.3.1.1.1.4 which also applies to the LPCI system.

7.3.1.2.1.5 Channel Integrity (IEEE-279, paragraph 4.5)

The LPCI system initiation channels (low water level or high drywell pressure) are

designed to meet the single failure criterion as discussed in Section 7.3.1.2.1.1 and

7.3.1.2.1.2 and thus satisfies the channel integrity objective of this paragraph.

The instrumentation provided for the loop selection logic does not initiate a protective

action and therefore this paragraph does not strictly apply to this instrumentation.

However, as previously described, redundancy in instrumentation and control logic circuits

have been provided so that is extremely unlikely that a failure within this functional logic

will prevent proper LPCI operation.

7.3.1.2.1.6 Channel Independence (IEEE-279, paragraph 4.6)

See Section 7.3.1.1.1.6 which also applies to the LPCI system. By definition (IEEE-279

paragraph 2.2) a channel loses its identity where single action signals are combined.

Therefore, since instrument channels are combined into a pair of single logic channel trip

systems this paragraph of IEEE-279 does not strictly apply for the loop selection logic.

7.3.1.2.1.7 Control and Protection Interaction (IEEE-279, paragraph 4.2)

See Section 7.3.1.1.1.7 which also applies to the LPCI system.

7.3.1.2.1.8 Derivation of System Inputs (IEEE-279, paragraph 4.8)

See Section 7.3.1.1.1.8 which also applies to the LPCI system. The inputs provided to

determine which loop should be used for LPCI injection are direct measures of the variables

required to make this decision.

QUAD CITIES - UFSAR Revision 5, June 1999 7.3-11 7.3.1.2.1.9 Capability for Sensor Checks (IEEE-279, paragraph 4.9)

See Section 7.3.1.1.1.9 which also applies to the LPCI system.

7.3.1.2.1.10 Capability for Test and Calibration (IEEE-279, paragraph 4.11)

See Section 7.3.1.1.1.10 which also applies to the LPCI system except as stated below. The

only portion of the LPCI logic which cannot be tested with the reactor at full power is the

recirculation pump trip portion of the loop selection logic.

7.3.1.2.1.11 Channel Bypass or Removal from Operation (IEEE-279, paragraph 4.11)

See Section 7.3.1.1.1.11 which also applies to the LPCI system.

7.3.1.2.1.12 Operating Bypasses (IEEE-279, paragraph 4.12)

A. Manual Bypasses

See Section 7.3.1.1.1.12 which also applies to the LPCI system.

B. Automatic Bypasses

The only automatic bypass of the LPCI system is the closure of the LPCI inboard injection valve on an isolation signal during the RHR shutdown cooling mode.

Indication of this is provided by an indicating light in the main control room.

7.3.1.2.1.13 Indication of Bypasses (IEEE-279, paragraph 4.13)

Indication of bypasses provided is as discussed in Section 7.3.1.2.1.12 above, and as

described in the core spray system Section 7.3.1.1.1.13.

7.3.1.2.1.14 Access to Means for Bypassing (IEEE-279, paragraph 4.14)

Access to switchgear, motor control center, and instrument valves is controlled as discussed

in Section 7.3.1.1.1.14. Access to other means of bypassing (that is, closure of pump suction

valves by means of a keylock switch) are located in the main control room and, therefore, under the administrative control of the operator.

QUAD CITIES - UFSAR 7.3-12 7.3.1.2.1.15 Multiple Trip Settings (IEEE-279, paragraph 4.15)

This is not applicable because all setpoints are fixed.

7.3.1.2.1.16 Completion of Protection Action Once Initiated (IEEE-279, paragraph 4.16)

See Section 7.3.1.1.1.16 which also applies to the LPCI system.

7.3.1.2.1.17 Manual Actuation (IEEE-279, paragraph 4.17)

Each piece of LPCI actuation equipment required to operate (pumps and valves) is capable

of manual initiation electrically from the control panel in the main control room.

7.3.1.2.1.18 Access to Setpoint Adjustments (IEEE-279, paragraph 4.18)

See Section 7.3.1.1.1.18 which also applies to the LPCI system.

7.3.1.2.1.19 Identification of Protective Actions (IEEE-279, paragraph 4.19)

See Section 7.3.1.1.1.19 which also applies to the LPCI system.

7.3.1.2.1.20 Information Readout (IEEE-279, paragraph 4.20)

Sufficient information is provided on a continuous basis so that the operator can have a

high degree of confidence that the LPCI function is available and/or operating properly.

7.3.1.2.1.21 System Repair (IEEE-279, paragraph 4.21)

See Section 7.3.1.1.1.21 which also applies to the LPCI system.

7.3.1.2.2 Failure Mode and Effects Summary

Since the LPCI system is by itself a single system and, as such, vulnerable to single failures in common components, a detailed failure mode and effects analysis is not presented here.

The failure mode and effects analysis presented for the core spray system applies to all portions of the system except the injection valves and specific portions of the loop selection

circuitry. As has been previously discussed, those single failures that could possibly disable

the LPCI system will not directly affect the core spray system. The low QUAD CITIES - UFSAR 7.3-13 Revision 8, October 2005 pressure core cooling system is designed such that for any single failure the availability of the following will be maintained:

1. Two core spray loops, or
2. One core spray loop and two LPCI pumps.

7.3.1.3 High Pressure Coolant Injection System Instrumentation and Control

Automatic initiation of HPCI occurs on low-low reactor water level or high drywell pressure in the absence of the reactor vessel high water level HPCI turbine trip signal. Low-low

reactor water level is detected by four independent transmitters. High drywell pressure is

detected by four independent pressure switches. All sensors are connected in one-out-of-

two-twice logic arrays. The reactor high water level switches are connected in a two-out-of-two logic. When the initiation signal is received, the HPCI turbine and its required auxiliary equipment will start and the required valves will open automatically, with the

exception of the steam supply valves 2301-4 & 5 and the turbine exhaust line vacuum

breaker valves 2399-40 & 41. These valves must always be opened manually from the control room switches after a manual closure or any valid isolation signal that has caused the valves to close. If the HPCI system starts due to a high drywell pressure signal and

automatically turns off at reactor high level, then the system will automatically restart at

reactor low low level.

[7.3-8]

In the event of a low water level in the condensate storage tank, or high level in the

suppression pool, the pump suction valves from the suppression chamber open and the

suction valve from the condensate storage tank closes. The valves are interlocked to

prevent the suction valve from the condensate storage tank from automatically opening whenever both suction valves from the suppression chamber are fully opened.

Automatic isolation of the HPCI system is discussed in Section 7.3.2.

Initiation for automatic trip of the HPCI turbine occurs (whenever the turbine stop valve is

not tripped) on high turbine exhaust pressure, low pump suction pressure, or high reactor

water level. The low pump suction and high turbine exhaust pressure trips are blocked

when a HPCI auto-initiation signal (reactor water low-low level or high drywell pressure) is

present. High turbine exhaust pressure is detected by two redundant pressure switches connected in a one-out-of-two logic. Low pump suction pressure is detected by a single

pressure switch. The low pump suction pressure trip is delayed 2.5 seconds to eliminate

short duration low suction transient trips. High reactor water level is detected by two redundant level sensors connected in a two-out-of-two logic. The pump discharge is

prevented from opening automatically whenever a turbine trip condition exists.

7.3.1.3.1 Conformance with IEEE-279

The following is a point-by-point comparison of the HPCI system with the requirements of

IEEE Std 279-1968 which has been summarized from NEDO-10139.[1] The automatic depressurization system is provided to reduce reactor pressure in case the HPCI system is not sufficient to maintain the reactor water level. Therefore, it is clear that the HPCI system is not required to meet all the requirements of IEEE-279 since it is backed up by the

independent automatic depressurization system. The following comparison is provided only

to show the adequacy of the HPCI system design. For more detailed information refer to the topical report. The GE topical report is a generic report for an entire product line and these statements should be used concurrently with design requirements described in other sections of the UFSAR (such as Section 3.5 for Missile Protection, 3.6 for Pipe Break, etc.)

that represent Quad Cities specific design requirements.

[7.3-9]

QUAD CITIES - UFSAR 7.3-14 Revision 7, January 2003 7.3.1.3.1.1 General Functional Requirements (IEEE-279, paragraph 4.1)

A. Auto-Initiation of Appropriate Action

Appropriate action for the HPCI control system is defined as the activation of equipment for introducing high pressure water into the reactor via the feedwater

line when reactor vessel level drops below a predetermined point, or the drywell

pressure increases above a predetermined value. This action occurs

automatically.

B. Precision

See Section 7.3.1.1.1.1 which applies equally to the HPCI and core spray systems. Sensors that initiate the HPCI system are the same type of sensor that initiates the core spray system.

C. Reliability

Reliability of the control system is compatible with the controlled equipment so that the overall system reliability is not limited by the controls.

D. Action Over the Full Range of Environmental Conditions

Refer to Section 3.11 for information on the current environmental qualification program.

7.3.1.3.1.2 Single-Failure Criterion (IEEE-279, paragraph 4.2)

The HPCI system by itself, is not required to meet the single-failure criterion. The control

logic circuits for the HPCI system initiation and control are housed in a single relay cabinet. The relay cabinet and normal power source for the automatic depressurization

system is independent of the HPCI system.

The HPCI initiation sensors and wiring up to the HPCI relay logic cabinet does, however, meet the single-failure criterion. Physical separation of instrument lines is provided so

that no single instrument rack destruction or single instrument line (pipe) failure can

prevent HPCI initiation. Wiring separation between divisions also provides tolerance to

single wireway destruction (including shorts, opens, and grounds) in the accident detection

portion of the control logic.

7.3.1.3.1.3 Quality of Components (IEEE-279, Paragraph 4.3)

See Section 7.3.1.1.1.3 which also applies generally to the HPCI system.

QUAD CITIES - UFSAR Revision 7, January 2003 7.3-15 7.3.1.3.1.4 Equipment Qualification (IEEE-279, paragraph 4.4)

No components of the HPCI control system are required to operate in the drywell

environment except for the temperature compensation columns of the vessel level sensors.

Errors introduced under steam leak (high drywell temperature and reactor

depressurization) for HPCI initiation are negligible as discussed in Section 7.3.1.1.1.1(B).

The HPCI steam line isolation valve located inside the drywell is a normally open valve and is therefore not required to operate except under special (test) conditions.

Other process sensor equipment for HPCI initiation is located in the reactor building and is

capable of accurate operation in ambient temperature conditions that result from abnormal (loss of ventilation and LOCA) conditions.

7.3.1.3.1.5 Channel Integrity (IEEE-279, paragraph 4.5)

The HPCI system instrument initiation channels meet the single-failure criterion as

discussed in Section 7.3.1.3.1.2 above and thus satisfy the channel integrity objective of this

paragraph.

By definition (IEEE-279, paragraph 2.2) a channel loses its identity where single-action

signals are combined. Therefore, since instrument channels are combined into a single trip

system this paragraph of IEEE-279 does not strictly apply for the HPCI control system.

7.3.1.3.1.6 Channel Independence (IEEE-279, paragraph 4.6)

Channel independence for initiation sensors monitoring each variable is provided by

electrical and mechanical separation. The A and C sensors for reactor vessel level are

located on one local instrument rack identified as Division I equipment and the B and D

sensors are located on a second instrument rack widely separated from the first and

identified as Division II equipment. The A and C sensors have a common pair of process

taps which are widely separated from the corresponding taps for sensors B and D.

Disabling of one or both sensors in one location does not disable the control for HPCI

initiation.

7.3.1.3.1.7 Control and Protection Interaction (IEEE-279, paragraph 4.7)

See Section 7.3.1.1.1.7 which also applies to the HPCI system.

7.3.1.3.1.8 Derivation of System Inputs (IEEE-279, paragraph 4.8)

The inputs that start the HPCI system are direct measures of the variables that indicate

need for high pressure core cooling; such as, reactor vessel low water level or high drywell pressure.

QUAD CITIES - UFSAR Revision 4, April 1997 7.3-16 7.3.1.3.1.9 Capability for Sensor Checks (IEEE-279, paragraph 4.9)

See Section 7.3.1.1.1.9 which also applies to the HPCI system.

7.3.1.3.1.10 Capability for Test and Calibration (IEEE-279, paragraph 4.10)

See Section 7.3.1.1.1.10 which also applies to the HPCI system.

7.3.1.3.1.11 Channel Bypass or Removal from Operation (IEEE-279, paragraph 4.11)

Calibration of a sensor which introduces a single instrument channel trip will not cause a protective function without the coincident trip of a second channel. There are no

instrument channel bypasses as such in the HPCI system. Removal of a sensor from

operation during calibration does not prevent the redundant instrument channel from

functioning if accident conditions occur. Removal of an instrument channel from service

during calibration is brief.

7.3.1.3.1.12 Operating Bypasses (IEEE-279, paragraph 4.12)

Manual Bypasses

The HPCI system can be bypassed by placing of the flow controller from AUTO to

MANUAL operation in the main control room or adjusting AUTO operation. The controller is in the main control room and therefore under the direct supervision of the control room

operator.

Automatic Trips/Isolations

The following is a list of automatic functions which can render the HPCI system

inoperative:

[7.3-10]

A. HPCI steam line isolation signal.

B. The following signals will cause a HPCI turbine trip irrespective of an initiation:

1. Reactor vessel water level high.
2. HPCI turbine overspeed.
3. Local manual trip lever.

C. The following signals will cause a HPCI turbine trip if an initiation signal is not present:

1. HPCI pump suction pressure low,
2. HPCI turbine exhaust pressure high.

QUAD CITIES - UFSAR Revision 7, January 2003 7.3-17 7.3.1.3.1.13 Indication of Bypasses (IEEE-279, paragraph 4.13)

Indication of bypasses provided is as previously discussed in Section 7.3.1.3.1.12 above.

7.3.1.3.1.14 Access to Means for Bypassing (IEEE-279, paragraph 4.14)

Access to switchgear, motor control centers, ATS cabinets, relays, and instrument valves is procedurally controlled.

7.3.1.3.1.15 Multiple Trip Settings (IEEE-279, paragraph 4.15)

This is not applicable because all setpoints are fixed.

7.3.1.3.1.16 Completion of Protective Action Once Initiated (IEEE-279, paragraph 4.16)

The final control elements for the HPCI system are essentially bistable, that is, motor

operated valves stay open or closed once they have reached the desired position, even

though their starter may drop out (which will occur when the limit switch is reached). In

the case of pump starts, the auto initiation signal is electrically sealed-in, except for the

turbine reset solenoid. The LOCA signal must be maintained long enough to latch the

turbine reset cylinder.

[7.3-10a]

Thus a protective action once initiated (for example, flow established) must go to completion

or continue until terminated by deliberate operator action or automatically stopped on high

vessel water level or system malfunction trip signals.

7.3.1.3.1.17 Manual Actuation (IEEE-279, paragraph 4.17)

Each piece of HPCI actuation equipment required to operate (pumps and valves) is capable

of manual initiation electrically from the control panel in the main control room. Failure of

logic circuitry to initiate the HPCI system will not affect the manual control of equipment.

However, failures of active components or control circuit failures which produce a turbine

trip may disable the manual actuation of the HPCI system. Failures of this type are continuously monitored by alarms as discussed in previous sections and as such cannot

realistically be expected to occur when HPCI operation is required.

7.3.1.3.1.18 Access to Setpoint Adjustments (IEEE-279, paragraph 4.18)

Section 7.3.1.1.1.18 also applies to the HPCI system.

[7.3-10b]

QUAD CITIES - UFSAR 7.3-18 Revision 7, January 2003 7.3.1.3.1.19 Identification of Protective Actions (IEEE-279, paragraph 4.19)

Protective actions (which are here interpreted to mean pickup of a single sensor relay) are

directly indicated and identified by action of the sensor relay which has an identification

tag and a clear glass window front which permits convenient visible verification of the relay

position. A sensor trip also actuates an annunciator so that no single channel trip (relay

pickup) will go unnoticed. This combination of annunciation and visible relay actuation is

considered to fulfill the requirements of this criterion.

7.3.1.3.1.20 Information Readout (IEEE-279, paragraph 4.20)

The HPCI control system is designed to provide the operator with accurate and timely

information pertinent to its status. It does not introduce signals into other systems that

could cause anomalous indications confusing to the operator. There are many passive as

well as active elements of this energize-to-operate system which are not continuously monitored for operability. For example, relay circuits are normally open and are not monitored for continuity on a continuous basis. Pressure and level sensors, although

continuously active are not continuously exercised and verified operable. Periodic testing is

the means provided for verifying the operability of these components and by proper selection of test periods to be compatible with the historically established reliability of the

components tested, complete and timely indications are made available. Sufficient

information is provided on a continuous basis so that the operator can have a high degree of

confidence that the HPCI function is available and/or operating properly.

7.3.1.3.1.21 System Repair (IEEE-279, paragraph 4.21)

See Section 7.3.1.1.1.21 which applies equally to the HPCI system.

In addition to the recognition of failed components during test, components which fail in the

direction so as to produce a trip condition are continuously monitored by alarm.

7.3.1.3.2 Failure Mode and Effects Analysis Summary

Since the HPCI system is by itself a single system, a detailed failure mode and effects

analysis is not warranted as it is recognized that there are single failures that could disable

the system.

As has been previously described, no single failure in the initiation instrumentation can

prevent HPCI operation if required.

It is also mentioned again that those single failures that could possibly disable the HPCI

system will in no way affect the ADS system and vice versa.

No instrumentation or equipment is shared by the ADS and HPCI systems. Reactor vessel water level sensors for HPCI initiation are associated with RPS and PCIS, and are QUAD CITIES - UFSAR 7.3-18a Revision 7, January 2003 separate from ADS. Level transmitters for ADS initiation are associated with ATWS/ECCS and are separate from HPCI. Separate switches on the shared sensors are used for the two systems. Both physical and electrical separation are QUAD CITIES - UFSAR Revision 8, October 2005 7.3-19 maintained so that no single failure of the level-sensing equipment or wiring (shorts or opens) can, in fact, disable either HPCI or ADS.

Therefore, it is concluded that no single failure can disable both the HPCI and the ADS

systems.

7.3.1.4 Automatic Depressurization System Instrumentation and Controls

The ADS system allows use of LPCI or core spray as a backup to HPCI by depressurizing the

reactor pressure vessel for small area breaks. Reactor vessel depressurization is accomplished

by blowdown through relief valves to vent steam to the suppression pool.

[7.3-11]

The ADS is initiated by instrumentation which monitors drywell pressure and reactor water

level. Automatic blowdown requires both that a drywell high pressure and reactor water level low-low signal persist for a two-minute period (analytical limit for initiation timer). In

addition, the design prevents blowdown until the discharge pressure of at least one LPCI

pump or one core spray pump exceeds 100 psig (analytical limit). This design provides direct

assurance that the low pressure ECCS pumps are operating prior to automatic

depressurization.

[7.3-12]

Four instrument channels monitor each initiating parameter. Two of the four channels

monitoring each parameter are assigned to one of the two logic divisions. The arrangement of these signals within each logic division is two-out-of-two (high pressure and low-low level) in

coincidence with two-out-of-two (high pressure and low-low level). The trip in one of these

coincidence signals is interlocked with, and permits the starting of, a timer which delays

actuation of the relief valves to permit operator intervention and to allow the HPCI to restore

reactor water inventory. The time delay setting was chosen to be long enough so that the

HPCI has time to start, yet not so long that core spray and LPCI systems are unable to

adequately cool the fuel if the HPCI fails to start.

The automatic depressurization system is also initiated when low-low reactor water level is

sensed continuously for a maximum of 9 minutes (analytical limit for actuation timer) and a

low pressure pump is running as previously stated. This reactor water level sensing logic is

two-out-of-two per division. The automatic depressurization system also has a keylocked, administratively-controlled, manually-actuated inhibit switch that prevents blowdown irrespective of any initiation signal. Inside panel 901(2)-32, there is a second keylocked, administratively-controlled, manually-actuated inhibit switch.

[7.3-13]

For additional reliability, each pair of circuits is provided with power from separate dc buses.

The instruments in the reactor vessel water level circuit and drywell pressure circuit do not

require electrical power to close or open the sensors in the initiation circuits, but the logic

circuitry requires 125 VDC power to operate. The single failure of one single switch in its

respective circuit will not cause an ADS actuation. An additional power source is also

available and is automatically switched over upon loss of the primary power source.

[7.3-13a]

7.3.1.4.1 Conformance With IEEE-279

The following is a point-by-point comparison of the automatic depressurization system (ADS)

with the design requirements of IEEE Std 279-1968 which has been summarized from GE

Topical Report, NEDO-10139.[1] For more detailed information refer to the topical report. The GE topical report is a generic report for an entire product line and these statements should be used concurrently with design requirements described in other sections of the UFSAR (such as Section 3.5 for Missile Protection, 3.6 for Pipe Break, etc.) that represent Quad Cities specific design requirements.

[7.3-14]

QUAD CITIES - UFSAR Revision 6, October 2001 7.3-20 7.3.1.4.1.1 General Functional Requirement (IEEE-279, paragraph 4.1)

A. Auto-Initiation of Appropriate Action

Appropriate action is defined as initiating the opening of a specified number of valves when loss of primary coolant is detected by reactor vessel low level, persists for approximately two minutes, and is confirmed by high drywell

pressure, provided that low pressure standby core cooling equipment is available

and operating or when reactor vessel low-low level is sensed for 9 minutes

continuously (analytical limit). The ADS design accomplishes the appropriate

action automatically.

B. Precision

The accuracy requirements for initiating ADS (like those for the core spray system) are not such that precision of measurement is required. Precision

provided by these instruments is adequate to give positive automatic

depressurization initiation before the vessel water level can go below a tolerable

point. The ADS control design achieves the degree of precision necessary to

insure appropriate initiation of the protective function when needed and

precludes inadvertent initiation under extremes of environment related errors in

instrumentation.

C. Reliability

The reliability of the auto depressurization control system is an estimated order of magnitude higher than the reliability of the actuated equipment (valves).

D. Action Over the Full Range of Environmental Conditions: fire, accidents, missiles, etc.

The corresponding section for the core spray system Section 7.3.1.1.1.2 applies here in all

respects except fire and missiles. A single cabinet houses the redundant relays that energize all the auto depressurization valves in unison. However, the circuits to the ADS

valves emerge from this cabinet in independent metal conduits and are carried through

separate penetrations into the drywell. Separate metal conduits are carried from the

penetrations to the individual valves distributed among the four main steam lines.

In view of the fact that wiring for the relief valve solenoids must survive the LOCA environment for an appreciable time, (at least several minutes to perhaps an hour), cable

has been selected which can easily tolerate this environment.

A destructive fire enveloping the control cabinet could disable all valve control circuits.

Such a fire is not considered credible from electrical sources because of the low current

available in the circuits involved and the fire resistant nature of the devices and wiring within the cabinet. Thus external, non-electrical fires are considered to be the only possible

fire damage source.

Separate routing of the ADS conduits within the drywell reduces to a very low probability

the possibility of missile damage to more than one ADS conduit or damage to the pilot solenoid assembly of ADS valves. The HPCI system will provide backup for the ADS under

all conditions unless the HPCI line is the source of the missile or jet in which case damage

to a single ADS valve or conduit is considered credible.

QUAD CITIES - UFSAR 7.3-21 If a valve were rendered inoperable by a jet of water and/or steam associated with a pipe break (Section 3.6), the redundancy of the ADS system provides adequate protection for all

possible break situations. This is true even for breaks in the feedwater line used for HPCI injection which is the worst case, since the HPCI function could then be impaired or lost.

The situation leaves all but one relief valve and all low pressure ECCS operable. Since the

plant has one extra relief valve, 100% automatic relief capacity is left. If a single additional

failure is added to this situation, the worst failure would be to fail one more relief valve

arbitrarily. This leaves LPCI pumps, two core spray loops and the ADS degraded by one

valve. Since the postulated break is located in the feedwater line, which is connected to the

reactor vessel above the core, the relief capacity, degraded by one valve, is adequate to

provide cooling protection.

Further, it should be noted that the situation described above would require an extremely

unlikely combination of circumstances.

In light of the above, it is concluded that ADS fulfills the minimum requirement of IEEE-

279 paragraph 4.1 without benefit of backup from HPCI.

7.3.1.4.1.2 Single Failure Criterion (IEEE-279, paragraph 4.2)

The single failure criterion of IEEE-279 is not directly applicable to ADS because HPCI and

ADS are diverse functional backups to each other insofar as depressurization is concerned.

However, ADS has been designed to accommodate all of the single failures listed under the

core spray systems with the exception of a single wireway destruction as described in

Section 7.3.1.1.1.5 or a single control cabinet section destruction.

It is not considered credible that any single event could occur within the automatic

depressurization cabinet that could disable more than one valve.

Inadvertent operation of the automatic depressurization system cannot result from failure

or malfunction of any single component including single shorts or single opens. Only one

valve can be opened by any single short.

7.3.1.4.1.3 Quality of Components (IEEE_279, paragraph 4.3)

See Section 7.3.1.1.1.3 which also applies to ADS.

7.3.1.4.1.4 Equipment Qualification (IEEE-279, paragraph 4.4)

See Section 7.3.1.1.1.4 which also applies to ADS insofar as the level sensors are concerned.

7.3.1.4.1.5 Channel Integrity (IEEE-279, paragraph 4.5)

See Section 7.3.1.1.1.5 which also applies to ADS.

QUAD CITIES - UFSAR Revision 7, January 2003 7.3-22 7.3.1.4.1.6 Channel Independence (IEEE-279, paragraph 4.6)

Channel independence for sensors exposed to each variable is provided by electrical and

mechanical separation. The A and C sensors for reactor vessel level are located on a stanchion adjacent to the Division I instrument rack, and the B and D sensors are located on a pair of stanchions adjacent to the Division II instrument rack. The A and C sensors have a common pair of process taps which are widely separated from the corresponding

taps for sensors B and D. Disabling of one or both sensors in one location does not disable the control for both of the automatic depressurization control channels.

There are two sensors of each type in one division mechanically and electrically

independent from those in the second division to initiate automatic depressurization.

Therefore, these sensors are redundant to each other. The logic for each trip channel is

four-out-of-four. So, the overall ADS trip logic becomes one of two, four-out-of-four logics.

In addition to the sensors that initiate automatic depressurization there are ADS

permissive sensors associated with the pump discharge pressure of the low pressure ECCS.

An interlock is provided in each trip system in order to give reassurance that low pressure

core coolant is available before ADS actually permits depressurization of the reactor vessel.

This interlock tends to degrade the reliability of ADS but is so arranged that this

degradation is reduced to a practical minimum. Two pressure switches (twelve total) on the

discharge of each core spray and each LPCI pump are connected through relays in

redundant groups so that each ADS trip system is blocked from actuating unless at least

one low pressure pump shows verified discharge pressure.

7.3.1.4.1.7 Control and Protection Interaction (IEEE-279, paragraph 4.7)

The automatic depressurization system is strictly an off-or-on system and no signal whose

failure could cause need of automatic depressurization can also prevent it from starting.

7.3.1.4.1.8 Derivation of System Inputs (IEEE-279, paragraph 4.8)

Inputs which start automatic depressurization system are direct measures of the variables

that indicate the need for and acceptable conditions for rapid depressurization of the

reactor vessel (such as, reactor vessel low water verified by high drywell pressure and at least one low pressure core cooling system developing adequate discharge pressure or when

reactor vessel low-low level is sensed for 9 minutes continuously (analytical limit)).

7.3.1.4.1.9 Capability for Sensor Checks (IEEE-279, paragraph 4.9)

All sensors are of the pressure sensing type and are installed with calibration taps and

instrument valves which allow for the application of a test pressure for calibration and/or

functional tests during normal plant operation or during shutdown.

QUAD CITIES - UFSAR Revision 7, January 2003 7.3-23 7.3.1.4.1.10 Capability for Test and Calibration (IEEE-279, paragraph 4.10)

The automatic depressurization system is not tested in its entirety during actual plant

operation but provisions are incorporated so that operability of all elements of the system

can be verified at periodic intervals.

7.3.1.4.1.11 Channel Bypass or Removal from Operation (IEEE-279, paragraph 4.11)

Calibration of each sensor will introduce a single instrument channel trip. This does not

cause a protective action without the coincident trip of three other channels. Removal of an instrument channel from service during calibration is brief and does not significantly

increase the probability of failure to operate. There are no channel bypasses as such in

ADS. Removal of a sensor from operation during calibration does not prevent the

redundant trip circuit from functioning if accident conditions occur because they will be

sensed by the redundant sensors. The manual reset switch can interrupt the automatic

depressurization for a limited time. However, releasing either one of the two reset switches

will allow automatic timing and action to resume. The ADS inhibit switches will prevent

blowdown if placed in the INHIBIT position. These switches are keylocked and

administratively-controlled.

7.3.1.4.1.12 Operating Bypasses (IEEE-279, paragraph 4.12)

See Section 7.3.1.4.1.11 which also generally applies to the ADS. Disabling of two selected

sensors would also disable the auto depressurization action and would result from selective shutting off of one or more sensor instrument valves for each of the two sets of four sensors.

This mechanism of disabling the system is not considered to be an operating bypass so no

exception to IEEE-279 is taken.

7.3.1.4.1.13 Indication of Bypasses (IEEE-279, paragraph 4.13)

The ADS inhibit switches as well as the manual opening of the control power breakers can

disable the automatic depressurization function. Placing either ADS inhibit switch in the

INHIBIT position, or a control power loss, is annunciated. Disabling of sensors by

deliberately shutting off instrument valves is not indicated, but such action is under the

operator's procedural control and cannot be done without appropriate authorization.

7.3.1.4.1.14 Access to Means for Bypassing (IEEE-279, paragraph 4.14)

Instrument valves are administratively controlled and cannot be operated without permission of responsible authorized personnel.

Reset switches are on the control panel in the main control rooms. Control power breakers are in dc distribution cabinets which are normally locked and under administrative

controls.

QUAD CITIES - UFSAR 7.3-24 7.3.1.4.15 Multiple Trip Settings (IEEE-279, paragraph 4.15)

Not applicable because all trip points are fixed.

7.3.1.4.1.16 Completion of Protection Action Once Initiated (IEEE-279, paragraph 4.16)

Each of the two trip systems for the automatic depressurization control seals in electrically

and remains energized until manually reset by one of the two reset switches.

7.3.1.4.1.17 Manual Actuation (IEEE-279, paragraph 4.17)

Each valve has its individual manual control switch which can operate the valve even

though the automatic control relays cannot operate for any reason including loss of control power fuses. Each valve has its own fused solenoid power circuit which is coordinated with

the breaker which provides power for ADS control. Manual control is therefore

independent of automatic control.

7.3.1.4.1.18 Access to Setpoint Adjustments (IEEE-279, paragraph 4.18)

See Section 7.3.1.1.1.18 which also applied to ADS.

7.3.1.4.1.19 Identification of Protective Actions (IEEE-279, paragraph 4.19)

See Section 7.3.1.1.1.19 which also applies to ADS.

7.3.1.4.1.20 Information Readout (IEEE-279, paragraph 4.20)

The information provided to the operator pertinent to ADS status are as follows:

A. Annunciators,

B. Valve position lights for each valve, and

C. Reactor vessel level indication.

From the previous text it can be seen that change of state of any active component from its

normal condition is called to the operator's attention; therefore, the indication is considered

to be complete and timely. Refer to Section 5.2.2 for a discussion of the acoustic monitors.

QUAD CITIES - UFSAR Revision 10, October 2009 7.3-25 7.3.1.4.1.21 System Repair (IEEE-279, paragraph 4.21)

As with core spray, ADS is designed to avoid the need for repair rather than for fast replacement of components. Thus reliability is built-in rather than approached by accelerated

maintenance. All devices in the system are designed for a 40-year lifetime under the duty

cycles to be imposed. Since this duty cycle is composed completely of testing at infrequent

intervals, the duration of active components other than sensors is more a matter of shelf life

than active life. However, all instrument components are selected for continuous duty plus

thousands of cycles of operation (far beyond that anticipated in actual service). Recognition and location of a failed component is accomplished during periodic testing.

7.3.2 Primary Containment Isolation Systems

7.3.2.1 Design Basis

The objective of the primary containment isolation system (PCIS) is to provide timely

protection against the onset and consequences of accidents involving the gross release of

radioactive materials from the primary containment. The PCIS system provides automatic

isolation of appropriate pipelines which penetrate the primary containment whenever certain

monitored variables exceed their preselected operational limits. To accomplish this objective, PCIS was designed using the following criteria:

[7.3-15]

A. Prevent the release of radioactive materials in excess of the limits in 10 CFR 100 (or 10 CFR 50.67 as applicable) as a result of the design basis accidents;

B. Function safely when required regardless of the failure of any single component; and C. Function independently of other plant controls and instrumentation.

7.3.2.2 Isolation Logic Description

The primary containment and reactor vessel isolation control system includes the sensors, trip

channels, switches, and the remotely activated valve closing mechanisms associated with the

valves which, when closed, isolate either the primary containment, the reactor vessel isolation

valves, or both.

[7.3-16]

Power for the trip systems and trip logics for Groups 1, 2, 3, and the RHR shutdown cooling isolation are supplied from the same two electrical busses that feed the reactor protection system (RPS). Refer to Section 7.2 for more information on RPS. The analog trip system (see

section 7.6) logic that is part of the trip logic is supplied from separate essential service motor

control centers. The trip logic for isolation Groups 4 and 5 are arranged differently. For these

Groups, there are two trip systems per group which has each trip system electrically supplied

by separate 125 Vdc sources. Only one trip system is required to provide an automatic

isolation for Group 4 and 5. Technical Specifications preserve system effectiveness even

during periods of maintenance and testing activities. The two series isolation valves are

supplied from different sources. One valve is powered from a reliable ac bus and the other

valve is powered by a dc bus. Series solenoid valves are typically powered from separate ac

buses. The MSIVs (described in detail in Section 6.2.4.3) use ac and dc power and pneumatic

pressure accumulators in the control scheme. Power cables are run in conduits from QUAD CITIES - UFSAR 7.3-26 Revision 10, October 2009 appropriate electrical sources to the motor or solenoid that operates each isolation valve.

The pneumatic control is provided to close the. MSIVs on loss of ac and dc power.

[7.3-17]

The PCIS logic is arranged as a dual logic channel system, similar to that of the reactor

protection system. The overall logic of the system is one-out-of-two-twice. Exceptions to

this basic logic arrangement are explained in the individual logic descriptions.

During normal operation of the isolation control system for Groups 1, 2, 3, and RHR shutdown cooling - when isolation is not required - sensor and trip contacts (essential to safety) are closed; trip channels, trip logics, and trip actuators are normally energized.

Whenever a trip channel sensor contact opens, its auxiliary relay de-energizes, causing

contacts in the trip logic to open. The opening of contacts in the trip logic de-energizes its trip actuators. When de-energized, the trip actuators open contacts in all the trip actuator

logics for that channel. If a trip then occurs in any of the trip logics of the other trip

channel, the trip actuator logics for the other channel are de-energized. With both trip

channels tripped, appropriate contacts open or close in the valve control circuitry to actuate

the valve closing mechanism. Automatic isolation valves that are normally closed receive an

isolation signal, as do those valves that are open. Once isolation is initiated, the valve

continues to close, even if the condition that caused the isolation signal clears. The

operator must operate switches in the control room to manually reset the isolation signal

and reopen a valve which has been automatically closed.

The trip logic for the following RWCU isolation has only two instrument or initiating device channels.

A. SBLC Activation Interlock (not a containment isolation related signal) B. RWCU Area Temperature High Non-regenerative heat exchanger outlet temperature high isolation is not a safety-related containment isolation, but a system isolation signal and only has one sensor.

The two channel logic for the above RWCU (Group 3) trips is acceptable, because maintenance and surveillances associated with this logic do not challenge safety systems.

The trip logic for isolation Groups 4 & 5 are different in the fact that there are two (2) trip

systems per isolation group, and the logic structure is normally de-energized. For Group 5, each trip system isolates both the Inboard and Outboard containment isolation valves.

Modification M04-1(2)-91-013B incorporated Regulatory Guide 1.75 and IEEE 384 criteria

for the electrical, and where possible, the physical separation of trip channels for Group 4 only. For Group 4, one trip system isolates only the Inboard valve while the second trip system isolates the Outboard valves.

In addition, the valves associated with the Group 1, 2, 3, 4, and RHR shutdown cooling isolations, as described in Table 6.2-7, will not automatically open when the isolation signal is reset.

[7.3-18]

A keylock bypass switch is provided to allow venting of the containment when an isolation

signal is present. This switch provides control room annunciation when it is not in the

normal position.

Primary containment isolation functions are initiated by groups, according to the trip

channel logic associated with each group. Additionally, manual switches on the control

panel in the control room are available for each isolation valve to back up all trip signals.

Figure 7.3-1 displays the various functions of the system and the signals that initiate their

operation.

[7.3-19]

QUAD CITIES - UFSAR Revision 10, October 2009 7.3-27 There are five groups of isolation valves as follows:

[7.3-20]

Group 1 - this group includes the isolation valves for the:

1. Four main steam lines.
2. Main steam line drain.
3. Reactor water sample line.

Group 2 - included in this group are the isolation valves for:

1. Drywell equipment drain discharge.
2. Drywell floor drain discharge.
3. Traversing in-core probe tubes.
4. Drywell purge inlet.
5. Drywell main exhaust.
6. Suppression chamber exhaust valve bypass.
7. Suppression chamber purge inlet.
8. Suppression chamber main exhaust.
9. Drywell Nitrogen purge inlet.
10. Nitrogen Makeup.
11. Nitrogen makeup to Drywell.
12. Nitrogen makeup to Suppression chamber.
13. Drywell exhaust to standby gas treatment.
14. Main primary containment vent to reactor building exhaust system.
15. Drywell exhaust valve bypass.
16. Drywell oxygen analyzer sample.
17. Torus oxygen analyzer sample.
18. Oxygen analyzer return.
19. Drywell pneumatic suction.
20. RHR reactor shutdown cooling suction.
21. RHR reactor LPCI/shutdown cooling injection (only when RHR is in operation in the shutdown cooling mode).
22. RHR discharge to radwaste.

QUAD CITIES - UFSAR Revision 10, October 2009 7.3-28 Group 3 - included in this group are the isolation valves for:

1. Reactor water cleanup.

Group 4 - included in this group are the isolation valves for:

1. HPCI steam line isolation.
2. HPCI turbine exhaust line vacuum breaker line isolation.

Group 5 - included in this group are the isolation valves for:

1. Reactor core isolation cooling (RCIC) steam line isolation.

In addition to the RHR shutdown cooling isolation received during a Group 2, the RHR shutdown cooling suction valves will close on a reactor high pressure condition.

In addition to the isolation valves listed above, the reactor building floor drain and

equipment drain pumps trip upon the receipt of a Group 2 isolation signal.

The analytical limits for the isolation signals are listed in Table 7.3-1. Table 6.2-7 shows

the valves affected by the system.

[7.3-20a]

The isolation functions and trip settings used for the electrical control of isolation valves

are discussed in the following paragraphs.

7.3.2.2.1 Low Reactor Vessel Water Level

A low reactor vessel water level could indicate that reactor coolant is being lost through a breach in the nuclear system process barrier and that the core is in danger of becoming

overheated as the reactor coolant inventory diminishes. There are two reactor vessel low

water level isolation trip settings used to initiate the isolation of the primary containment

and the reactor vessel.

[7.3-21]

The first reactor vessel low water level isolation trip setting, which occurs at a higher water level than the second setting, initiates closure of all Group 2 and Group 3 isolation valves

in major process pipelines. The main steam line isolation valves (Group 1) are left open to

allow the removal of heat from the reactor core.

This setting which, coincidentally is the same as the reactor vessel low water level scram

setting, was selected to initiate isolation at the earliest indication of a possible breach in

the nuclear system process barrier yet far enough below normal operational levels to avoid

spurious isolation.

The second and lower reactor vessel low (low-low) water level isolation trip setting

completes the isolation of the primary containment and reactor vessel by closure of the

Group 1 isolation valves.

This setting was selected to be low enough to prevent actuation of the ECCS during normal

operation or during normally expected transients, yet high enough to complete its isolation

in time for the operation of the ECCS to provide effective core cooling.

QUAD CITIES - UFSAR 7.3-29 Revision 10, October 2009 7.3.2.2.2 Main Steam Line High-High Radiation

For a discussion of this topic, refer to Section 11.5.2.

[7.3-22]

7.3.2.2.3 Main Steam Line Space High Temperature

High temperature in the space where the main steam lines are located, outside of the

primary containment, could indicate a breach in a main steam line. The automatic closure of

Group 1 valves prevents the excessive loss of reactor coolant and the release of significant

amounts of radioactive material from the nuclear system process barrier.

[7.3-23]

Due to a small section of RWCU piping in the space where the main steam lines are located, outside of the primary containment, two of the four main steam line high temperature switch channels (A and B) also provide an automatic isolation of Group 3 valves of the RWCU system. Additional Group 3 high temperature isolation of the RWCU system is discussed in UFSAR Section 7.3.2.2.14. Area leak detection allows isolation at lower power levels than would isolate on Reactor Water Level Low.

[7.3-24]

The main steam line space high temperature trip is set far enough above the temperature

expected during operations at rated power to avoid spurious isolation, yet low enough to

provide early indication of a steam line break.

7.3.2.2.4 Main Steam Line High Flow

Main steam line high flow could indicate a break in a main steam line. The automatic

closure of the Group 1 valves prevents the excessive loss of reactor coolant and the release of

significant amounts of radioactive material from the nuclear system process barrier.

[7.3-25]

The main steam line high flow trip setting was selected high enough to permit testing of one

main steam line for operability of the respective MSIV at reduced power without causing an automatic isolation of the rest of the steam lines, yet low enough to permit early detection of a steam line break (Reference Section 6.2.6.3.1).

7.3.2.2.5 Low Steam Pressure at Turbine Inlet

[7.3-26]

Low steam pressure at the turbine inlet while the reactor is operating could indicate a

malfunction of the reactor pressure regulator in which the turbine control valves or turbine

bypass valves inadvertently fully open. This action causes rapid depressurization of the reactor. Also, in the event of a steam line break the vessel would rapidly depressurize. From

even partial load operating conditions, the rate of temperature decrease could exceed the

allowable vessel temperature rate of change. Such depressurization without adequate

preventive action, could require thorough vessel analysis or core inspection prior to returning

the reactor to power operation. In lieu of an analysis of the conditions following a rapid

depressurization, the steam pressure at the turbine inlet is monitored. Steam pressure, upon falling below a preselected value with the reactor in the RUN mode, initiates a time

delay relay. If steam pressure remains below the preselected value during the delay time, an

isolation of the Group 1 isolation valves is initiated. The low steam pressure isolation setting

was selected far enough below normal turbine inlet pressures to avoid spurious isolation, yet

high enough to provide timely detection of a pressure regulator malfunction. The total

channel response time, from the time main steamline pressure drops to below the low

pressure setpoint to QUAD CITIES - UFSAR Revision 6, October 2001 7.3-30 the time a Group I isolation is initiated, is not greater than 0.5 seconds (analytical limit).

Although this isolation function is not required to satisfy any of the safety design bases for this system, it is included here to make the isolation functions list complete.

7.3.2.2.6 Primary Containment (Drywell) High Pressure

[7.3-27]

High pressure in the drywell could indicate a breach of the nuclear system process barrier

inside the drywell.

The automatic closure of various Group 2 valves prevents the release of significant amounts

of radioactive material from the primary containment.

The primary containment high pressure isolation setting was selected to be as low as

possible without inducing spurious isolation trips.

[7.3-28]

High Drywell pressure makes up half of the required trip for the HPCI vacuum breaker

isolation (Group 4) logic. The logic for this Group 4 isolation is one-out-of-two-taken twice, on low reactor pressure and high drywell pressure. The HPCI turbine exhaust line isolation is not required on HPCI steam line break, but will isolate on indications of a large

break LOCA inside the drywell.

[7.3-29]

7.3.2.2.7 Primary Containment (Drywell) High Radiation

High radiation in the drywell indicates an abnormal situation due to a line break or other

abnormal occurrence. To preclude the release of potentially highly contaminated material

from the containment, this isolation signal automatically closes the Group 2 isolation

valves.

7.3.2.2.8 Reactor Core Isolation Cooling Turbine Space High Temperature

High temperature in the vicinity of the RCIC turbine could indicate a break in the RCIC

steam line. The automatic closure of the RCIC isolation valves prevents the excessive loss

of reactor coolant and the release of significant amounts of radioactive material from the

nuclear system process barrier. The high-temperature isolation setting was selected far

enough above anticipated normal RCIC system operational levels to avoid spurious

operation, but low enough to provide timely detection of a RCIC turbine steam line break.

[7.3-30]

QUAD CITIES - UFSAR Revision 6, October 2001 7.3-31 7.3.2.2.9 Reactor Core Isolation Cooling Turbine High Steam Flow

A RCIC turbine high steam flow signal could indicate a break in the RCIC turbine steam

line. The automatic closure of the RCIC isolation valves prevents the excessive loss of reactor coolant and the release of significant amounts of radioactive materials from the

nuclear system process barrier. When RCIC turbine high steam flow is detected, the RCIC

turbine steam line is isolated. The high steam flow trip setting was selected high enough to

avoid spurious isolation yet low enough to provide timely detection of a RCIC turbine steam

line break. A time-delay relay with a setting of 3 to 9 seconds (analytical limit) is used to

prevent spurious isolations (on receipt of high steam flow) during turbine startup.

[7.3-31]

The logic arrangement used for this function is a one-out-of-two and is an exception to the

usual logic arrangement because high steam flow is the alternate method of detecting an

RCIC turbine steam line break.

7.3.2.2.10 Reactor Core Isolation Cooling Turbine Steam Line Low Pressure

The RCIC turbine steam line low pressure signal is used to automatically close the two

isolation valves in that line so that steam and radioactive gases will not escape from the

RCIC turbine shaft seals into the reactor building after steam pressure has decreased to such a low value that the turbine cannot be operated. The isolation setpoint is chosen at a

pressure below that at which the RCIC turbine can operate effectively. A loss of pressure in the steam supply to the RCIC turbine could also indicate a steam line break. The low pressure signal therefore, backs up the other RCIC line break detection signals.

[7.3-32]

7.3.2.2.11 High Pressure Coolant Injection Turbine Space High Temperature

High temperature in the HPCI Room could indicate a HELB in the HPCI system and

causes a HPCI steam supply line Group 4 isolation. The automatic closure of the HPCI

steam supply isolation valves prevents the excessive loss of reactor coolant and the release

of significant amounts of radioactive material from the nuclear system process barrier. The

high temperature isolation setting was selected far enough above ambient to avoid spurious

isolations, but low enough to provide timely detection of a break.

[7.3-33]

The instrument sensors are located near the steam supply line and the turbine exhaust

rupture disc. The 2 instruments at each location are utilized for inputs to each trip

channel. Each trip channel will, therefore, detect smaller leaks at the 2 different locations

or a larger leak, as indicated by an increase in the overall room temperature.

The logic utilized for this Group 4 isolation is two-out-of-two in each trip channel. This

logic provides the advantage that a single spurious instrument trip will not isolate HPCI, and multiple failures (to trip) are required to prevent an isolation following a HELB

accident. Utilizing this logic, instead of one-out-of-two-twice logic, is justified based on the

redundancy of the HPCI High Room Temperature and the HPCI High Steamline Flow

trips.

QUAD CITIES - UFSAR Revision 6, October 2001 7.3-32 7.3.2.2.12 High Pressure Coolant Injection Turbine High Steam Flow

HPCI turbine high steam flow could indicate a break in the HPCI turbine steam lines. This

instrumentation senses high flow from taps inside the drywell in order to monitor flow from

any potential break outside the drywell. Breaks in areas other than the HPCI Room are

detectable. The automatic closure of the HPCI steam supply isolation valves prevents

excessive loss of reactor coolant and the release of significant amounts of radioactive

materials from the nuclear system process barrier. A time delay relay with a setting of

greater than or equal to 3 seconds and less than or equal to 9 seconds (analytical limit) is used to prevent spurious isolation during turbine startup.

The HPCI turbine high steam flow setting was selected high enough to avoid spurious isolation, yet low enough to prevent excessive inventory loss from the reactor vessel.

The instrumentation for each of the 2 trip channels include a transmitter and 2 trip units (one trip unit detects high steam flow or a break in the low pressure instrument sensing

line, and the other trip unit detects a break in the high pressure instrument sensing line).

The use of one-out-of-one logic in each trip channel is justified based on the use of highly

reliable instrumentation (EQ qualified transmitter with analog trips), redundancy of the

2 trip channels, and the redundancy with the HPCI room temperature trip.

7.3.2.2.13 High Pressure Coolant Injection Turbine Steam Line Low Pressure

Low reactor pressure, as measured in the HPCI turbine steam supply line, is used to isolate the HPCI steamline so that steam and radioactive gases will not escape from the reactor

pressure vessel and/or containment through the HPCI system.

HPCI turbine seals would become ineffective at preventing leakage from the turbine casing

at low steam pressures. The isolation setpoint is chosen at a pressure below where HPCI is

needed to mitigate the consequences of a small or intermediate break LOCA and above the

pressure where the turbine and turbine seals cease to function. The HPCI steam supply

line would be isolated by this instrumentation following a large break LOCA.

The use of two-out-of-two logic for each trip channel prevents an isolation in the event of a

single, spurious instrument trip. Use of this logic, instead of one-out-of-two-twice logic, is

justified by the use of high quality EQ qualified transmitters, analog trip instruments, and

the redundancy of the two trip channels.

Low reactor pressure, as measured in the HPCI turbine steam supply line, also makes up

half of the required trips for the HPCI vacuum breaker isolation logic.

The HPCI vacuum breaker isolation valves are not required to isolate or mitigate the

consequences of a HPCI steam line break, but will isolate on indications of a large break LOCA inside the containment to prevent a radiological release through the HPCI system.

The logic in each trip channel for this isolation is one-out-of-two, taken twice on low reactor

pressure and high drywell pressure. The use of 4 instruments in each trip channel make this a highly reliable trip logic. All 4 instruments measuring one of the two parameters

would have to fail to prevent an isolation.

QUAD CITIES - UFSAR Revision 10, October 2009 7.3-33 7.3.2.2.14 RWCU Piping Area High Temperature

Two RWCU Auto-Isolation Analog Trip System panels provide detection, alarm and isolation

signals for RWCU pipe breaks. The main reason this instrumentation signal was provided is to isolate RWCU breaks at lower reactor power levels when reactor feedwater flow can make

up reactor water level losses and prevent automatic isolation on reactor water low level.

[7.3-34]

The trip logic employs a one-out-of-two taken once logic as there are only two trip channels.

Detection by any single RTD provides actuation of both the inboard and outboard RWCU

isolation valves.

The area temperature trip settings are selected to insure the RWCU HELB analysis is

bounded and environmental conditions are not more severe than the worse case accident

previously analyzed.

7.3.2.2.15 Reactor Vessel Pressure High The reactor vessel pressure high function provides equipment protection to prevent an RHR intersystem LOCA scenario. This function isolates the RHR shutdown cooling suction valves. The pressure is sensed on the "B" recirculation loop suction line where RHR shutdown cooling takes it suction.

The isolation employs a one-out-of-two taken once trip logic since there are only two trip channels. Detection by any single sensor provides actuation of both inboard and outboard suction valves. The setpoint selected for this value assures that the pressure rating of the RHR shutdown cooling piping and components will not be exceeded when the suction valves are open. This function also serves as a permissive for RHR to operate in the shutdown cooling lineup mode.

7.3.2.3 Primary Containment Isolation System Instrumentation

Sensors providing inputs to the primary containment and reactor vessel isolation control

system are dedicated to that function. Trip channels are physically and electrically separated

to reduce the probability that a single physical event will prevent isolation. Trip channel

sensors for one monitored variable that are grouped near each other provide inputs to

different isolation trip systems. The sensors are described in the following paragraphs.

[7.3-35]

A. Reactor vessel water level signals (PCIS isolation) are initiated from four level transmitters via four indicating- and four nonindicating analog trip switches that

are part of the analog trip system. The transmitters sense the difference between

the pressure of a constant reference column of water and the pressure due to the

actual water level in the vessel. The four indicating switches are used to identify

that water level has decreased to the low water level isolation setting. The four

nonindicating switches are used to identify that water level has decreased to the

low-low water level isolation settings. The four switches for each level setting are

arranged in pairs; each switch in a pair provides a signal to a different isolation

logic channel.

[7.3-36]

Two instrument sensing lines, attached to taps above and below the water level on the reactor vessel, are required for the differential pressure measurement for each

pair of transmitters. The two pairs of sensing lines terminate outside the primary

containment and inside the reactor building; they are physically separated from each other and tap off the reactor vessel at widely separated points. This QUAD CITIES - UFSAR Revision 10, October 2009 7.3-34 arrangement assures that no single physical event can prevent isolation, if required. Cables from the level sensors are routed to the analog trip cabinets.

Temperature equalizing columns are used to reduce errors in level measurement

that can occur with changes in reactor water temperature.

B. Main steam line radiation is monitored by four radiation monitors, which are described in Section 11.5.2

[7.3-37] C. High temperature in the vicinity of the main steam lines is detected by 16 bimetallic temperature switches located along the main steam lines between the drywell wall and the turbine. The detectors are located or shielded so that they are

sensitive only to air temperature and not the radiated heat from hot equipment.

[7.3-38]

D. High flow in each main steam line is sensed by four differential pressure transmitters that sense the pressure difference across the flow restrictor in that

line. Each transmitter provides an input signal to an indicating analog trip unit.

[7.3-39]

The logic is arranged as two trip systems, both of which must trip to initiate isolation. Each trip system has two trip logics, either of which can trip the parent

trip system. Each trip logic receives an input from a high steam flow trip channel

for each steam line.

E. Main steam line low pressure is sensed by four bourdon-tube pressure switches which sense pressure downstream of the outboard main steam isolation valves. The

sensing point is located as close to the turbine stop valves as possible. The switches

are arranged as two trip systems both of which must trip to initiate isolation. Each

trip system receives inputs from two main steam line low pressure trip channels, either of which can trip the system.

[7.3-40]

F. Primary containment pressure is monitored by four nonindicating pressure switches which are mounted on instrument racks outside the drywell. Instrument

sensing lines connect the pressure switches located in the reactor building to the

drywell atmosphere. Cables are routed from the switches to the control room via

the auxiliary electrical room. The switches are grouped in pairs, physically

separated, and electrically connected to the isolation control system so that no

single event will prevent isolation due to primary containment high pressure.

[7.3-41]

The containment pressure is also monitored by four additional nonindicating electronic pressure switches per division which were used to isolate the ACAD

system under high drywell pressure conditions. The switches are grouped in pairs, physically separated and electrically connected to the isolation control system so

that no single event will prevent the isolation. Each pair of switches is fed by a

pressure transmitter which is piped to the drywell air space.

[7.3-42]

When the ACAD dilution air injection subsystem was abandoned, the ACAD isolation valves that the electronic pressure switches isolated were abandoned in

place and physically deactivated. Each pair of pressure switches will still cause a

Drywell Hi Pressure annunciator in the control room to illuminate. However, the

Group 6 isolation function is no longer active. Reference UFSAR Section 6.2.5.

G. Primary containment high radiation is monitored by two detector assemblies mounted in penetrations outside the drywell which feed two non-indicating

radiation switches, each with two contacts, mounted in racks in the control room.

Each switch is fed from a separate radiation sensor which is part of the QUAD CITIES - UFSAR Revision 10, October 2009 7.3-35 containment atmosphere monitoring (CAM) system. The switches are physically separated and electrically connected to the isolation control system.

[7.3-43] H. High temperatures in the vicinity of the RCIC turbine are sensed by four temperature switches arranged in a one-out-of-two-twice logic.

[7.3-44] I. High flow in the RCIC turbine steam line is sensed by two differential pressure switches which monitor the differential pressure across a 90 o elbow installed in the RCIC turbine steam supply pipeline. The tripping of either channel initiates

isolation of the RCIC turbine steam line following a time delay of 3 to 9 seconds (analytical limit). This is an exception to the usual sensor requirement. The

reason for the exception was given in the explanation of the RCIC turbine high

steam flow isolation function.

[7.3-45]

J. Low pressure in the RCIC steam line is sensed by four pressure switches upstream of the RCIC turbine line isolation valves. The switches are electrically

connected as a "1 of 2 twice" trip logic. The four pressure switches will actuate to

energize only one trip system.

[7.3-46]

K. High temperature in the area of the HPCI turbine is sensed by four (4) temperature switches arranged in a two out-of-two logic in each of the two trip

channels.

[7.3-47]

L. High flow in the HPCI turbine steam line is sensed by two differential pressure switches which monitor the differential pressure across a 90 o elbow installed in the HPCI turbine steam pipeline. Each transmitter provides an input to two (2)

trip units in the analog trip system. Each trip unit controls one of the HPCI

steam supply line isolation valves. This is an exception to the usual sensor

requirement. The reason for the exception was given in the explanation of the

HPCI turbine high steam flow isolation function.

[7.3-48]

M. Low pressure in the HPCI turbine steam line is sensed by four pressure transmitters monitoring upstream of the isolation valves. Each transmitter

provides input to a trip unit in the analog trip system. The trip units are

arranged in two trip systems (two trip units per trip system), with each trip

system connected to one HPCI turbine steam line valve. Both trip units in a trip

system must activate to isolate a steam line valve.

[7.3-49]

N. The relay contacts in each trip channel are arranged in a 1-out-of-2-twice on high drywell pressure and low reactor pressure for the HPCI vacuum breaker

isolation logic. Each trip channel closes 1 of the 2 steam supply or vacuum

breaker isolation valves.

O. RWCU Piping Area High Temperature is sensed by five RTDs per channel. Two RTDs are located in the RWCU Heat Exchanger room, one in the Phase

Separator Tank Area, and two in the "D" Heater Bay. The RTDs provide input

signals to analog trip units located in the reactor building. Any one of the five

RTDs in each of the two channels can initiate an automatic isolation.

[7.3-50] P. Reactor vessel high pressure is sensed by two pressure switches from two different taps on the "B" recirculation loop suction line piping. The pressure switches are electrically connected to a common relay that provides contacts for both the inboard and outboard RHR shutdown cooling suction valves. The same pressure switches provide contacts to the logic controlling the 1(2)-1001-29A and B shutdown cooling injection valves. These contacts provide logic input when pressure is below the shutdown cooling permissive pressure.

QUAD CITIES - UFSAR Revision 10, October 2009 7.3-36 Sensor trip channel and trip logic relays are high reliability relays equivalent to type HFA relays made by GE. The relays are selected so that the continuous load will not exceed 50%

of their continuous duty rating.

[7.3-51]

The physical and electrical arrangement of the primary containment and reactor vessel

isolation control system was selected so that no single physical event will prevent isolation.

The location of Group 1 and 2 valves inside and outside the primary containment provides

assurance that the control system for at least one valve on any pipeline penetrating the

primary containment will remain capable of automatic isolation.

Electrical cables for isolation valves in the same pipeline are routed separately. Motor

operators for valves inside the primary containment are totally enclosed and those outside

the primary containment have weatherproof-type enclosures. Solenoid valves, whether

used for direct valve isolation or as air pilots, are equipped with watertight enclosures.

All cables and valve operators can function in the most unfavorable ambient conditions

anticipated for normal operations. Temperature, pressure, humidity, and radiation are all considered in the selection of equipment for the system. Cables used in high radiation

areas have radiation-resistant insulation. Shielded cables are used where necessary to

eliminate interference from magnetic fields.

Special consideration was given to isolation requirements during a LOCA inside the

drywell. The PCIS components that are located inside the primary containment that must operate during a LOCA are the cables, control mechanisms, and the valve operators for the

isolation valves inside the drywell. Primary containment isolation system components

located within the primary containment associated with design basis events during or after

which they must perform mitigating functions are covered by the Environmental

Qualification Program Described in Section 3.11.

7.3.2.4 Design Evaluation

The primary containment isolation control system, in conjunction with other safety

systems, is designed to provide timely protection against the onset and consequences of

accidents involving the gross release of radioactive materials from the fuel and nuclear

system process barriers. It is the objective of Chapter 15 to identify and evaluate

postulated events resulting in gross failure of the fuel barrier and the nuclear system

process barrier. The consequences of such gross failures are described and evaluated in

that chapter.

[7.3-52]

The design practice for Quad Cities Station is to select tentative isolation trip settings that

are far enough above or below normal operating levels that spurious isolation and operating inconvenience are avoided. Analyses are performed to verify that the release of radioactive material following postulated gross failures of the fuel and nuclear system process barrier is kept within acceptable bounds. The Technical Specification allowable values and the

associated instrument trip setpoints have been based on the methods prescribed in NES-

EIC-20.04.

Chapter 15 shows that the actions initiated by PCIS, in conjunction with other safety

systems, are sufficient to prevent releases of radioactive material from exceeding the values

given as guidance in applicable regulations.

RWCU High Area Temperature Isolation instrumentation was installed to detect RWCU

line breaks. Credit for these instruments is not taken in any transient or accident analysis

because this line break is bounded by larger MSL or recirculation breaks. Administrative

controls are required to provide technical requirements for operability of this

instrumentation to preclude reliance on manual trips during RWCU HELB scenarios.

QUAD CITIES - UFSAR Revision 10, October 2009 7.3-37 This RWCU system isolation instrumentation mitigates the HELB to limit offsite releases and maintains HELB environmental conditions within analyzed parameters.

Temperatures in the spaces occupied by various steam lines and steam-driven equipment outside the primary containment are the only essential variables of significant spatial

dependence that provide inputs to PCIS. The large number of temperature sensors and their

dispersed arrangement near the steam lines requiring this type of break protection provides assurance that a significant break will be detected rapidly and accurately. One of the four groups of main steam line space temperature switches is located in the ventilation exhaust from the steam line space between the drywell wall and the secondary containment wall. This

assures that abnormal air temperature increases are detected regardless of the location of a

leak in that space.

Section 15.6 evaluates a gross breach in the main steam line outside the primary containment

during operation at rated power. The evaluation shows that the main steam lines are automatically isolated in time to prevent a release of radioactive material in excess of the

values given as guidance in applicable regulations and to prevent the loss of coolant from being great enough to allow uncovering of the core. These results are true even if the longest

closing time of the valve is assumed.

The shortest closure time of the main steam line isolation valves is 3 seconds. The transient

resulting from a simultaneous closure of all main steam line isolation valves in 3 seconds

during reactor operation at rated power is considerably less severe than the transient

resulting from inadvertent closure of the turbine stop valves (elapsed time approximately 0.1

seconds) coincident with failure of the turbine bypass system (see Section 15.2.3.1).

Because essential variables are monitored by trip channels arranged for physical and

electrical independence, and because a dual trip system arrangement is used to initiate closure of automatic isolation valves, no single failure, maintenance operation, calibration

operation, or test can prevent the system from initiating valve closure, for Groups 1, 2, 3, and the RHR shutdown cooling isolation. An analysis of the isolation control system shows that the system does not fail to respond to essential variables as a result of single electrical failures

such as short circuits, grounds, and open circuits. These single failures result in a failure of

only one trip system. Isolation is initiated upon a trip of the remaining trip system.

The Group 4 and 5 isolation circuits each contains two normally de-energized trip systems.

These systems isolation valves remain open unless a line break in the respective HPCI or

RCIC system is sensed. The HPCI and RCIC logic systems were not originally designed to

meet single failure criteria because of the redundancy of the core cooling systems. The Group

4 and 5 dual trip systems do provide a level of redundancy and reliability for mitigation of a

high energy line break. The three sensor functions within the trip systems provide redundant

methods of detecting a line break. The RCIC low steam supply pressure isolation is the only

exception in that the instruments will only trip one trip system. This is acceptable as this function is considered an operational interlock for turbine operation within the reactor

building, and this low pressure isolation is only a backup to the other line break isolation

detection instrument channels. The Group 4 isolation logic only has been updated by

Modification M04-1(2)-91-013B to include the electrical and where possible physical

separation of the IEEE-384 and Regulatory Guide 1.75.

[7.3-52a]

The RWCU High Temperature Auto Isolation system was added to eliminate the reliance on

operator action for manual isolation for an RWCU HELB scenario described in GE SIL No. 604. The original licensing analysis for RWCU pipe breaks was found adverse to quality

and a commitment was made to the NRC to provide a high area temperature based actuation

for RWCU. This new isolation actuation was designed to meet the intent of the requirements

as defined for a plant protection system and is consistent with IEEE 279-1968 criteria.

[7.3-53]

QUAD CITIES - UFSAR Revision 10, October 2009 7.3-37a The reactor vessel pressure high function provides equipment protection to prevent an RHR intersystem LOCA scenario. No credit for this interlock is assumed in any accident or transient analysis.

The redundancy of trip channels provided for all essential variables provides a high

probability that whenever an essential variable exceeds the isolation setting, the system will

initiate isolation. In the unlikely event that all trip channels for one essential variable in one

trip system fail in such a way that a system trip does not occur, the system could still respond properly as other monitored variables exceed their isolation settings. In addition, isolation of

the process lines could be accomplished manually by the operator.

QUAD CITIES - UFSAR Revision 10, October 2009 7.3-38 The sensors, logic and circuitry used for primary containment isolation system are not used for any process systems, where the malfunction of these process systems will prevent a

containment isolation, when an isolation is warranted.

The wall of the primary containment effectively separates adverse primary containment

environmental conditions which might otherwise affect both isolation valves in a pipeline.

Therefore, environmental conditions inside the drywell will not affect the ability to isolate a

given line. The previously discussed electrical isolation of control circuitry prevents

failures in one part of the control system from propagating to another part. See Section 8.3

for electrical distribution information. Electrical transients have no significant effect on

the functioning of the isolation control system.

Calibration and test adjustments for pressure and level switches are located on the

switches themselves. These switches are located in the turbine building, reactor building, and cable spreading room. To gain access to the adjustments on each switch, a cover plate, access plug, or sealing device must be removed by personnel before any adjustment in trip

settings can be effected. Calibration and maintenance of instruments are done in

accordance with approved plant procedures with the approval of the shift engineer to

reduce the probability that operational reliability will be degraded by operator error.

[7.3-54]

The various power supplies used for the isolation system logic circuitry and for valve operation provide assurance that the required isolation can be accomplished in spite of power failures. If ac power for valves inside the primary containment is lost, dc power is

available for operation of valves outside the primary containment. The main steam

isolation valve control arrangement will not inhibit the isolation function due to the loss of

ac and/or dc power. Because both solenoid-operated pilot valves must be de-energized, loss

of a single power supply will neither cause inadvertent isolation nor prevent isolation if

required. The logic circuitry for Groups 1, 2, 3, and the RHR shutdown cooling isolation is powered by separate reactor protection system (RPS) buses for separate divisions. The power supplies for the Group 4 isolation channels are: 1.) Division II 125 VDC with

transfer to Division I on loss of power, and 2.) Division II 120 VAC. A loss of a single RPS

bus power here results in a single trip system trip. In no case does a loss of a single power

supply prevent isolation.

7.3.2.5 Inspection and Testing

All parts of the PCIS are testable during reactor operation. Isolation valves can be tested to

assure that they are capable of closing by operating manual switches in the control room and observing the position lights and any associated process effects. Testable check valve

controls are designed to allow verification that valve disks are free to open and close. The

trip channel and trip system responses can be functionally tested by applying test signals to

each trip channel and observing the trip system response. Testing of the main steam line

isolation valves is discussed in Section 6.2.6.3.

[7.3-55]

7.3.2.6 Conformance to IEEE-279

The following is a point-by-point comparison of the containment isolation control system

with the Requirements of IEEE Std 279-1968 which has been summarized from GE Topical

Report, NEDO-10139.[1] For more detailed information refer to the topical report.

[7.3-56]

QUAD CITIES - UFSAR Revision 9, October 2007 7.3-39 7.3.2.6.1 General Functional Requirements (IEEE-279, paragraph 4.1)

A. Auto-Initiation of Appropriate Action

The control system action from sensor to final control signal to the valve actuator is capable of initiating appropriate action and of doing it in a time commensurate

with the need for valve closure. Total time, from the point where a process out-

of-limits condition is sensed to the energizing or de-energizing of appropriate valve actuators, is less than 200 milliseconds (logic response time excluding sensor). The closure time of valves ranges upward from a minimum of 3 seconds for the main steam isolation valves, depending upon the urgency for isolation

considering possible release of radioactivity. Thus it can be seen that the control

initiation time is at least an order of magnitude lower than the minimum

required valve closure time.

B. Precision

Accuracies of each of the sensing elements is sufficient to accomplish the isolation initiation within required limits without interfering with normal plant

operation.

C. Reliability

The reliability of the PCIS is compatible with and higher by at least an order of magnitude than the reliability of the actuated equipment (valves).

D. Action Over the Full Range of Environmental Conditions

The similar item listed under core spray (Section 7.3.1.1.1.1) applies here in all respects to all isolation control equipment, except the manual control switches

for the HPCI and RCIC isolation valves. Since both of the control switches for

the redundant valves are in the same control panel in the main control room, it

is conceivable that destruction of this cabinet by fire or missile could affect the

control of both valves in these two lines in such a way as to prevent them from

closing. However, it is highly unlikely that such an event could occur

coincidentally with an independent event requiring system isolation such as a

steam line break. Refer to Commonwealth Edison's, 10 CFR 50, Appendix R, Program and UFSAR, Section 3.5.

7.3.2.6.2 Single Failure Criterion (IEEE-279, paragraph 4.2)

The single failure criterion of IEEE-279 is fully complied with in the design of the PCIS.

7.3.2.6.3 Quality of Components and Modules (IEEE-279, paragraph 4.3)

See Section 7.3.1.1.1.3 which also applies to the PCIS, with the exception that most of the

isolation control is de-energize to trip, instead of energized to trip, and is thus more likely

to call attention to the failures that may occur in coil circuits, connections, or contacts.

QUAD CITIES - UFSAR Revision 5, June 1999 7.3-40 7.3.2.6.4 Equipment Qualifications (IEEE-279, paragraph 4.4)

See Section 7.3.1.1.1.4 which also applies to PCIS.

7.3.2.6.5 Channel Integrity (IEEE-279, paragraph 4.5)

See Section 7.3.1.1.1.5 which also applies to PCIS. However, the fail-safe design of the

isolation control and operation of a grounded ac system makes it less likely to fail to

operate.

7.3.2.6.6 Channel Independence (IEEE-279, paragraph 4.6)

Channel independence for sensors exposed to each process variable is provided by electrical

and mechanical separation. Physical separation is maintained between redundant elements of the redundant control systems where it will add to reliability of operation. The

manual control switches for the HPCI and RCIC isolation valves are an exception to this

objective, but they are sufficiently separated to give a high degree of reliability and meet a

literal interpretation of paragraph 4.6 of IEEE-279.

7.3.2.6.7 Control and Protection Interaction (IEEE-279, paragraph 4.7)

The isolation control system is a strictly on-off system, and no signal whose failure could

cause a need for isolation can also prevent it.

7.3.2.6.8 Derivation of System Inputs (IEEE-279, paragraph 4.8)

The inputs which initiate isolation valve closure are direct measures of variables that

indicate a need for isolation (such as reactor vessel low level, drywell high pressure, and

pipe break detection). Pipe break detection utilizes methods of recognition of the presence

of a material that has escaped from the pipe, rather than detecting actual physical changes

in the pipe itself.

7.3.2.6.9 Capability for Sensor Checks (IEEE-279, paragraph 4.9)

The reactor vessel instruments can be checked one at a time by application of simulated signals. These include level, pressure, radiation, and flow. Temperature sensors along the

main steam lines are not testable except during shutdown, but they are sufficient in number so that testing between refueling outages is not necessary to achieve the reliability

level required. Temperature sensors can be checked periodically by removing them and

applying heat to the sensitive zone, and also by oven calibration, which requires removal

from the circuit during calibration and replacement by calibrated units.

QUAD CITIES - UFSAR Revision 6, October 2001 7.3-41 7.3.2.6.10 Capability for Test and Calibration (IEEE-279, paragraph 4.10)

All active components of PCIS, with the exception of the main steam line high temperature

sensors and the main steam line radiation sensors, can be tested and calibrated during

plant operation.

The radiation sensors can be cross-checked against their companions for verification of

operability and since they are used with reference to background, they do not require actual

sensitivity verification on a frequent basis.

7.3.2.6.11 Channel Bypass or Removal from Operation (IEEE-279, paragraph 4.11)

Calibration of each sensor will introduce a single instrument channel trip, except in the case of the Unit 1 RWCU Automatic Isolation Area RTDs which can be placed in bypass during calibration. The introduction of a single instrument channel trip during calibration

does not cause a protective function without the coincident trip of at least one other

instrument channel, except in the case of the HPCI and RCIC where leak detection

temperature sensors have one-out-of-two logic on differential temperature and where leak

detection flow sensors have one-out-of-two logic. The RWCU Automatic Isolation on Area

Temperature Hi is also a one-out-of-two logic.

[7.3-57]

7.3.2.6.12 Operating Bypasses (IEEE-279, paragraph 4.12)

The only bypasses in PCIS are the main steam line low-pressure bypass and the main steam line tunnel temperature switches. The main steam line low-pressure bypass is

imposed by the mode switch when not in the run mode. The mode switch cannot be left in

this mode with neutron flux measuring power above 15% of rated power without imposing a

scram. Therefore the bypass is considered to be removed in accordance with the intent of

IEEE-279, although it is a manual action that removes it rather than an automatic one. In

the case of the motor operated valves, automatic or manual closure can be prevented by

shutting off electric power. The MSIV steam tunnel temperature bypass switches located

on the main control board 901(2)-4 allow the RWCU to continue to run during MSIV

temperature switch calibration and testing.

7.3.2.6.13 Indication of Bypasses (IEEE-279, paragraph 4.13)

The bypass of the main steam line low-pressure isolation signal is not indicated directly in

the control room except by the position of the mode switch handle. This switch is under

strict operator control. Its specific bypass functions are a matter of operator training and, as such, do not reasonably need to be brought to the operator's attention each time he places the switch in startup mode. Since the bypass is not removed by any automatic action

it is positively in effect any time the mode switch is in position to impose it.

For bypass of the MSIV room area temperature switches, individual indicating lights are

provided on a new RWCU isolation area temperature monitoring panels. Annunciator

indication is provided when the switches are placed in the bypass configuration.

QUAD CITIES - UFSAR Revision 6, October 2001 7.3-42 7.3.2.6.14 Access to Means for Bypassing (IEEE-279, paragraph 4.14)

The mode switch affects the main steam line low pressure PCIS function, and it is centrally

located on the operators main control console.

Two handswitches, one for inboard isolation logic and one for outboard isolation logic, are

installed on the 901(2)-4 panels to allow bypassing the MSIV room temperature relays contacts This is necessary to allow for RWCU operation during a shutdown when MSIV

room temperature switches are removed from service, and the RWCU system is required for

outage related operation.

7.3.2.6.15 Multiple Trip Settings (IEEE-279, paragraph 4.15)

Paragraph 4.15 of IEEE-279 is not applicable because all setpoints are fixed.

7.3.2.6.16 Completion of Protection Action Once Initiated (IEEE-279, paragraph 4.16)

All isolation decisions are sealed-in downstream of the decision making logic, so valves go to

the closed position, which ends protective action. Manual reset action is provided by a

three-position reset switch, so that inboard valves can be reset independent of outboard

valves.

7.3.2.6.17 Manual Actuation (IEEE-279, paragraph 4.17)

All isolation valves are capable of manual actuation independent of active components of the automatic actuation circuitry, with the exception of the motor starters for the motor

operated valves.

7.3.2.6.18 Access to Setpoint Adjustments (IEEE-279, paragraph 4.18)

The discussion given in Section 7.3.1.1.1.18 is also applicable to PCIS.

7.3.2.6.19 Identification of Protective Actions (IEEE-279, paragraph 4.19)

The statements made in Section 7.3.1.1.1.19 are applicable to PCIS.

7.3.2.6.20 Information Readout (IEEE-279, paragraph 4.20)

The information presented to the operator are as follows:

A. Annunciation of each process variable which has reached a trip point,

B. Computer readout of trips on main steam line tunnel temperature or main steam line excess flow, QUAD CITIES - UFSAR Revision 6, October 2001 7.3-42a C. Control power failure annunciation on each channel, QUAD CITIES - UFSAR Revision 9, October 2007 7.3-43 D. Annunciation of steam leaks in each of the five systems monitored such as, main steam, reactor water cleanup, RHR, HPCI, and RCIC; and

E. Open and closed position lights for each isolation valve. This information is considered to fulfill the requirements for information readout.

7.3.2.6.21 System Repair (IEEE-279, paragraph 4.21)

Those components which are expected to have a moderate need for replacement are

designed for convenient removal. Pressure sensors, vessel level sensors, etc. can be

replaced in a reasonable length of time, but these devices are considered to be permanently

installed although they have nonwelded connections at the instrument, which will allow

replacement.

7.3.3 Secondary Containment Isolation System

The objective of the secondary containment system, in conjunction with other systems, is to

limit the release of radioactive materials to be below the limits in 10 CFR 100 (or 10 CFR 50.67 as applicable) as a result of the design basis accidents. For more information on the design basis refer to Section 6.2.

[7.3-58]

The secondary containment isolation includes:

A. Closing the reactor building ventilation isolation valves;

B. Tripping the reactor building supply and exhaust fans; and

C. Starting the standby gas treatment system (SBGTS).

The initiating signals are:

A. Low reactor water level using a one-out-of-two-twice logic;

B. High drywell pressure using a one-out-of-two-twice logic;

C. High reactor building ventilation exhaust radiation using a one-out-of-two logic;

D. High refuel floor radiation using a one-out-of-two logic;

E. Reactor building ventilation radiation monitors downscale using a two-out-of-two logic; F. Refuel floor radiation monitors downscale using a two-out-of-two logic; and

G. High drywell radiation using two-out-of-two logic.

The reactor building ventilation isolation and fan trip are actuated via auxiliary contacts

from the SBGTS logic. See Section 6.5 for more information on SBGTS.

QUAD CITIES - UFSAR Revision 5, June 1999 7.3-44 7.3.4 References

1. General Electric Topical Report NEDO-10139, June 1970.

(Sheet 1 of 1)

Revision 10, October 2009 QUAD CITIES - UFSAR Table 7.3-1 ANALYTICAL LIMITS FOR GROUP ISOLATION SIGNALS Valve Isolation Group Isolation Signal Analytical Limit

[Note 1] Group 1 Reactor Low-Low Water Level >

-59 in. Steamline High Flow <

140% of rated flow Steamline Low Pressure >

785 psig in RUN mode Steam Tunnel High Temperature <200°F Group 2 Reactor Low Water Level >

0 in. Drywell High Pressure <

+2.5 psig Drywell High Radiation <

100 R/hr Group 3 Reactor Low Water Level >

0 in. Steam Tunnel High <200°F RWCU Area High Temperature <185°F Group 4 HPCI Steamline Low Pressure >

100 psig HPCI Steam Supply Valves

ONLY HPCI Steamline High Flow HPCI Area High Temperature

<300% rated flow

<170°F Group 4 HPCI Steamline Low Pressure >

100 psig

  • HPCI Turbine Exhaust Vacuum Breaker Valves

ONLY Drywell Pressure High

  • Signals existing simultaneously

<2.5 psig

  • Group 5 RCIC Steamline Low Pressure >

50 psig RCIC Steamline High Flow <

300% rated flow RCIC Area High Temperature <170°F RHR Shutdown Cooling Reactor High Pressure [Note 2]

> 135 psig Note 1: Analytical Limit shown unless noted otherwise Note 2: Pressure sensed on Reactor Recirculation loop B suction line

QUAD CITIES - UFSAR Revision 5, June 1999 7.4-17.4. SAFE SHUTDOWN The following section describes the instrumentation and control system aspects of the containment cooling mode of the residual heat removal (RHR) system. This section also

provides a description of shutdown outside the control room.

[7.4-1]

7.4.1 Containment Cooling Mode of the Residual Heat Removal System The containment cooling function is provided by the residual heat removal (RHR) system

after the core is flooded. Suppression pool water can be recirculated through the heat

exchangers for cooling. The cooled water can be used to spray the drywell and/or torus. For

a complete description of the design basis, system functions and components, refer to

Section 6.2

[7.4-2]

Containment cooling mode of RHR is initiated manually from the control room by

alignment of the proper combination of valves, pumps, and heat exchangers. No automatic

start function is provided. When a LPCI initiation signal is present, the use of the containment cooling permissive switch is required for containment cooling valve alignment

and the RHRSW permissive switch is required to start RHRSW.

However, in order to initiate or maintain containment cooling, the following conditions

must be met or the signal bypassed by use of the containment cooling 2/3 level and ECCS

initiation bypass switch:

A. Reactor water level inside the core shroud must be at least 2/3 core height. This parameter is measured by one level transmitter per division.

B. Reactor water level inside the annulus is above the ECCS initiation setpoint.

This parameter is measured by two level switches per division arranged in one-

out-of-two twice logic.

C. Drywell pressure is below the ECCS initiation setpoint. This parameter is measured by two pressure switches per division arranged in one-out-of-two twice

logic.

Additionally, to initiate or maintain drywell and/or torus spray the following condition must

also be met:

A. Drywell pressure is above the low limit setpoint. This parameter is measured by two pressure switches per division arranged in one-out-of-two twice logic.

This additional condition does not have a bypass switch. Once containment cooling has been placed in operation, if any of the preceding requirement do not continue to be either met or bypassed, the associated valves will close to allow full LPCI flow.

7.4.2 Shutdown Outside the Control Room In the unlikely event that the control room becomes uninhabitable, provisions have been

made to permit shutdown of the reactor outside of the control room. A number of QUAD CITIES - UFSAR Revision 5, June 1999 7.4-2 automatic features incorporated in the plant design allow the reactor to be brought to a safe shutdown condition. The following description outlines a course of action which achieves a

safe and orderly cold shutdown condition. Alternate action sequences are possible.

[7.4-3]

Immediately prior to control room evacuation, the operator actuates the reactor manual

scram switches on the control panel to insert all control rods, and observes the control rod

position indicators on the display panel. The control rods may also be inserted from outside

the main control room by several methods. One method is to manually trip both reactor

protection system (RPS) motor-generator (M-G) sets by opening the power supply circuit breakers at the 480 V motor control centers in the turbine building. The position of the

scram valves for the individual control rod drives can be verified in the reactor building at the control rod drive modules. Table 7.4-1 lists key parameters available outside the

control room and their locations.

Reactor vessel pressure and water level are indicated locally in the reactor building on

instrument racks 2201(2)-5 and 2201(2)-6 and other racks as indicated in Table 7.4-1. The

steam pressure regulator will continue to automatically regulate reactor pressure by

allowing steam flow through the main turbine and its bypass system to the condenser.

Decay heat from the reactor will continue to be dissipated to the condenser through the turbine system until the turbine generator trips. At which time the turbine bypass valves

will open and dump steam directly to the main condenser. Steam dumping to the

condenser continues until the amount of decay heat being generated within the core is not

sufficient to maintain reactor pressure. Thermal losses from the reactor system, combined with the normal steam flow to the turbine gland seals and air ejector, will eventually

exceed the decay heat and result in a gradual cooldown and depressurization of the reactor

to approximately 850 psig at which time the main steam isolation valves will close

automatically.

The operator will continue reactor vessel depressurization and cooldown by remote-manual

actuation of the relief valves resulting in blowdown to the suppression pool. The number of

valves, and the opening frequency and duration, will be determined by monitoring the

reactor pressure at instrument rack 2201(2)-5 to insure that the vessel cooldown rate does

not exceed 100~F per hour. Remote-manual actuation of the relief valves is accomplished by closing the contacts on the relief valve controllers which are also located on instrument

rack 2201(2)-5.

While the reactor is blown down to the suppression pool, one RHR pump, heat exchanger, and RHR service water pump may be placed in service to cool the suppression pool water

and prepare for shutdown. The equipment, motor-operated valves, pumps, etc., may be actuated manually in the reactor building and at appropriate breakers at the 480-V motor

control centers and 4160-V switchgear as required. Once the reactor has been

depressurized to approximately 50 psig, the RHR system is placed in the shutdown cooling

mode and reactor cooldown will continue.

The required communications for accomplishing this shutdown can be maintained outside

the control room using remote phone equipment, sound powered telephones, two-way

radios, etc. (see Section 9.5.2). During the entire shutdown process, no re-entry into the

main control room is required. Instrumentation outside the control room enables the

operator to monitor the reactor vessel level, pressure, and temperature during cooldown.

Therefore, a safe operational shutdown of the reactor from a normal operating condition to

a cold shutdown condition can be accomplished without access to the main control room.

(Sheet 1 of 1)

QUAD CITIES - UFSAR

Table 7.4-1

REACTOR VESSEL PRESSURE AND LEVEL INDICATORS AVAILABLE OUTSIDE THE CONTROL ROOM

Variable Monitored Location Reactor Pressure Rack 2201(2)-5,6 Reactor Pressure Panel 2201(2)-70A Reactor Pressure Panel 2201(2)-70B Reactor level Rack 2201(2)-5,6 Reactor level Rack 2201(2)-7,8 Reactor level Panel 2201(2)-73 A,B

QUAD CITIES - UFSAR Revision 7, January 2003 7.5-17.5DISPLAY INSTRUMENTATION The following section describes display instrumentation required by the operator for normal operation and safe shutdown of the unit, including post-accident conditions.

Included is a discussion of instruments meeting the requirements of Regulatory Guide

1.97[1], a description of the safety parameter display system, and a summary of the detailed control room design review.7.5.1Post-Accident Monitors Certain instruments have been designated as post-accident monitors, and as such have been determined to comply with Regulatory Guide 1.97

[1]. These instruments are identified in the station's work control system data base.

[7.5-1]7.5.1.1Description Post-accident monitoring instruments are assigned to meet one of three design categories

described in detail in Regulatory Position 1.3. Category 1 requirements are the most

stringent, with requirements very similar to safety-related instruments. Category 2

requirements are not quite as stringent, but many of the same standards are

recommended. Category 3 instruments are commercial grade.In accordance with Regulatory Guide 1.97, process variables used in post-accident monitoring are grouped into 5 types: A, B, C, D, and E.

Type A, those variables to be monitored that provide the primary information required to permit the control room operators to take the specific manually

controlled actions for which no automatic control is provided and are required for

safety systems to accomplish their safety function for design basis accident events.

Primary information is information that is essential for the direct accomplishment

of the specified safety functions; it does no include those variables that are

associated with contingency actions that may also be identified in written

procedures. A variable included as Type A does not preclude it from being included

as Type B, C, D or E, or vice versa.Type B, those variables that provide information to indicate whether plant safety functions are being accomplished. Plant safety functions are (1) reactivity control

(2) core cooling (3) maintaining reactor coolant system integrity, and (4)

maintaining containment integrity (including radioactive effluent control).

Variables are listed with designated ranges and category for design and

qualification requirements. Key variables are indicated by design and qualification

Category 1.

Type C, those variables that provide information to indicate the potential for being breached or the actual breach of the barriers to fission product releases. The

barriers are (1) fuel cladding, (2) primary coolant pressure boundary, and (3)

containment.

QUAD CITIES - UFSAR Revision 7, January 2003 7.5-2 Type D, those variables that provide information to indicate the operation of individual safety systems and other systems important to safety. These variables

are to help the operator make appropriate decisions in using the individual systems

important to safety in mitigating the cause of an accident.

Type E, those variables to be monitored as required for use in determining the magnitude of the release of radioactive materials and in continually assessing such

releases.Type A, B, and C variables relate to the determination of the safety condition of the plant and provide the operator with the information to perform tasks needed to mitigate

accidents. The following parameters have been identified as Type A variables per

Regulatory Guide 1.97

[1]:1.Coolant level in the reactor; 2.Reactor pressure; 3.Drywell pressure; 4.Suppression chamber pressure; 5.Suppression pool water level; and 6.Suppression pool water temperature.

The instruments monitored by these variables meet the intent of Category 1 requirements per Regulatory Guide 1.97

[1], or deviations from these requirements have been justified.

The station's work control system data base identifies the instrument numbers and the variable type associated with these parameters.

The seismic criteria are described in Section 3.10.7.5.1.2Analysis A review of the post-accident monitoring instruments indicated that Quad Cities Station is in compliance with the intent of Regulatory Guide 1.97. Control room instrumentation

provides sufficient information for operators to identify, mitigate, and monitor all design

basis accidents.

The following sections provide details of Quad Cities acceptability with respect to seismic, power, environmental, and separation requirements.7.5.1.2.1Seismic QualificationSafety-related instruments installed prior to Regulatory Guide1.97 that either fulfilled the requirements of Regulatory Guide 1.97, Revision 2, Category 1, or were previously

designated as seismic by the 1980 FSAR Safety-Related and ASME Classification Valve, Equipment, and Instrument List, the Master Equipment List, or the instrument data QUAD CITIES - UFSAR 7.5-3 sheets, did not undergo further seismic qualification. Replacement instruments, or new instruments installed to meet Regulatory Guide 1.97, meet the seismic requirements of IEEE 344-1975 and station requirements. Safety-related instrument racks have been seismically upgraded by adding bracing as required (refer to Section 3.10).7.5.1.2.2Environmental QualificationIn order to show that electrical equipment important to safety is capable of functioning in a harsh environment, CECo provided a response to IEB 79-01B for Quad Cities Station Units

1 and 2. Environmental zone maps were established which identified the temperature, pressure, and radiation values in various locations of the station (refer to Section 3.11).

That equipment which performed a safety-related or Regulatory Guide 1.97 Category 1 or

Category 2 function, and was required to function while exposed to the harsh environment following the associated design basis event, was included in the program to be

environmentally qualified for its respective location. Equipment located in a mild

environment, regardless of its function, was not required to have additional environmental

qualification over and above its required service conditions.

The analysis applied the 10 CFR 50.49.k rule allowing the use of instrumentation qualified under the IEB 79-01B program. Instruments not covered under the IEB 79-01B program, but required to fulfill Category 1 or Category 2 requirements of Regulatory Guide 1.97, are

qualified under the station environmental qualification (EQ) program (refer to Section

3.11).Required instrument cables are included in the environmental qualification program.

Under this program, the cable tabulations were correlated to catalog instrument cables by

manufacturer and cable type. The purchase specifications for these cable types were then

reviewed to identify the approved vendors. The environmental qualification (EQ) program

included original station design instrumentation cable.7.5.1.2.3Redundancy of Power Power sources for instrumentation have been verified for their ability to provide power under post-accident conditions. Each instrument bus has a main source and at least one

backup or reserve source of power. See Section 8.3 for power supply information.

Each Category 1 variable is redundant to ensure that at least one channel is available to provide the necessary information to the operator. Instrumentation for every Category 1

variable, with the exception of valve position indication, has a redundant loop that receives

power from an alternate bus.

Neither Category 2 nor Category 3 instrumentation requires redundant monitoring channels. Therefore, only one power source for these categories of monitoring

instrumentation is required. Even though this station received its construction permit

prior to the categorization of power sources as Class 1E or non-1E, the power sources and

the reserve sources provide the required reliability to meet the intent of Regulatory Guide

1.97.This station was licensed before Regulatory Guide 1.75 established the requirements for physical independence of electrical systems. Existing instrumentation used for post-QUAD CITIES - UFSAR Revision 11, October 2011 7.5-4 accident monitoring does not follow these separation requirements. New instrument loops added after July 31, 1985 to fulfill a Category 1 requirement comply with the requirements

of Regulatory Guide 1.75.7.5.2Process Computer The process computer supplies information to the operators via video displays and printers.

7.5.2.1Description The process computer system (PCS) consists of the hardware and software necessary to run various nuclear steam supply system (NSSS) and balance of plant (BOP) programs.

Quad Cities Units 1 and 2 utilize separate process computer systems. An interface exists

between the computer systems for each unit to facilitate the transfer of data from inputs

common to both units. Components duplicated for both units include the control room request and output devices. Color graphics displays have been added to PCS in order to

display the status of pipelines, valves, pumps, and other generating station data.

Historical plant data is maintained on a separate computer.

[7.5-2]Each unit has devices which allow requests for information to be entered. In addition, devices for displaying alarms and requested information are provided. An audible alarm

horn and reset button are also provided.

[7.5-3]The basic PCS consists of a distributed process computer system that provides on-line monitoring of over 1500 input points (digital, pulse, and analog) representing significant

plant process variables. The system scans digital and analog inputs at specified intervals and issues appropriate alarm indications and messages if monitored analog values exceed

predefined limits or if digital trip signals occur. It performs calculations with selected input data to provide the operator with essential core performance information through a

variety of logs, trends, displays, and summaries. The computer outputs include various

front panel displays (digital lights, trend recorders and color graphic displays).

In general, the process computer system drives all peripherals that display or log real time data, while a separate computer drives all devices, which run the nuclear program for core

calculations. Typical peripherals include operator workstations, printers and color graphic

displays.A separate computer system is used to provide historic data storage and retrieval functions, and to support data link access to offsite users. The computer used is a high

performance workstation with a large disk storage capacity. Data is transferred from the

PCS over a high speed data link.

Core performance calculations are performed on a separate computer system. This program is updated each refueling outage with data from the fuel vendor. This data is

evaluated during initial power ascension subsequent to the outage and is approved along

with other startup tests by On-site Review.

Computer data is available in station emergency facilities.

QUAD CITIES - UFSAR Revision 11, October 2011 7.5-5 Supporting the emergency plan is real time on-line computer software which uses plant parameters, meteorological data, and radiation monitoring inputs to aid in determining

accident classifications. (Refer to Section 13.3.)

The computer is powered by an uninterruptible power supply (UPS). (See Section 8.3.)

[7.5-5]The PCS includes hardware and software necessary for the safety parameter display system (SPDS).

[7.5-6]7.5.2.2Operator Functions This section describes various on-demand programs available to the operator.

Two trend recorders are available. The status of these recorders can be checked. Points can be assigned and the range of the trend can be selected or canceled.

Various methods of displaying information to the operator are provided by the system.A.Tabular displays of selected points showing current values and point states.BTabular displays of selected points showing historical data.

C.Graphical display of selected points in various formats.

D.Summary displays of alarms which can be filtered if desired.

E.System health summary displays showing the state of various computer sub-systems and programs.

A program for collecting, displaying, and transferring TIP data for calibrating Local Power Range Monitors (LPRMs).(See Section 7.6)

Methods for alarming points at multiple levels for both high and low alarms.

Tools to allow the operator to set alarms other than the predefined alarm setpoints.

Tools to allow the operator to inhibit alarms and substitute values for bad data.

Historical archives that save all data at 1 second and a subset of data at 10 millisecond intervals for retrieval. The size of the 1-second archive will be at least 14 days and the 10-

millisecond archive will be at least 3 days.

Tools to allow historical archives to be saved for future retrieval of plant data from events of interest.

An event re-call archive that will provide for data archives to be created for events of interest when a specified triggering event occurs.

A method to calibrate computer points.

[7.5-7 thru 7.5-13]

QUAD CITIES - UFSAR Revision 11, October 2011 7.5-67.5.3Safety Parameter Display System Supplement 1 of NUREG 0737 required all operating plants to provide a Safety Parameter Display System (SPDS) in the control room. The purpose of SPDS is to provide a concise

display of critical plant variables to aid in rapidly and reliably determining the safety

status of the plant. NUREG 0737 required that SPDS provide, as a minimum, information

concerning:

[7.5-14] A.Reactivity Control;B.Reactor core cooling and heat removal from the primary system; C.Reactor coolant system integrity; D.Radioactivity control; and E.Containment conditions.

These functions have been designated as Critical Safety Functions. The parameters required for these functions include:A.Reactivity control1.Average power range monitor 2.Source range monitor QUAD CITIES - UFSAR 7.5-7B.Core cooling1.Reactor water level 2.Core spray system statusC.Reactor coolant system integrity1.Reactor vessel pressure 2.Drywell pressure 3.Containment activity 4.Safety Relief Valve (SRV) Position 5.Isolation valve statusD.Radioactivity control1.Main stack monitor 2.Off-gas pretreatment monitor 3.Reactor building ventilation radiation monitor 4.Liquid discharge monitorsE.Containment conditions1.Drywell pressure 2.Drywell temperature 3.Suppression pool level 4.Suppression pool temperature 5.Containment isolation valve status7.5.3.1Description The SPDS provides color graphics displays in the control room and technical support center for key plant parameters. The system takes its input from several sources for each

parameter and determines which sensors are valid. It then averages the valid sensors to

determine the best value for each of the displayed parameters.

[7.5-15]Colors have the following significance:A.Red-this color indicates an alarm condition with a parameter being in an abnormal state, QUAD CITIES - UFSARRevision 11, December 2011 7.5-8B.Yellow-this color indicates an alert condition,C.Cyan (Light blue)-this color means input is invalid or inoperable, and D.Green-this color indicates a normal condition of a parameter.

The SPDS at Quad Cities is a software package incorporated into the process computer, a non-safety-related system utilizing computer inputs for data. The computer has been

suitably isolated from safety-related process inputs.

[7.5-16]Invalid data to SPDS is indicated by the color cyan. With the exception of radioactive

release, all parameters are monitored by multiple sensors. When all sensors for a

parameter are lost, the bar chart or box for that parameter turns cyan. The bar charts will

indicate full scale. This does not mean that the parameter is reading full scale, but only

that the computer input for that parameter is not valid.

In addition to the main display for SPDS, one or more displays for each parameter will be

provided to assist the operator in determining what raw signals were used to determine

that parameter and what methodology was used in arriving at the final SPDS value for a

parameter.7.5.3.2Analysis A human factors review of SPDS was conducted as part of the detailed control room design review (DCRDR). The purpose of the SPDS review was to ensure that the design of the

installed SPDS complied with sound human factors engineering principles, and to verify theparameter selection by referring to the task analysis data collected during the DCRDR

and the criteria established in NUREG 0737, Supplement 1.

[7.5-17]The review evaluated the appropriateness and completeness of the information available

through the SPDS, the effectiveness of the display format and coding techniques, the

location and positioning of the CRTs in the control room, the readability of the display given hardware and environment factors, and the adequacy of procedures and

documentation for interpreting the display.

To assure that the parameters displayed on SPDS adequately monitor plant safety status during emergency conditions (which is accomplished by monitoring the critical safety

functions), a comparison was made between the DCRDR task analysis and the SPDS

display parameters.

The findings of the DCRDR evaluation confirmed that the parameters displayed on SPDS indicate the accomplishment or maintenance of plant safety functions. Discrepancies

identified during the data collection phase represented minor modifications to SPDS. The

verification and validation of SPDS confirmed that the final product adequately met the

criteria of NUREG 0737, Supplement 1.

QUAD CITIES - UFSAR Revision 3, December 1995 7.5-97.5.4Detailed Control Room Design Review The purpose of the DCRDR was to assess and evaluate the control room work space, instrumentation, controls, and other equipment from a human engineering perspective.

The process took into account both system demands and operating capacities and then

identified essential and select control room improvements which would correct inadequate

or unacceptable items. The ultimate goal was to ensure that proper human engineering

principles and practices were incorporated into the design of the control room to help

ensure the ability of control room operators to prevent accidents or cope with accidents if

they occur.

[7.5-18]The investigative process included the following elements:A.A control room survey which compared control room design features with CECo Human Factors Guidelines;B.A verification of instrumentation and control availability and the verification that operator task performance is not affected by the operator/control board

interface; andC.A validation of the control room functions to ensure the functions allocated to the control room operating crew can be accomplished within the structure of the

defined emergency operating procedures and the design of the control room as it

exists.

QUAD CITIES - UFSAR Revision 3, December 1995 7.5-107.5.5References1."Instrumentation for Light-Water Cooled Nuclear Power Plants to Assess Plant and Environs Conditions During and Following an Accident," NRC Regulatory Guide 1.97, Revision 2, December1980.2."Supplement 1 to NUREG-0737 "Requirements for Emergency Response Capability," Generic Letter 82-33. 3."Quad Cities Station Supplement 1 to the Detailed Control Room Design Review Find Summary Report Volumes 1 and 2," December 1985, Commonwealth Edison Company.

QUAD CITIES - UFSAR Revision 9, October 2007 7.6-1 7.6 CORE AND VESSEL INSTRUMENTATION

This section describes core and vessel instrumentation system. Included are nuclear

instrumentation systems and vessel instrumentation. Refueling interlocks are described in

Section 9.1.

7.6.1 Nuclear Instrumentation

7.6.1.1 Design Bases

The nuclear instrumentation is designed to:

[7.6-1]

A. Provide the operator with the information required for optimum, safe operation of the reactor core; and

B. Provide inputs to the reactor protection system (RPS) and the rod block circuitry to assure that the local power density, power oscillations and bulk power level do not exceed preset limits.

In order to meet the design requirements, the nuclear instrumentation must:

A. Detect, measure, and indicate neutron flux from the source range level through the power range level;

B. Annunciate an alarm on component failures; and

C. When reactor power is in the power range:

1. Indicate local neutron flux;
2. Compute and indicate average reactor power; and
3. Detect and suppress core power oscillations.

Specific design requirements are listed for each nuclear instrumentation subsystem.

7.6.1.2 General Description

The nuclear instrumentation uses three types of neutron monitors. The neutron flux level for

operation in the region of subcritical to an intermediate flux level at which the reactor is critical is

monitored by the source range monitor (SRM). The intermediate range monitor (IRM) is used

from a neutron flux of just above criticality to approximately 10% of full power (refer to Figure

7.6-1). From about 3% power to full power operation, the local power range monitor (LPRM) is

used. The detectors for the SRM and IRM subsystems are withdrawn from the core during power

range operation. The detectors for the power range are fixed in place. An in-depth report covering the incore neutron monitoring system is documented in topical report APED-5706, Revision 1 (April 1969).

QUAD CITIES - UFSAR Revision 12, October 2013 7.6-2 During operation in the power range, the LPRM signals are used by separate subsystems:

1. LPRM flux level is indicated, and a high flux alarm is annunciated if the level reaches a preselected point.
2. The average power range monitors (APRMs) average the outputs of selected LPRMs in such a manner that indication of average reactor power is provided.

The APRM generates scram signals on high-high APRM flux level.

3. During control rod motion, the average of a set of LPRMs adjacent to the selected control rod is used by the Rod Block Monitor (RBM) to limit increases in local

power.

4. The OPRMs utilize LPRM signals to detect and alarm core power instabilities that have the potential of occurring in the high power / low flow portion of the

operating domain. The OPRMs are designed to automatically suppress the

detected oscillations prior to exceeding the MCPR safety limit by providing a

Reactor Protection system (RPS) trip function.

Figure 7.6-2 presents a block diagram of the various nuclear instrumentation ranges as

they are functionally assembled.

A traversing incore probe (TIP) may be inserted in the core to obtain an axial neutron flux

distribution at each LPRM detector location. The information obtained from the TIP is used to calibrate the LPRM system and to provide a relative flux distribution for the core to

the process computer.

7.6.1.3 Source Range Monitoring Subsystem

7.6.1.3.1 Design Bases

In order to meet the general design requirement to provide the nuclear information needed

for knowledgeable and efficient reactor startup and low flux level operation, the SRM must:

A. Provide a minimum signal-to-noise ratio of 3:1 and a minimum count rate of 3 cps with all control rods inserted prior to initial power operation. (For the

original core, this included the contribution of neutron-emitting sources - see

7.6.1.3.2);

B. Show a measurable increase in output signal from at least one detector before the neutron flux multiplication exceeds a factor of 2000 during the most limiting

startup control rod withdrawal condition; and

C. Provide a signal overlap of at least one half ( 1/2) decade to the IRM signal with the SRM detectors in the fully-inserted position.

7.6.1.3.2 System Description

The SRM subsystem is used to provide the necessary information for reactor startup from

subcritical to an intermediate flux level and for refueling operations. The system consists of four miniature fission chambers which are operated in the pulse counting mode. These QUAD CITIES - UFSAR Revision 12, October 2013 7.6-3 detectors have a nominal sensitivity of 2 x 10

^-3^ cps/nv (nv is neutrons per square centimeter per second) and are located radially in the core as shown in Figure 7.6-3. The detectors are attached to drive mechanisms which can position the chambers from the fully-inserted location (approximately core center) to a position approximately 2 feet below the

reactor core.

[7.6-2]

The detector drive system consists of a detector drive, a flexible drive shaft, a motor

module, and a drive tube for each detector. The drive is mounted through an adapter to the

instrumentation nozzle well below the vessel in a location that does not interfere with the control rod operation and maintenance. The drive tube is a long hollow tube which acts as

a guide. A long, slender shuttle tube is mounted on the upper end of the drive tube. This

combination tube, housing the fission chamber detector assembly, is driven up and down

inside the dry tube.

[7.6-3]

A flexible drive shaft transmits power to the gearbox of the detector drive assembly from the motor module located approximately 20 feet away. Four limit switches provide detector

position information and also interlock the motor power circuits to establish insert and

retract limits.

Seven neutron-emitting antimony-beryllium sources were located radially within the

reactor core as indicated in Figure 7.6-3. These sources were designed to provide at least three cps in each SRM channel with the reactor in the cold, xenon-free, fully-shutdown

condition prior to initial power operation. This requirement continued to be met during

routine reactor operation by reactivation of the radioactive source (Sb-124) through capture

of reactor neutrons. These sources have been removed, since photoneutron production is

high enough to provide the required neutron flux without these sources.

The SRM detector assembly consists of a fission chamber attached to a low-loss quartz fiber

insulated transmission cable terminated with a connector. The detector cable is connected

below the reactor vessel to a triple-shielded cable which carries the detector electrical output to the monitor circuitry. The output of the four SRM detectors is amplified and the

signal is conditioned. The resulting signal, proportional to the logarithm of the counts per

second occurring in the detector is continuously displayed to the reactor operator on log count rate meters. The time derivative of this signal is formed and displayed to the reactor

operator on four reactor period meters which have an inverse scale and indicate the period in seconds. A recorder is available to the operator to allow recording of all four log count rate signals. Annunciators are activated under various conditions, for example, short

reactor period or high count rate.

Each of the four SRM channels initiates a rod block (see Section 7.7.) with the mode switch

in STARTUP/HOT STANDBY or REFUEL under the following conditions:

A. SRM detectors not fully inserted into the reactor core with the SRM count level below 163 cps (allowable value);

B. SRM count level high, greater than 2.8x 10 5 cps (allowable value); or

C. SRM channel inoperative.

The SRM detector position rod block is actuated by a position indicator on the retract mechanism. The SRM channel inoperative rod block is effective whenever the high voltage

supply drops below a preset level, one of the channel modules is not plugged in, or the

channel is not in its OPERATE mode. A rod block signal from any one of the four channels

prevents rod withdrawal.

QUAD CITIES - UFSAR Revision 6, October 2001 7.6-4 Any one of the four SRM channels may be bypassed by operation of a bypass switch on the control panel. An automatic bypass of the SRM channel detector position rod block occurs

when the count rate is greater than 100 cps.

Reactor startup is begun with the unbypassed SRM chambers fully inserted. Withdrawal of control rods increases the reactivity of the reactor core and hence, the multiplication of

source neutrons. Although the removal of a given individual control rod may not show as a measurable increase on all chambers, the approach to criticality through distributed control

rod withdrawal will be indicated by an appreciable increase in the count rate. Both the log

count rate meters and the period meters provide indication of the approach to criticality, criticality and, with further withdrawal of control rods, supercriticality. After sufficient rod

withdrawal to obtain a useful reactor period (on the order of 60 - 90 seconds) the reactor

power is allowed to increase exponentially.

The SRM chambers may be withdrawn from the fully-inserted position when the count rate

is greater than 100 cps on the chamber to be withdrawn. To continue the reactor startup, withdrawal of the SRM detectors must be gradual, maintaining the SRM count levels

between the low level (100 cps) and high level (10 5 cps) rod block set points. Each SRM chamber can be withdrawn individually, and it may be stopped at any intermediate point in

its travel. Withdrawn SRMs which are selected will be automatically inserted on a reactor

scram. [7.6-4]

The useful range of the SRM channels is from 10 10 6 cps, which corresponds to a flux range of 10 4 - 5 x 10 8 nv. [7.6-5]

7.6.1.3.3 Design Evaluation

The number and location of the SRM detectors and neutron-emitting sources have been

analytically and experimentally determined to be sufficient to result in a count rate of 3 cps with all rods inserted in the cold, xenon-free condition prior to initial power operation.

Verification of conformance to the minimum count rate was made at the time of fuel

loading. The sources are not necessary following extended power operation. The detector sensitivity and monitor electronic characteristics have been chosen to guarantee a

minimum signal to noise ratio of 3:1.

The primary safety function of the SRM system is to verify that an adequate neutron flux

background exists during an approach to criticality. The number of SRM channels was

selected to permit positive detection of an approach to criticality performed by withdrawing

control rods in the region most remote from chambers. In this worst case, the nearest

unbypassed SRM channel would show a factor of 1.1 signal increase at the time criticality is

achieved.

Since the SRM detectors can be retracted as reactor startup is continued, a large overlap of

indication is possible during transition from the SRM to the IRM. Figure 7.6-1 depicts the

overlap between the two monitoring subsystems. Even with the SRM detectors fully

inserted, an overlap of approximately one decade is provided. The SRM/IRM detector range overlap reduces the uncertainty in the neutron level indication during the transition from the SRM to the IRM. The Technical Specifications allow for the verification of SRM and IRM overlap prior to fully withdrawing the SRMs.

[7.6-5a]

The detector is designed to function in the environment in which it is to be located.

Any SRM component or power supply failure is annunciated. Failure of any SRM channel

during low flux operations with the mode switch in REFUEL or STARTUP/HOT QUAD CITIES - UFSAR 7.6-5 Revision 9, October 2007 STANDBY will initiate a rod block, thus preventing control rod withdrawal. The bypass switch arrangement permits only one SRM channel to be bypassed, guaranteeing the

required detection capability during source range reactor operation.

The SRM detector position rod block assures that reactivity insertion will not be made

under very low flux level conditions unless the SRM detectors are inserted to the optimum

position for flux detection. Administrative controls exist to ensure that at least two SRMs

are fully inserted and operable prior to control rod withdrawal for startup.

[7.6-6]

7.6.1.4 Intermediate Range Monitoring Subsystem

7.6.1.4.1 Design Basis

The intermediate range monitoring (IRM) subsystem is designed to:

[7.6-7]

A. Detect and indicate neutron flux level in a range between the SRM detection capability and the power range instrumentation capability (approximately 10 8 - 10 12 nv); and

B. Generate trip signals to prevent fuel damage from a single operator error or a single equipment malfunction.

7.6.1.4.2 System Description

The IRM subsystem is composed of eight miniature fission chambers located radially in the

core as shown in Figure 7.6-4. The figure also shows the assignment of IRM detectors to

each RPS logic channel. The assignment is made to provide coverage of each quadrant of

the reactor core with one detector in each channel bypassed. The detectors are attached to drive mechanisms which can position them from the fully-inserted location (approximately

core center) to a position approximately 2 feet below the reactor core. The drive systems

are identical to those used in the SRM subsystem and the detectors are similar, except for

the range of measurement. The detectors are not withdrawn from their fully inserted

position until the mode switch has been turned to the RUN position. Withdrawn

previously-selected IRMs will be inserted automatically on a scram.

[7.6-8]

The output of each fission chamber is processed through a wide-band amplifier to a voltage

variance circuit (Campbelling or root mean square technique)

[1] and a signal conditioner to produce an output which is linearly proportional to the reaction rate in the chamber. This

output is provided to a trip unit and is used to drive one channel in one of four recorders.

[7.6-9]

The IRM subsystem can detect flux levels from the upper end of the SRM range to

approximately 1.5 x 10 13 nv (34% of full power).

A neutron flux of 5 x 10 7 nv (upper source range) will provide a signal of approximately 0.1 full scale on the lowest IRM range.

In order to handle the wide range of IRM detection, the IRM equipment is provided with a remote range switch which selects various ranges of attenuation of the detector signal. As QUAD CITIES - UFSAR 7.6-6 the neutron flux level changes during reactor startup, the operator manually up-ranges the IRM.

The IRM subsystem provides trip signals for both the RPS and the rod block circuitry; all

the trips but one, as described in the following, are effective only with the mode selector

switch in the REFUEL or STARTUP/HOT STANDBY positions.

Each IRM detector provides a trip signal to the RPS scram logic circuitry under the

following conditions:

A. IRM high-high flux level,

B. IRM channel inoperative, and

C. IRM channel high flux level or inoperative with its companion APRM downscale in the RUN mode.

In order for a scram to occur, a scram trip signal must be received in both RPS logic

channels. The scram-initiating high-high level trips provide automatic shutdown capability

for operation from just critical to the lower portion of the power range.

When the reactor mode switch is in REFUEL or STARTUP/HOT STANDBY, the IRM

subsystem provides a rod block signal to the rod block circuitry under the following

conditions:

A. IRM high flux level,

B. IRM inoperative,

C. IRM downscale on any range but the lowest, and

D. IRM detectors not fully inserted into the core.

Any one of the eight IRM channels can initiate a rod block.

Any one IRM detector channel in each RPS logic channel may be manually bypassed, making ineffective the scram and rod block associated with that individual IRM channel.

7.6.1.4.3 Design Evaluation

The number and location of the IRM detectors have been analytically and experimentally

determined to provide sufficient intermediate range flux level information under the worst

permitted bypass and chamber failure conditions. Figure 7.6-1 shows the range capability

of the IRM channels. The ability of the monitor output to provide an accurate

measurement of the detector reaction rate over the flux range of interest has been verified

by experimentation with the root mean square technique

[1]. Intermediate range monitor channel redundancy includes a margin which allows for component failure, and also allows

continued reactor operation with one IRM bypassed in each RPS logic channel. The scaling

arrangement in the IRM subsystem assures that for all unbypassed IRM channels, the

scram and rod block trips are no more than a factor of 10 above the IRM level at that time.

This assures that, should scram or rod block action be needed due to rapid or unintentional

neutron flux increases, the trip signal will be generated before the flux QUAD CITIES - UFSAR 7.6-7 increases by a factor greater than 10, thus providing a conservative margin to fuel damage.

A range of rod withdrawal accidents has been analyzed. The most severe case involves all initial conditions in which the reactor is just subcritical and the IRM subsystem is not yet

onscale. This condition exists at the three-quarter rod density illustrated in Figure 7.6-5 (rod density is the total notches inserted in the core divided by the number of notches which

would be inserted when all rods are fully inserted). Full withdrawal of the control rod indicated will result in the power distribution indicated in Figure 7.6-6; it should be noted

that this is an out of sequence rod which would normally be blocked by the rod worth

minimizer (see Section 7.7). Figure 7.6-5 indicates the location of rod withdrawn and the distance to the IRM chambers in the two RPS logic channels which will initiate a scram

with the IRM channels nearest to the withdrawn rod bypassed.

[7.6-10]

Comparison of the power distribution shown in Figure 7.6-6 indicates that the ratio of the

resultant neutron flux at the farthest detector to the neutron flux peak is 2.2 x 10

-4. Because the trip of the IRM channel associated with this detector is set to operate at a flux

of less than 6 x 10 8 nv (rod blocked if not set on proper range) the flux in the power peak is less than 2.7 x 10 12 nv. At this flux level, the power at the peak is limited to 7.7% of rated average power; hence, it will be within thermal limits, even if the recirculation pumps are

shut down.

[7.6-11]

The overlap between the IRM and the power range monitoring subsystem is sufficient to

guarantee a safe transition between the instrumentation ranges (Figure 7.6-1). Overlap

between the SRM and IRM ranges is discussed in Section 7.6.1.3.

The IRM detector position rod block is effective during periods of reactor operation when

the IRM is required for flux level indication.

The IRM detectors are chosen with characteristics which permit reliable performance in the

reactor environment.

IRM failures are annunciated, and during low flux level reactor operation, result in a RPS

single logic channel trip and rod block. Thus, further insertion of reactivity is prevented, and a reactor scram would be initiated by any condition resulting in a trip of the other RPS

logic channel.

7.6.1.5 Power Range Monitoring Subsystem

7.6.1.5.1 Local Power Range Monitoring Subsystem

7.6.1.5.1.1 Design Basis

In order for the power range monitoring subsystem to meet the general design

requirements for power range flux monitoring and to prevent excessive local and bulk

power densities, the local power range monitoring (LPRM) subsystem must:

A. Continuously monitor over its design range the local neutron flux, and alarm on excessive conditions; QUAD CITIES - UFSAR Revision 7, January 2003 7.6-8 B. Permit evaluation of the critical core parameters (fuel thermal limits) to an accuracy consistent with core design and established limits; and

C. Permit demonstration of compliance with the critical core parameters (critical power ratio) with a speed and ease consistent with efficient operation of the

plant.

7.6.1.5.1.2 System Description

The LPRM subsystem output signals are used to demonstrate that the core is operating

within the established limits on peak power density and minimum critical power ratio (MCPR). This system provides the information needed for evaluating the detailed characteristics of the power distribution or for other technical evaluations. The LPRM

subsystem provides input to the average power range monitoring (APRM) subsystem, Oscillation Power Range Monitoring (OPRM) subsystem and rod block monitor (RBM) subsystem which are described below.

[7.6-12]

The LPRM subsystem, which uses dc measurement techniques, consists of miniature fission

chambers located within the reactor core, electronic signal conditioning equipment located

in the control room, and a TIP calibration system.

Each LPRM has a high neutron flux level alarm and a common annunciator located on the

control board.

Figures 7.6-7 and 7.6-8 indicate the core location of the LPRM strings. Each LPRM string

consists of four miniature fission chambers which are spaced vertically at 3-foot intervals.

The top and bottom chambers are located 1.5 feet from the core boundaries, thereby

providing uniform core coverage in the axial direction. Also included in each detector string

is a calibration tube which accepts the TIP used to measure the axial flux distribution and

calibrate the LPRM subsystem (see Figure 7.6-8).

Figure 7.6-9 illustrates that, due to the equivalence of locations resulting from symmetry, the LPRM subsystem monitors all unique locations within the central region of the core

when the core is operated with quadrant symmetric control rod patterns.

The LPRM flux amplifiers are calibrated using data from the TIP calibration system, heat

balance data and some analytical data. The basic process involves:

A. Running the TIP system and accumulating axial profile data;

B. Normalizing the axial profile data;

C. Determining for each detector elevation the average nodal heat flux in four adjacent fuel nodes at detector elevations; and

D. Adjusting flux amplifiers until meter readings are proportional to heat flux.

These calculations are performed using the process computer (see Section 7.5.2). When calibrated, the LPRM signals are proportional to the average nodal power in the four

adjacent fuel nodes at the detector elevation. The LPRM amplifier signals adjacent to a

control rod selected are displayed to the reactor operator on 16 centrally-located meters on

the 901(2)-5 panel. This directs the attention of the operator to the local power level prior QUAD CITIES - UFSAR 7.6-9 to and during rod motion. These signals are also used by the RBM. When rods near the core periphery are selected, two or three detector strings may be used. When rods on the

core periphery are selected, the RBM system is bypassed. In both previous cases, the

readings are zeroed on the corresponding unused meters. The operator may view any desired region of the core by selecting of the control rod in the area of interest. A selected

set of LPRM signals is used as an input to each of the six APRM channels.

[7.6-12a]

7.6.1.5.1.3 Design Evaluation

The number and location of LPRM detectors provides the capability of determining local

heat flux in all unique locations in the central region of the core. Although each unique

location in each core quadrant is not specifically monitored, the quadrant symmetry (illustrated in Figure 7.6-9) effectively provides knowledge of the flux level throughout the

core. [7.6-13]

The previously described method of calibration using the TIP provides a method of

correlating LPRM measurements with local thermal conditions; thus, the LPRM

measurements are a valid representation of local thermal conditions.

Each individual LPRM channel will annunciate an alarm upon detection of a flux level

exceeding a preset limit. Thus the operator receives warning of local high or low flux

conditions or LPRM component failure.

The LPRM detectors are selected with characteristics which guarantee reliable operation in

the reactor environment; reactor temperature, pressure, neutron and gamma flux, and

detector electrical requirements were considered in detector selection.

The use of the LPRM signals in the RBM provides a positive assurance that local thermal

peaks which could cause fuel damage will be prevented.

7.6.1.5.2 Average Power Range Monitoring Subsystem

7.6.1.5.2.1 Design Basis

The APRM subsystem must continuously indicate core average flux level and initiate trips

to prevent excessive average power density. In order to fulfill its design requirement, the

APRM subsystem must:

A. Initiate trip signals which scram the reactor automatically before the neutron flux level exceeds specified values;

B. Initiate a rod block trip signal, thereby preventing core average power increases to excessive levels with reduced recirculation flow (the rod block trip setpoint will

be lower than the scram setpoint);

C. Provide a continuous indication and record of the bulk thermal power of the reactor in the power range; QUAD CITIES - UFSAR 7.6-10 D. During the worst permitted bypass and chamber failure conditions, generate a scram signal during neutron flux level transients before fuel damage has

occurred; and E. Continue to perform its function following any single component failure within the subsystem. In order that the APRM satisfy this requirement, there must be

two operable APRMs in each RPS logic channel. In a practical sense, this

requirement results in three APRM channels for each bus to permit bypassing

for calibration and maintenance during operation.

7.6.1.5.2.2 System Description

The APRM subsystem consists of electronic equipment which averages the output signals

from selected groups of LPRM flux amplifiers. Figures 7.6-10 and 7.6-11 illustrate the APRM subsystem for the reactor. As shown on these figures, the system consists of six

channels. Each of these channels averages the output signals from either 20 or 21 LPRM

flux amplifiers.

Three of the APRM channels provide trip inputs to one RPS logic channel, and the other

three APRM channels feed the other logic channel (see Section 7.2).

Each APRM channel provides a scram trip signal to RPS and a rod block trip under the

following conditions:

[7.6-14]

A. High neutron flux (flow referenced and fixed level) (rod block only),

B. High-high neutron flux (flow referenced and fixed level) (scram only),

C. APRM channel inoperative,

D. APRM channel reading downscale with the mode switch in RUN. (rod block only). (Refer to 7.2 and 7.7 for a further description of mode switch interlocks).

In order for a scram to occur, a scram trip signal must be received by both RPS logic

channels. Any one of the six APRM channels can initiate a rod block.

Switches located on the main control panel reactor console allow the operator to bypass the

trips from one of the APRM channels in each of the RPS logic channels; the bypass is

effective for both the scram and rod block trip signals.

The rod block set point is automatically varied with recirculation flow (with mode switch in

RUN) as shown in Figure 7.6-12. The slope of the trip vs. flow relationship is determined

by the characteristic bulk power vs. flow relationship of the reactor which was determined

experimentally. The absolute magnitude of the trip set point was established to prevent

operation significantly above the flow control characteristic that includes the point 100%

flow and 100% power.

The APRM channel output signals are continuously displayed on recorders located on the control board. The output signals are adjusted so that the meter deflections indicate

percent of rated bulk thermal power. Bulk thermal power is determined using heat balance

techniques. Adjustment of the APRM channel readings is not possible from the QUAD CITIES - UFSAR Revision 6, October 2001 7.6-11 control board and does not affect the output signals of the LPRM amplifiers which are averaged in the APRM channel.

If an LPRM used to provide input to an APRM channel fails, the operator can manually

bypass this invalid input. The APRM channel then properly averages the inputs from the

remaining LPRM channels. If the number of bypassed LPRMs used as inputs to an APRM channel exceeds a preset number, the APRM instrument inoperative alarm is actuated.

This feature assures that the APRM system will adequately perform its safety function of

terminating average neutron flux level transients through scram initiation. In addition to

the automatic input monitoring, administrative controls require at least 50% of all LPRMs

and at least 2 LPRMs per level for an APRM to be operable. The "too-few" input alarm

feature also automatically provides a high degree of assurance that the APRM system will

be capable of preventing fuel damage due to rod withdrawal errors.

The readout equipment for the APRM system is located in the control room. The APRM

outputs are displayed on continuous recorders shared with the IRM channels. Also located on the control board are the bypass switches described previously. Outputs from the

reactor recirculation flow sensors are used to provide the reference flow information.

Amplifiers are used to average the signals from the LPRM detectors in each of the six

APRM channels. Other equipment is used to automatically vary the upscale rod block and scram trip points with recirculation pump drive flow (which is indicative of bulk core flow)

as necessary to meet the design criteria. This equipment is located in the control room.

The flow-dependent bias which determines trip level is subject to both positive and negative

errors originating in the flow monitoring equipment. However, the equipment limits the

trip bias so that the trip level can never exceed the intended level for 100% flow regardless

of the magnitude of positive errors in flow signal. Negative errors are in the conservative

direction.

7.6.1.5.2.3 Design Evaluation

As shown in Figures 7.6-10 and 7.6-11, the LPRM inputs to the APRM channels provide a

wide sampling of local flux levels on which to base an average power level measurement.

The fact that three APRM channels are provided for each RPS logic channel assures that at

least two independent average power measurements will be available under the worst

permitted bypass or failure conditions. The six APRM channels provide continuous

indications of core average power level based on different samplings of local flux levels. The

APRM provides valid average power measurements during typical rod or flow induced

power level maneuvering as shown by Figures 7.6-13 and 7.6-14, which are the results of

analysis.

Using a plant heat balance technique, the APRM measurements are calculated such that

indications are within + 2% of the thermal power when the power level is greater than or equal to 25 percent of rated; this calibration is maintained by procedure.

The effectiveness of the APRM high flux scram signals in preventing fuel damage following

a single component failure or a single operational error is evaluated in each section of this

report where system failures are analyzed. In all such failures, no fuel damage occurs.

Since only two APRM channels in each RPS logic channel are required for effective

detection of bulk power level transients, the same effectiveness is attained even under the

worst permitted bypass conditions.

QUAD CITIES - UFSAR Revision 14, October 2017 7.6-12 The APRM rod block setpoint is set lower than the scram setpoint; thus, reactivity additions due to rod withdrawal errors are terminated well before fuel damage limits are

approached.

The APRM component failures which result in upscale, downscale, or instrument inoperative conditions are annunciated. The reduction of LPRM inputs for any APRM

channel below a preset number gives an alarm, rod block, and a logic channel trip. These

features warn of loss of APRM capability.

7.6.1.5.3 Rod Block Monitor

7.6.1.5.3.1 Design Basis

The RBM is designed to initiate a rod block under the worst permitted bypass and chamber

failure conditions to prevent local fuel damage during the worst single rod withdrawal error

starting from any permitted power and flow condition.

7.6.1.5.3.2 System Description

The system uses the signals from the LPRM strings adjacent to the selected control rod (Figure 7.6-15) and the recirculation flow sensors. The signals from the A and C levels are

averaged in one channel and the signals from the B and D levels are used in the second channel. The RBM output is automatically adjusted upon rod selection so that its output is

equal to the reading of a preselected APRM channel. This gain setting is held until a new

control rod is selected. An in-depth description of the RBM system is given in topical report

APED 5706, "In-Core Neutron Monitoring System for General Electric Boiling Water

Reactors," Revision 1, April 1969.

Two RBM channels are provided; either channel, independently, will prevent rod

withdrawal under the following conditions:

A. High neutron flux (flow referenced);

B. One of the two channels inoperative; and

C. Channel reading downscale with the mode switch in RUN.

One of the two RBM channels may be manually bypassed.

The RBM high trip setpoint varies linearly with recirculation flow as does the APRM rod block setpoint. However, the 100% flow intercept depends on the power-flow characteristic along with which reactor is operating. For the exact setpoints see the current core operating limits report. The RBM is bypassed below 30%.

[7.6-15]

QUAD CITIES - UFSAR Revision 14, October 2017 7.6-13 7.6.1.5.3.3 Design Evaluation Since the RBM utilizes the signals for the LPRMs, it is capable of determining the approach

of local thermal flux conditions which could result in local fuel damage. The fact that either

RBM channel can, independently, initiate a rod block, provides assurance that a rod

withdrawal error will be terminated even with one RBM channel bypassed.

[7.6-16]

The effectiveness of the RBM to prevent local fuel damage as a result of a single rod

withdrawal error has been analytically determined on a fuel cycle specific basis. Results from cycle specific analyses determine the appropriate RBM setpoint needed to assure the

design basis function. Depending on the cycle specific analysis results, rod withdrawal

error events may achieve acceptable results with no control rod blocking by the RBM. For

these specific cycles, the RBM setpoint as described in the core operating limits report, is

raised such that an APRM rod block will occur prior to the high trip RBM rod block. The

initial condition is conservatively defined such that the reactor is operating at maximum

permitted power with MCPR and peak power density at the steady-state limits in a region

adjacent to a fully-inserted control rod; no credit is taken for the action of the rod worth

minimizer (see Section 7.7). The response of the least responsive RBM channel is

calculated as a function of rod withdrawal distance. The MCPR and peak power density are

also calculated as a function of rod position.

[7.6-17]

7.6.1.5.4 Traversing Incore Probe

The TIP system includes five TIP machines, each of which has the following components:

[7.6-18]

A. One traversing incore probe,

B. One cable drive mechanism,

C. One 10-position indexing mechanism, and

D. Nine guide tubes (one to a common core location).

The system allows calibration of LPRM signals by correlating TIP signals to LPRM signals

as the TIP is positioned in various radial and axial locations in the core. TIP machine availability requirements can be found in the current cycle's COLR. The guide tubes inside the reactor are divided into groups. Each group has its own associated TIP machine.

A TIP machine uses a fission chamber attached to a flexible drive cable, which is driven

from its lead shielded storage chamber outside the primary containment by a pinion gear

box assembly. The flexible cable is contained by guide tubes that continue into the reactor core. The guide tubes are specially prepared to provide a durable, low-friction surface and

are a part of the LPRM detector assembly. The indexing mechanism allows the use of a

single detector in any one of the nine different tube paths. A tenth tube is available as a

spare.

The Unit 1 control system includes five Automated TIP Control Units (ATCUs) that provide

both manual and automatic operation. The TIP signals are amplified and displayed on the ATCU screens. The ATCUs provide the TIP scan data to the process computer. A single ATCU can be set as a master ATCU to initiate a full TIP set scan.

QUAD CITIES - UFSAR Revision 10, October 2009 7.6-14 The Unit 2 control system includes five Drive Control Units (DCUs) that provide both manual and semiautomatic operation. The TIP signals are amplified and displayed on a meter and input via the DCUs to the process computer. Core position versus neutron flux is recorded on an x-y plotter.

For Unit 1, the cable drive mechanism contains the drive motor, the cable takeup reel, and a position encoder to provide position indication to the ATCU for positioning the TIP at specific locations along the guide tube. For Unit 2, the cable drive mechanism contains the drive motor, the cable takeup reel, and analog probe position indicator for the recorder, and

a counter to provide digital pulses to the control unit for positioning the TIP at specific

locations along the guide tube.

The cable drive mechanism inserts and withdraws the TIP and its cable from the reactor

and provides detector position indication signals. The drive mechanism consists of a motor and drive gear box which drives the cable in the manner of a rack and pinion. A two-speed

motor provides a high speed for insertion and withdrawal and a low speed for scanning the

reactor core.

For Unit 1, the encoder is driven directly from the output shaft of the cable drive motor.

The encoder and a flux amplifier output are used to plot neutron flux versus incore position of the TIP. The ATCU utilizes the position encoder data to position the TIP in the guide tube with a linear position accuracy of plus-or-minus one inch. The ATCU can control TIP positions at the top of the core, for initiation of scan, and at the bottom of the core, for

changing to fast withdrawal speed.

For Unit 2, the analog position indicator and the counter (digital) are also driven directly

from the output shaft of the cable drive motor. The analog position signal from a

potentiometer and a flux amplifier output are used to plot neutron flux versus incore position of the TIP. The DCU control logic utilizes the digital counter output to position the TIP in the guide tube with a linear position accuracy of plus-or-minus one inch. The DCU can control TIP positions at the top of the core, for initiation of scan, and at the bottom of

the core, for changing to fast withdrawal speed.

A position limit switch provides an electrical interlock release when the probe is withdrawn

clear of the indexing mechanism to allow the TIP to be indexed to the next guide tube

location. The limit switch is actuated when the end of the TIP passes a switch in the guide tube in use. The cable drive motor includes an ac voltage-operated brake to prevent

coasting of the TIP after a desired incore position is reached.

Each 10-position indexing mechanism functions as a circular transfer machine with nine

usable indexing points. Eight of these locations are for the guide tubes associated with that

particular TIP machine. The final location is for the guide tube common to all the TIP

machines. Indexing to a particular tube location is accomplished manually at the control

panel by means of a position selector switch which energizes the electrically-actuated

rotating mechanism. The tube transfer mechanism is part of the indexing mechanism and

consists of a fixed circular plate containing 10 holes on the reactor side of the primary

containment which mates to a rotating single-hole plate. The rotating plate aligns and

mechanically locks with each fixed hole position in succession. The indexing mechanism is

actuated by a motor-operated rotating drive. Electrical interlocks prevent the indexing

mechanism from changing positions until the probe cable has been completely retracted

beyond the transfer point. Additional electrical interlocks prevent the cable drive motor

from moving the cable until the transfer mechanism has indexed to the preselected guide

tube location.

QUAD CITIES - UFSAR Revision 10, October 2009 7.6-15 A valve system is provided with a valve on each guide tube entering the primary containment. These valves are closed except when the TIP system is in operation. A ball valve, manual valve and a cable-shearing valve are mounted in the guide tubing just outside of the primary containment. A valve is also provided for gas purge line to the

indexing mechanisms. A guide tube ball valve opens only when the TIP is being inserted.

The shear valve is used only if a containment isolation occurs when the TIP is beyond the

ball valve and cannot be withdrawn. The shear valve, which is controlled by a manually-operated keylock switch, can cut the cable and close off the guide tube. The shear valves

are actuated by detonation squibs. The continuity of the squib circuit is monitored by

indicator lights in the control room. An additional manual ball valve is installed between

the automatic ball valve and the drywell penetration.

The guide tube ball valve is normally de-energized and in the closed position. When the TIP starts forward the valve is energized and opens. As it opens it actuates a set of

contacts which gives a signal light indication at the TIP control panel and bypasses an

inhibit limit which automatically stops TIP motion if the ball valve does not open on

command. A Group II containment isolation signal initiates TIP drive withdrawal. Once the probe is retracted to the IN SHIELD position, then the ball valve will close. Ball valve position is displayed in the control room and loss of power to the shear valve circuitry and

the actuation of any shear valve are both annunciated.

[7.6-19] The entire TIP system including its controls is not safety-related, except for the tubing and valves on the outside of each primary containment penetration, which are mechanically safety-related through the outermost valve. The TIP tubing does not directly communicate with the reactor vessel or the containment air space. Thus the TIP system response to a PCIS Group 2 initiation does not require a safety system design. Refer to Section 6.2.4.5 for a detailed discussion of the TIP system response to a containment isolation.

7.6.1.5.5 Oscillation Power Range Monitoring (OPRM) Subsystem The Oscillation Power Range Monitoring (OPRM) subsystem is a microprocessor-based

monitoring and protection system, which will:

  • detect a thermal-hydraulic instability,
  • provide an alarm on detection of an oscillation (based on period based algorithm only), and
  • initiate an Automatic Suppression Function (ASF) trip to suppress an oscillation prior to exceeding fuel safety limits.

The subsystem design, technical details, equipment qualification, and validation are

discussed in Reference 4. The NRC has accepted the above reference and has also issued a

safety evaluation report (Reference 5).

7.6.1.5.5.1 Design Basis 7.6.1.5.5.1.1 Safety Design Bases Boiling water reactor cores may exhibit thermal-hydraulic instabilities in certain portions

of the core power and flow operating domain. General Design Criterion 10 (GDC 10)

requires that the reactor core be designed with appropriate margin to assure that

acceptable fuel design limits will not be exceeded during any condition of normal operation

including the effects of anticipated operational occurrences. GDC 12 requires assurance

that power oscillations which can result in conditions exceeding specified acceptable fuel QUAD CITIES - UFSAR Revision 10, October 2009 7.6-15a design limits are either not possible or can be reliably and readily detected and suppressed.

The OPRM is provided to meet the requirements of these GDCs by adding a detect and

suppress feature to the Reactor Protection System.

7.6.1.5.5.1.2 Power Generation Design Bases

The power generation design basis of OPRM consists of assuring that spurious scrams do

not occur. This objective is accomplished in part by establishing an exclusion region, as

discussed below in Section 7.6.1.5.5.2, where the thermal-hydraulic oscillations are not

postulated to occur.

7.6.1.5.5.2 System Description

Detailed description of OPRM subsystem design and physical arrangements are provided in

the Generic Topical Report (Reference 4). Basic and station specific information is

summarized here.

The OPRM subsystem consists of 4 OPRM trip channels, each channel consisting of two OPRM modules. Each OPRM module receives input from a group of LPRMs combined into

localized monitoring cells. It also receives input from the Average Power Range Monitor (APRM) power and Reactor Recirculation flow signals to automatically enable the trip

function of the OPRM module. A block diagram showing the relationship of OPRM with

other nuclear instrumentation is shown in Figure 7.6-2. A block diagram showing the

OPRM subsystem interconnections is shown in Figure 7.6-17.

The OPRMs are capable of detecting thermal-hydraulic instabilities within the reactor core.

The OPRMs are designed to provide an alarm and initiate an automatic suppression

function (ASF) trip to suppress oscillations prior to exceeding the MCPR safety limit. The OPRMs are auto enabled at the specified reactor recirculation flow and reactor power

setpoints. The ASF outputs initiate an ASF trip through the RPS based on the existing

plant trip logic and configuration. The OPRM System provides annunciator windows, SER

messages and indicating lights for pre-trip conditions and other alarm functions such as

Trip, Alarm, Trouble, Inop, Bypass and Trip Enabled to be displayed in the Main Control

Room (MCR).

Each OPRM subassembly includes a signal processing module, Automatic Suppression Function (ASF) Trip Relay Assembly, OPRM Annunciator Relay Assembly, two Digital

Isolation Blocks (DIBs) and Enable and Bypass Selector Switches.

The OPRM trip circuits may be bypassed by a selector switch. The bypass is accomplished through hardwired bypass of ASF trip relay contact by a selector switch actuated auxiliary

relay contact and through actuation of OPRM logic circuits and software. The bypass

condition of the OPRM module is indicated by the sequence of events monitor and by

indicating lights. The OPRMs may be manually enabled by a selector switch for any

recirculation flow and reactor power levels.

A. Modes of Operation

The OPRM has two modes of operation, operate and test. In the operate mode, it

performs all of its normal trip and alarm functions as well as broadcasting status

information to fiber optic output ports. The test mode is utilized for test, calibration, setpoint adjustment and downloading of the event buffer. In the test mode, the

OPRM's trip output is bypassed and the OPRM module is considered inoperable.

QUAD CITIES - UFSAR Revision 10, October 2009 7.6-15b Entry into the test mode is controlled by a key switch and is annunciated in the control room.

B. Event Buffer

When a trip occurs, data immediately prior to and following the trip is captured in

an event buffer. This buffer may be downloaded to aid in the analysis of the trip.

The event buffer can also be captured and downloaded at any time for non-trip

analysis by placing the OPRM in the test mode.

C. Maintenance Terminal

A portable maintenance terminal is utilized for system testing, calibration, and data

collection. It is connected to the OPRM via fiber optic cables. This maintains

isolation between the safety related OPRM and the non-safety related maintenance

terminal.

With the OPRM in its operate mode, the maintenance terminal may only be used to

collect data, which is broadcast by the OPRM at fixed intervals. Communications in

this mode are one way, namely OPRM to maintenance terminal, via the fiber optic

connections. The OPRM will not respond to commands from the maintenance

terminal when in the operate mode. Thus, the maintenance terminal cannot affect

OPRM operation.

In the OPRM test mode, bi-directional, fiber optic communications are established

between the OPRM and its maintenance terminal. In this mode, commands may be

seen from the maintenance terminal to the OPRM to perform such actions as altering the OPRM configuration and setpoints, downloading event buffers and error

logs, and testing various OPRM functions. Additional conventional test cables may

be connected between the maintenance terminal and a test port on the OPRM to

provide simulated analog signals for use in calibration and testing. To access this

test port, a shorting plug must be removed from the OPRM. Removal of the shorting

plug causes the OPRM module to become inoperable and is annunciated in the

control room.

D. Power Supply

Power supplies for the OPRMs are the same as those for the APRM and LPRM

Group channels. These power supplies provide the required voltage sources for

OPRM signal processing modules, DIBs, ASF Trip Relay Assemblies, OPRM

Annunciator Relay Assemblies, the new flow units, analog isolators and the existing

APRM, RBM and LPRM channels.

E. Physical Arrangement The OPRM signal processing modules are installed in APRM and LPRM Pages of the Power Range Neutron Monitoring System (PRNMS) Panel (see Figure 7.6-17).

Selector switches required for the manual enable functions are installed in the

PRNMS panel. Bypass selector switches are installed in the 901(2)-5 panel.

Indicating lights for the enable and bypass functions are installed in the 901(2)-5

panel. Automatic Suppression Function (ASF) Trip Relay Assemblies, OPRM

Annunciator Relay Assemblies, Analog Isolators, Digital Isolation Blocks, and

manual enable switches are installed in the PRNMS Panel.

QUAD CITIES - UFSAR Revision 9, October 2007 7.6-15c F. Exclusion Region The OPRM is required to be operable in order to detect and suppress neutron flux oscillations in the event of thermal-hydraulic instability. As described in Reference 4, the region of anticipated oscillation is defined by reactor thermal power (RTP) 30% and core flow <60% of rated core flow. The station specific region of anticipated oscillation is defined by RTP 25% and core flow <60% of rated core flow to reflect changes in rated output following extended power uprate (EPU) implementation. It is not necessary for the OPRM to be operable with reactor thermal power <25%.

G. Algorithm Reference 4 describes three separate algorithms for detecting stability related oscillations: the period based detection algorithm, the amplitude based algorithm, and the growth rate algorithm. The OPRM System hardware implements these algorithms in microprocessor-based modules. These modules execute the algorithms based on LPRM inputs and generate alarms and trips based on these calculations.

These trips result in tripping the Reactor Protection System (RPS) when the appropriate RPS trip logic is satisfied. Only the period based detection algorithm is used in the safety analysis. The remaining algorithms provide defense in depth and additional protection against unanticipated oscillations.

H. Trip Function The OPRMs are designed to provide an alarm (based on period-based algorithm only) and initiate an automatic suppression function (ASF) trip to suppress oscillations prior to exceeding the MCPR safety limit. The OPRMs are auto enabled at the specified reactor recirculation flow and reactor power setpoints. The OPRM initiates an ASF trip through the RPS based on the existing plant trip logic and configuration. The OPRMs provide alarm for pre-trip conditions and other alarm functions such as Trouble, Inop, and Trip Enabled to be displayed in the main control room. Table 7.6-1 lists the OPRM trip functions and setpoints.

I. Alternate Backup Method At times when OPRM channels may be inoperable, and until they can be restored to operable status, an alternate method of detecting and suppressing thermal hydraulic instability oscillations can be used. This alternate method is described in Reference

6. It consists of increased operator awareness and monitoring for neutron flux oscillations when operating in the region where oscillations are possible. If indications of oscillation, as described in Reference 6, are observed by the operator, the operator will take the actions described by procedures, which include initiating a manual scram of the reactor.

J. Component Qualification Considerations The OPRM devices are designated Class 1E, Seismic Category I and are qualified to the applicable portions of IEEE-381 and IEEE-344.

QUAD CITIES - UFSAR Revision 9, October 2007 7.6-15d K. Single Failure Considerations Since the OPRMs perform a protective function, they are required to withstand a single failure. To ensure acceptable defense against single random failures, the combination of architecture, wiring practices and use of isolation devices is applied to provide required redundancy, isolation and physical independence.

There is an OPRM channel associated with each of the four RPS trip system divisions. OPRMs in each RPS division are electrically isolated and physically separated from OPRMs in other RPS divisions. Within each OPRM channel there are two OPRM modules. The use of two OPRM modules per channel provides redundancy against an OPRM hardware failure in the same channel. The redundant OPRM modules in the same RPS division share the same Class 1E power supplies as those used by the safety-related APRM modules in that RPS division. However, each OPRM module is electrically isolated from the companion module in the same channel.

Common software failures do not lend themselves well to single failure analyses.

System reliability and safety requirements are examined in the description of the software design process and quality assurance considerations as discussed in Reference 4.

L. Redundancy, Diversity, and Separation Since the OPRM's operation is based on interface with PRNMS and RPS, the redundancy, diversity and separation requirements are the same as the requirements for these systems. The LPRM analog signals, which are locally wired, are provided to OPRMs with the same redundancy and separation as provided to the APRMs and LPRM groups. However, unlike the APRM logic where the output of APRMs 3 and 4 is shared between two different RPS divisions, there is a sufficient number of OPRMs such that the outputs of two OPRMs are assigned to the trip logic of a single RPS division. This configuration provides the required redundancy and maintains channel separation requirements. The assignment of OPRMs and existing APRMs for each RPS division is as follows:

RPS Division OPRM APRM A1 1,3 1,3 A2 2,7 2,3 B1 5,8 4,5 B2 4,6 4,6 7.6.1.5.5.3 Design Evaluation The OPRM subsystem is designed to alarm when a stability-related thermal-hydraulic oscillation is detected (based on period-based algorithm only), and to initiate an ASF trip when the oscillations are large enough to threaten fuel safety limits. The system settings assure adequate trip sensitivity while providing adequate margin to avoid inadvertent trips and spurious alarms. The OPRM system functions meet the requirements of GDC 12, and hence, acceptably address the related requirements of GDC 10 for ensuring reactor QUAD CITIES - UFSAR Revision 14, October 2017 7.6-15e safety in the event of power instabilities. The OPRM software development methodology is consistent with the guidance provided in Regulatory Guide 1.152, which endorses IEEE Std

7-4.3.2-1993 for ensuring software quality. The OPRM design assures high reliability as it

is governed by Quality Assurance requirements and applicable industry standards. The

system performs self-health tests on a continuous basis.

Reference 6 describes the licensing basis and methodology that demonstrates the adequacy of the hardware and software to meet the functional requirements. The requirements of Reference 6 were later supplemented with the need to perform cycle-specific DIVOM calculations. For Quad Cities Unit 1 AREVA reload cores this is accomplished with the RAMONA5-FA methodology of Reference 7. For Quad Cities Unit 2 Westinghouse reload cores this is accomplished with the methodology from References 8 & 9. The application of the process for determining OPRM setpoints is summarized in detail in the reload safety analysis report.

7.6.2 Reactor Vessel Instrumentation

The following section describes instrumentation associated with the reactor pressure vessel.

This includes those instruments which measure vessel water level, reactor pressure, vessel

metal temperature, and head flange leakage.

7.6.2.1 Design Bases and Design Features

A. Design Bases

The reactor vessel instrumentation is designed to fulfill a number of requirements

pertaining to the vessel itself or the reactor core. The instrumentation must:

[7.6-20a]

1. Provide the operator with sufficient information in the control room to protect the vessel from undue stresses;
2. Provide information which can be used to assure that the reactor core remains covered with water and that the separators are not flooded;
3. Provide redundant, reliable inputs to the reactor protection system to shut the reactor down when fuel damage limits are approached; and
4. Provide a method of detecting leakage from the reactor vessel head flange.

B. Design Features

1. Provide inputs to ECCS and ATWS to assure initiating and interlocking signals occur as required; and
2. Provide signals to operate the reactor relief valves.

QUAD CITIES - UFSAR Revision 13, October 2015 7.6-16 7.6.2.2 Description The reactor vessel instrumentation system provides sensing, indication and alarms of various reactor parameters to the operators and inputs these signals to various control and

protective systems. For details of reactor vessel instrumentation refer to P&IDs M-35 and M-77. The parameters monitored by this instrumentation system and addressed in this section are:

A. Reactor vessel temperature,

B. Reactor vessel pressure,

C. Reactor vessel level,

D. Reactor feedwater flow,

E. Reactor steam flow, and

F. Reactor vessel flange leak detection.

The instruments described in the section may have, depending on their functions, various

classifications. The classification of all instruments are listed in the station's work control

system data base. Those instruments designated as post-accident monitors are described in

Section 7.5.

7.6.2.2.1 Reactor Vessel Temperature Thermocouples are attached to the reactor vessel to measure the temperature at a number of points. These points were chosen to provide data representative of thick, thin, and

transitional sections of the vessel. The data obtained from this instrumentation provides

the basis for controlling the rate of heating or cooling the vessel so that the stress set up

between sections of the reactor vessel is held within allowable limits. The stress is computed from the temperature difference between the various points. The temperatures of the various vessel locations are recorded on a multipoint recorder. The thermocouples

are copper constantan, insulated with braided glass, and clad with stainless steel. They are

positioned under pads attached to the reactor vessel.

[7.6-21]

7.6.2.2.2 Reactor Vessel Pressure

Reactor vessel pressure is both indicated and recorded in the control room and is indicated

in the plant at two separate instrument racks on the mezzanine floor of the reactor

building. Additionally, reactor pressure is monitored to provide control signals for the RPS

high pressure trip, the core spray and low pressure coolant injection (LPCI) low pressure

emergency core cooling system (ECCS) injection permissive and LPCI loop select logic, automatic relief valve operation, and anticipated transient without scram (ATWS) system

operation.

[7.6-22]

QUAD CITIES - UFSAR Revision 7, January 2003 7.6-17 The reactor pressure inputs to the RPS are from pressure transmitters/analog trip units.

The pressure is tapped off the vessel through two sensor lines on opposite sides of the

reactor vessel. The sensor lines are extended outside the drywell to separate instrument

racks. The pressure sensors are grouped on the two independent sensing lines so that a

single event will not jeopardize the ability of the RPS to initiate a scram.

Core spray and LPCI reactor vessel low pressure ECCS injection permissive pressure

switches and ATWS pressure transmitters are grouped into separate divisions and

connected to the same two sensing lines used for the RPS pressure sensors. The ATWS

pressure transmitters are mounted locally on the reactor building mezzanine floor.

Two additional separate instrument lines, attached to the same taps on opposite sides of the reactor vessel, are also extended outside the drywell to separate instrument racks.

These lines are used for the separate divisions of the LPCI loop select logic and for control room indication.

A. Automatic relief valve control, core spray and LPCI injection permissive, and LPCI loop select signals are derived from bourdon-tube pressure switches.

Anticipated Transient Without Scram signals and Reactor Protection System Signals are developed from diaphragm operated pressure transmitters.

B. Two divisions of reactor pressure indicators and/or recorders in the control room receive signals from both bourdon-tube and diaphragm transmitters.

The logic and sequencing, bypasses and interlocks, actuated devices, and system design bases of the systems to which these instruments connect, are discussed in their respective

UFSAR instrumentation and control or system functional description sections:

A. Emergency core cooling systems (HPCI, LPCI 7.3.1, 6.3 mode of RHR, ADS,and core spray)

B. Reactor protection system 7.2, 4.6

C. Anticipated transient without scram 7.8, 15.8

D. Safety relief valve 5.4

7.6.2.2.3 Reactor Vessel Water Level

Reactor vessel water level is indicated and recorded in the control room which is measured by differential pressure transmitters. Level is also indicated locally on two separate racks

on the reactor building mezzanine floor and two separate racks on the reactor building ground floor, which is measured by differential pressure indicators and differential pressure transmitters.

[7.6-23] Reactor vessel water level provides ECCS initiation signals by non-indicating differential pressure transmitters, which also provide trip functions in the Anticipated Transient Without Scram (ATWS) system. The water level is also monitored by level transmitters coupled to the same sensing lines to provide (ATS) signals for the RPS, PCIS and HPCI systems.

In addition, reactor water level is sensed by redundant level transmitters that provide

inputs to the analog trip instrumentation (Section 7.6.2.5).

QUAD CITIES - UFSAR Revision 12, October 2013 7.6-18 Level instruments provide inputs to other systems and are described in sections listed below:

A. Reactor Protection System 7.2

B. Anticipated Transient Without Scram 7.8

C. Emergency core cooling system 7.3.1, 6.3

D. Diesel start 8.3

E. Reactor core isolation cooling 5.4

F. Primary Containment Isolation System 7.3.2

G. Feed pump and turbine trip 7.7, 10.2, 10.4

In response to NRC NUREG-0737 and Generic Letter 84-23, the Yarway columns inside the

drywell have been replaced with two condensate pots per loop and the reference legs were

rerouted through new drywell penetrations to minimize the amount of piping inside the

drywell. This modification was performed to address concerns with potential reference leg

flashing due to elevated temperatures within the containment following an accident.

In response to NRC IN 93-27 and Bulletin 93-03, a Reactor Vessel Level Instrumentation

System (RVLIS) Backfill Subsystem was installed. This subsystem of RVLIS establishes a deaerated water barrier that prevents non-condensable gases in the condensate pot from

diffusing into the reference leg water. The Backfill Subsystem also maintains the

condensate pot water level when non-condensable gases have built up in the condensate pot

steam space. The Backfill Subsystem takes water from the CRD drivewater header, regulates the flow at 4-6 lbm/hr, and injects the water into the reference legs on the inboard

side of the drywell penetration root valve. Only one reference leg from a single condensate

pot is equipped with Backfill injection to avoid excessive thermal hydraulic and thermal

stress to the condensate pot and reactor nozzle.

The sensors and transmitters are grouped so that a single event will not jeopardize the

ability of the RPS to initiate a scram.

The water level in the reactor is controlled by the reactor feedwater level control system.

The primary level sensors for feedwater level control are on separate condensing chambers than those for RPS level functions. The sensors are calibrated in a range which is sensitive

to minor level changes. An isolated third reactor water level signal input is used to

increase the feedwater level signal reliability from sensor failures. A majority based value

is determined from the three level inputs and used to control feedwater flow. The

feedwater control system is discussed in Section 7.7.

Two other redundant transmitters for the two-thirds core height containment cooling permissive interlock use the same condensing chambers as the feedwater control system (Sections 7.4 and 5.4).

In addition to level indicators provided on the sensing lines described above, a separate

level transmitter (with a reference leg condensing chamber connected to the reactor head)

provides (non-ESF) control room indication of level in the upper-most part of the vessel.

This would be used, for example, when filling the vessel prior to head removal.

QUAD CITIES - UFSAR Revision 14, October 2017 7.6-19 7.6.2.2.4 Reactor Feedwater Flow Reactor feedwater flow is monitored by flow transmitters coupled to flow nozzles in the feedwater lines. See Section 7.7 for a further discussion of the reactor feedwater flow

control (level control) system.

[7.6-24]

In addition to the flow nozzles, feedwater flow is also monitored by the Cameron Leading

Edge Flow Meter (LEFM) CheckPlus System. The LEFM CheckPlus System consists of an

electronics cabinet and spool pieces installed in each of the three feedwater supply lines.

Each spool piece contains ultrasonic flow transducers, pressure tap for pressure

transmitters and RTDs (resistance temperature detector) that feed signals back to the electronics cabinet. The LEFM CheckPlus System is only used for feedwater flow

measurement and does not provide input to any control system.

7.6.2.2.5 Reactor Steam Flow

Reactor steam flow is monitored by flow transmitters coupled to the flow restrictors in each

main steam line. Individual steam flows are used by the feed water level control system to

determine total steam flow (section 7.7.5). High main steam line flow (indicative of a main

steam line break) is used as an input to the primary containment isolation system isolation

valve control (Section 7.3.2).

[7.6-25]

7.6.2.2.6 Reactor Vessel Flange Leak Detection

Integrity of the seal between the reactor vessel body and head is continuously monitored at the drain line that is connected to the flange face between the two large concentric O-rings.

The drain line is normally closed. Leakage from the reactor vessel through the inner O-

ring collects in a level-switch chamber and annunciates an alarm. Pressure buildup is also

annunciated. A solenoid-operated valve permits draining the leak system piping so a

measurement of the severity of this leak can be made as the chamber refills.

7.6.2.3 Design Evaluation

Reactor vessel temperature and pressure are sensed and indicated in the control room to

provide the operator with the information required to prevent excessive vessel stresses.

Both the vessel temperature sensors and pressure sensors are provided in quantities which

allow a margin for sensor failures. Pressure sensors used for control room indication and

recording have a history of reliable performance.

[7.6-26]

Thermocouples on the reactor vessel were particularly important during the first few cycles

of heating and cooling of the reactor vessel. Once a good record was obtained and analyzed, the limiting rates of temperature change were related to the temperature observations from

a relatively few thermocouples and from bulk coolant temperature. Redundant

thermocouples are installed to ensure that the operator always has adequate information to

operate the reactor safely. The thermocouples meet the requirements of USAS-C96.1.

QUAD CITIES - UFSAR Revision 12, October 2013 7.6-20 Reactor vessel water level is measured to provide information which can be used to assure that the core is covered and that the separators are not flooded. The use of the level signals in the RPS, ECCS, and the feedwater control systems assures that either the proper level is

maintained, or that the reactor will be shut down automatically.

Tests have been conducted to determine the stability of the vessel level instrumentation in

the presence of rapidly decaying pressures. These tests were conducted at 1500 psig on a standard temperature-compensated head chamber. A series of test runs, starting at 1500 psig, verified the level instrumentation assembly could withstand a depressurization rate of

200 psi/s for the first 3 seconds. At this point, the surface of the water started simmering.

Thereafter, the rate was 100 psi/s. Thus, the pressure was dropped rapidly without

interfering with the stability of the constant head chamber level and the accuracy of the

connected level instrumentation.

Redundant level indicating sensors and transmitters are provided, and there are a

sufficient number of sensing lines so that plugging of a line will not cause a failure to

scram. The arrangement provides assurance that vital protection functions will occur, if

necessary, in spite of a failure in the system.

The feedwater control system level sensors are independent of the RPS level sensors. A failure in the level control which causes the water level to exceed set limits will in no way

influence the level signals feeding the RPS. Feedwater control system failures are

discussed in Section 15.1 and 15.6.

Protection against reactor vessel overfill is provided by reactor high water level trip signals.

Protective actions automatically initiated by reactor high water level include: closure of

the main turbine stop valves (which scrams the reactor and trips the main turbine),

tripping the feedwater pumps, and tripping the HPCI and RCIC systems. These trips protect steam handling equipment from damage due to gross water intrusion. In addition, the high water level trip also serves to maintain fuel thermal margins during the feedwater

controller failure event (as discussed in Section 15.1.2). Redundant logic is used to prevent

a single channel from causing inadvertent trips.

In addition to reactor vessel water level, reactor pressure is sensed for core protection

purposes. A damaging core power transient resulting from a reactor vessel pressure rise is

prevented through the control actions initiated by the reactor pressure signal. The four pressure sensors used by the RPS are arranged so that a plugged line or any other single

failure will not prevent a reactor scram initiated by high pressure.

The reactor vessel flange leak detection system gives immediate qualitative information

about a leak sensed by a pressure buildup. These sensors' sensitivities are such that

degradation of the seal is noted long before excessive leakage occurs. Quantitative information as to the leak rate gives the operator the information necessary for a prudent

evaluation of repair urgency.

7.6.2.4 Surveillance and Testing

All reactor vessel instrumentation inputs to RPS and ECCS are derived from pressure or

differential pressure measurements. The sensing devices are piped so that they may be individually actuated with a known signal during shutdown or operation to initiate a

protection system single logic channel trip. The level switches have indicators so that the

readings can be compared to check for nonconformity.

[7.6-27]

QUAD CITIES - UFSAR Revision 12, October 2013 7.6-21 During equilibrium conditions, either hot or cold, thermocouples monitor an approximately uniform temperature; this information is used to detect abnormalities

The reactor feedwater system control scheme is a dynamic system and malfunctions become

self-evident. The system can at all times be cross-compared with the other level

measurements.

7.6.2.5 Analog Trip Instrumentation

The analog trip instrumentation system consists of an analog sensor (transmitter) and

master/slave trip unit setup which ultimately drives a trip relay. The use of these types of

instruments, including calibration intervals, is described in General Electric Topical Report

NEDO-21617-A.

[2] The instruments in this system meet the EQ requirements of 10 CFR 50.49. [7.6-28]

The power feeds to the transmitters and trip units were selected so that when power is

available to an ECCS pump, power will also be available to the controlling trip unit.

[7.6-29]

Physical location of the components and cable routing is such that divisional separation

criteria is maintained.

[2] [7.6-30]

The analog trip instruments serve as a part of other systems (see the appropriate system

sections):

A. Reactor protection system Section 7.2

B. Primary containment isolation system Section 7.3

C. High pressure coolant injection/core spray Sections 6.3, 7.3

D. Residual heat removal Sections 5.4, 6.3, 7.3

E. Reactor Core Isolation Cooling Section 5.4

F. Feed Pump and Turbine Trip Sections 7.7, 10.2, 10.4

G. Anticipated transient without scram Section 7.8

QUAD CITIES - UFSAR Revision 14, October 2017 7.6-22 7.6.3 References

1. DuBridge, R.A., et al., "Reactor Control Systems Based on Counting and Campbelling Techniques, Full Range Instrumentation Development Program, Final Progress Report," AEC Research and Development Report, U.S. Atomic Energy Commission

Contract AT (04-3)-189, Project Agreement 22 GEAP-4900 (July (1965).

2. "Analog Transmitter/Trip Unit Systems for Engineered Safeguard Sensor Trip Units," G.E. Topical Report, NEDO-21617-A, December 1978.
3. Deleted.
4. CENPD-400-P, Rev. 01, Generic Topical Report for the ABB Option III Oscillation Power Range Monitor.
5. C. Thadani to L. A. England, "Acceptance for Referencing of Topical Reports NEDO-31960 and NEDO-31960, Supplement 1, BWR Owners' Group Long-Term Stability

Solutions Licensing Methodology," (TAC No. M75928) dated July 12, 1993 (SER

attached).

6. NEDO-32465, Licensing Topical Report, BWR Owners' Group Reactor Stability Detect and Suppress Solutions Licensing Basis Methodology and Reload Applications.
7. BAW-10255PA Revision 2, "Cycle-Specific DIVOM Methodology Using the RAMOA5-FA Code," AREVA NP, May 2008.
8. CENPD-294-P-A, "Thermal-Hydraulic Stability Methods for Boiling Water Reactors" Westinghouse Report, July 1996.
9. CENPD-295-P-A, "Thermal Hydraulic Stability Methodology for Boiling Water Reactors" Westinghouse Report, July 1996.

Revision 9, October 2007 QUAD CITIES - UFSAR Table 7.6-1 OPRM SYSTEM TRIPS TRIP FUNCTION TRIP SETPOINT CONFIRMATION COUNT SETPOINT ACTION OPRM Alarm N/A 17* Annunciator OPRM Trip ** ** Annunciator, Automatic Suppression Function (ASF) trip signal to RPS OPRM Bypass Selector Switch Contact N/A Annunciator OPRM Trouble\Inop OPRM Annunciator Relays N/A Annunciator System Enable Nominal Setpoints: 25% thermal power

< 60% recirculation drive flow N/A Annunciator

  • Initial Value - can be varied to meet operating needs ** Refer to cycle specific values in COLR QUAD CITIES - UFSAR 7.7-1 7.7 OTHER INSTRUMENTATION

This section discusses instrumentation and control systems whose functions are not

essential for the safety of the plant. These systems include the following:

A. Reactor control rod control systems including:

1. Control rod adjustment control,
2. Rod block interlocks,
3. Rod position indication system (RPIS), and
4. Control room indicators and alarms.

B. Rod worth minimizer (RWM);

C. Recirculation flow control and economic generation control;

D. Pressure regulator and turbine generator controls;

E. Feedwater (reactor level) controls; and

F. Condenser, condensate, and condensate demineralizer controls.

7.7.1 Reactor Control Rod Control Systems

7.7.1.1 Design Bases

The reactor control rod control system, in conjunction with the recirculation flow control

system discussed in Sections 7.7.3 and 5.4.1, is designed to:

[7.7-1]

A. Provide capability to control reactor power level;

B. Provide capability to control the power distribution within the reactor core;

C. Prevent a single component malfunction or single operator error from causing damage to the reactor or reactor coolant system;

D. Prevent a malfunction from interfering with reactor protective functions; and

E. Provide the reactivity control capability required to prevent fuel damage by meeting the specific core characteristics, parameters, and limitations described

in Sections 4.2, 4.3, and 4.4.

QUAD CITIES - UFSAR 7.7-2 7.7.1.2 Control Rod Adjustment Control (Reactor Manual Control System)

7.7.1.2.1 Control Rod Adjustment Control

Withdrawing a control rod increases core reactivity, causing reactor power to increase until

the increased boiling, void formation, and fuel temperature balance the change in reactivity caused by the rod withdrawal. An increased boiling rate tends to raise reactor vessel

pressure, causing the pressure regulator to open the turbine control valves to maintain a

constant turbine inlet pressure. When a control rod is inserted, the converse effect takes

place. [7.7-2]

The hydraulic portion of the control rod drive system is described and evaluated in Section 4.6. Each control rod has its own drive, including separate control and scram devices. Each

rod is electrically and hydraulically independent of the others, except that a common

discharge volume is used for scram operation. Each rod has an individual pressure source

for scram operation. Rod position is mechanically controlled by the design of the rod drive

piston and collet assembly.

Scram operation of all rods is completely independent of the circuitry involved in rod

positioning during normal operation. Scram operation is described in Section 7.2.

Electrical power for the reactor manual control system (RMCS) is received from an

instrument bus which is fed from an emergency ac bus. The control rod drive system is

actuated, for normal operation, by energizing solenoid-operated valves which direct the

drive water to insert or withdraw the rod.

Control rods are operated one at a time and are withdrawn in preplanned symmetrical

patterns. The allowable patterns have been chosen such that control rod worths will remain below the fuel damage limits, and power distribution in the core will be properly

balanced. The rod selected for withdrawal is electrically controlled so that withdrawal is

not more than 6 inches - one notch - at a time. The one notch withdrawal restriction

may be overridden by the operator by simultaneously manipulating two switches.

7.7.1.2.2 Rod Block Interlocks

Protection is afforded to prevent inadvertent control rod movement (rod block). Refer to

Figure 7.7-1.

With the mode switch in SHUTDOWN, no control rod can be withdrawn. This enforces

compliance with the intent of the shutdown mode.

The circuitry is arranged to initiate a rod block regardless of the position of the mode switch

for the following conditions:

[7.7-3]

A. Any average power range monitor (APRM) upscale rod block alarm - the purpose of this rod block function is to avoid conditions that would require

reactor protection system action if allowed to proceed. The APRM upscale rod

block alarm setting is selected to initiate a rod block before the APRM high neutron flux scram setting is reached. The APRM system is also recirculation

flow referenced in the RUN mode to initiate trip signals to inhibit rod QUAD CITIES - UFSAR Revision 9, October 2007 7.7-3 withdrawal to prevent operating the reactor at excessive power levels with reduced recirculation flow. B. Any APRM inoperative alarm - this assures that no control rod is withdrawn unless the average power range neutron monitoring channels are either in

service or properly bypassed. C. Either rod block monitor (RBM) upscale alarm - this function is provided to stop the erroneous withdrawal of a control rod so that local fuel damage does not

result. Although local fuel damage poses no significant threat in terms of

radioactive material released from the nuclear system, the trip setting is selected

so that no local fuel damage results from a single control rod withdrawal error

during power range operation. The RBM system is also recirculation flow

referenced and operates when power is above 30%. D. Either RBM inoperative alarm - this assures that no control rod is withdrawn unless the RBM channels are in service or properly bypassed.

E. Neutron monitoring system recirculation flow unit either upscale or downscale (inoperative) alarm - this assures that no control rod is withdrawn unless the

recirculation flow units, which are necessary for the proper operation of the RBMs, are operable.

F. Neutron monitoring system recirculation flow unit comparator alarm or inoperative - this assures that no control rod is withdrawn unless the difference

between the outputs of the flow units is within limits and the comparator is in service.

G. Scram discharge volume high water level - this assures that no control rod is withdrawn unless enough capacity is available in the scram discharge volume to accommodate a scram. The setting is selected to initiate a rod block prior to the

scram signal that is initiated on scram discharge volume high water level.

H. Scram discharge volume high water level scram trip bypassed - this assures that no control rod is withdrawn while the scram discharge volume high water

level scram function is out of service.

I. RWM rod insert block and rod withdrawal block - the purpose of these functions is to reinforce procedural controls that limit the reactivity worth of control rods

under low power conditions. The rod block settings are based on the allowable

control rod worth limits established for the design basis rod drop accident.

Adherence to prescribed control rod patterns is the normal method by which this

reactivity restriction is observed.

J. Rod position indication system inoperative - this assures that no control rod is moved unless the rod position information system is in proper operation.

K. Rod movement timer switch malfunction.

L. Rod select power switch in OFF position when movement timer switch is in the HOME or START position above 30% core thermal power as indicated by

APRMs. Below 30% power, the rod out permit light remains on, but no rod

withdrawal is possible without a rod selected.

With the mode switch in RUN, the following conditions initiate a rod block:

QUAD CITIES - UFSAR 7.7-4 A. Any APRM downscale alarm - this assures that no control rod is withdrawn during power range operation unless the average power range neutron

monitoring channels are operating properly or are correctly bypassed. All

unbypassed APRMs must be on scale during reactor operations in the RUN

mode.

B. Either RBM downscale alarm - this assures that no control rod is withdrawn during power range operation unless the RBM channels are operating properly

or are correctly bypassed. Unbypassed RBMs must be on scale during reactor

operations in the RUN mode.

With the mode switch in STARTUP/HOT STANDBY or REFUEL the following conditions

initiate a rod block:

[7.7-4]

A. Any source range monitor (SRM) detector not fully inserted into the core when the SRM count level is below the retract permit level and any IRM range switch

on either of the two lowest ranges - this assures that no control rod is withdrawn unless all SRM detectors are properly inserted when they must be

relied upon to provide the operator with neutron flux level information.

B. Any SRM upscale level alarm - this assures that no control rod is withdrawn unless the SRM detectors are properly retracted during a reactor startup. The

rod block setting is selected at the upper end of the range over which the SRM is

designed to detect and measure neutron flux.

C. Any SRM inoperative alarm - this assures that no control rod is withdrawn during low neutron flux level operations without having proper neutron

monitoring capability available, in that all SRM channels are in service or

properly bypassed.

D. Any intermediate range monitor (IRM) detector not fully inserted into the core -

this assures that no control rod is withdrawn during low neutron flux level

operations unless proper neutron monitoring capability is available, in that all

IRM detectors are properly located.

E. Any IRM upscale alarm - this assures that no control rod is withdrawn unless the intermediate range neutron monitoring equipment is properly upranged

during a reactor startup. This rod block also provides a means to stop rod

withdrawal in time to avoid conditions requiring reactor protection system action (scram) in the event that a rod withdrawal error is made during low neutron flux

level operations.

F. Any IRM downscale alarm except when range switch is on the lowest range -

this assures that no control rod is withdrawn during low neutron flux level operations unless the neutron flux is being properly monitored. This rod block

prevents the continuation of a reactor startup if the operator upranges the IRM

too far for the existing flux level; thus, the rod block ensures that the

intermediate range monitor is onscale if control rods are to be withdrawn.

G. Any IRM inoperative alarm - this assures that no control rod is withdrawn during low neutron flux level operations unless proper neutron monitoring

capability is available in that all IRM channels are in service or properly

bypassed.

QUAD CITIES - UFSAR Revision 9, October 2007 7.7-5 H. Mode switch in STARTUP/HOT STANDBY and the refueling platform over the reactor - this assures that no control rod is withdrawn when fuel is being loaded

into the reactor.

I. Fuel on any refueling hoist and the refueling platform over the reactor - this assures that no control rod is withdrawn when fuel is being loaded into the

reactor.

J. Selection of a second control rod when one control rod is already withdrawn while the mode switch is in REFUEL - this assures that no more than one

control rod is withdrawn during control rod and/or control rod drive

maintenance.

To permit continued power operation during the repair or calibration of equipment for

selected functions which provide rod block interlocks, a limited number of manual bypasses

are permitted as follows:

[7.7-5] A. One SRM channel, B. Two IRM channels, C. Two APRM channels, and D. One RBM channel.

IRM and APRM reactor protection system and rod block bypasses are initiated using

joystick switches in the control room. There is one IRM bypass switch and one APRM

bypass switch for each reactor protection logic channel. Each of the two IRM bypass

switches can be positioned to bypass the trip and rod block functions for one of four IRM

channels, and each of the two APRM bypass switches can be positioned to bypass the trip

and rod block functions for one of three APRM channels. A light in the control room

indicates the bypassed condition.

The bypass circuits are separated such that only one IRM and one APRM can be bypassed

in a single reactor protection logic channel at the same time. Actuation of all four bypass

switches would bypass a total of four neutron monitoring instruments - one IRM and one

APRM bypass in each reactor protection channel and the corresponding IRM and APRM

bypasses in the rod block channels. Under these circumstances, no other IRM or APRM

bypass is possible without first removing an existing bypass. This bypass restriction

ensures that adequate monitoring of the core is maintained.

The SRM detector position rod block is automatically bypassed as the neutron flux

increases beyond a preset low level count rate (100 cps) on the SRM instrumentation. The

bypass allows the detector to be withdrawn, as a reactor startup is continued, until the low level count rate is reached. An automatic bypass of the entire SRM rod block circuit occurs

when all IRM range switches reach range eight or above.

[7.7-6]

An automatic bypass of the RBM rod block occurs whenever the power level is below a

preselected level, or whenever a peripheral control rod is selected. Either of these two

conditions indicates that local fuel damage is not threatened, and that RBM action is not required.

With the exception of OPRM, the same neutron monitoring equipment (APRM, IRM, SRM, and RBM) that is used in the reactor protection system is also used in the rod block

circuitry. One half of the total QUAD CITIES - UFSAR Revision 9, October 2007 7.7-6 number of APRMs, IRMs, SRMs, and RBMs provides inputs to one of the rod block logic circuits, and the remaining half provides inputs to the other logic circuit. One neutron monitoring system recirculation flow unit provides a rod block signal to one logic circuit; the remaining flow unit provides an input to the other logic circuit. The flow unit comparator provides trip signals to each flow unit trip circuit. In addition to the arrangement just described, both RBM trip channels provide input signals into a separate circuit for the nonannunciating rod block control. Scram discharge volume high water level signals are provided as inputs into one of the two rod block logic circuits. Both rod block logic circuits sense when the high water level scram trip for the scram discharge volume is bypassed.

The rod withdrawal block from the RWM trip affects a separate circuit that trips the nonannunciating rod block control. The rod insert block from the RWM function prevents

energizing the insert bus for both notch insertion and continuous insertion.

The APRM and RBM rod block settings are varied as a function of recirculation flow.

Analyses (Section 15.4) show that the APRM or RBM settings selected are sufficient to

avoid both reactor protection system action and local fuel damage as a result of a single

control rod withdrawal error. Mechanical switches in the SRM and IRM detector drive

systems provide the position signals used to indicate that a detector is not fully inserted.

Additional detail on all the neutron monitoring system trip channels is available in Section

7.6.

The rod block from scram discharge volume high water level utilizes a thermal-type level

sensor installed in each scram discharge instrument volume.

[7.7-7]

An additional thermal-type level sensor is installed on each scram discharge volume to

provide an alarm in the control room on high level in the discharge volume as a warning to

the operator. This indication has no automatic actuation or block functions.

7.7.1.2.3 Rod Position Indication System

Control rod position information is obtained from the rod position indication system (RPIS),

which utilizes reed switches in the control rod drive that open or close as a magnet attached

to the rod drive piston passes during rod movement. Reed switches are provided at each

3-inch increment of piston travel. Since a notch is 6 inches, indication is available for each

half-notch of rod travel.

A reed switch is also provided at a location that is beyond the limits of normal rod

movement. If the rod drive piston moves to these overtravel positions, an alarm is sounded in the control room. The overtravel alarm provides a means to verify that the drive-to-rod

coupling is intact, because with the coupling in its normal condition, the drive cannot be

physically withdrawn to the overtravel position. Coupling integrity can be checked by attempting to withdraw the drive to the overtravel position.

7.7.1.2.4 Control Room Indicators and Alarms

The following control room indicators and alarms are provided to allow the operator to

know the status of the control rod system and the control circuitry:

[7.7-8]

QUAD CITIES - UFSAR 7.7-7 A. Rod position,

B. Withdraw bus energized,

C. Insert bus energized,

D. Withdrawal permissive,

E. Rod drift,

F. Notch override,

G. Stabilizer valve selector switch position,

H. Settle bus energized,

I. Rod drive flow control valves' position,

J. Rod drive water pressure control valve position,

K. Drive water pump low suction pressure (alarm only),

L. Drive water filter high differential pressure (alarm only),

M. Charging water (to accumulator) low pressure (alarm only),

N. Control rod drive temperature,

O. Scram discharge volume not drained (alarm only),

P. Scram valve pilot air header low pressure,

Q. Rod worth minimizer conditions are displayed (Section 7.7.2),

R. Nuclear instrumentation system trips are displayed (Section 7.6), and

S. Scram discharge volume high level (alarm only).

7.7.1.3 Design Evaluation

The circuitry described for the reactor manual control system is completely independent of the circuitry controlling the scram valves. This separation of the scram and normal rod control functions prevents failures in the reactor manual control circuitry from affecting the scram circuitry. The scram circuitry is discussed in Section 7.2, Reactor Protection System.

Because each control rod is controlled as an individual unit, a failure that results in the

energization of any of the insert or withdraw solenoid valves can affect only one control rod.

The effectiveness of a reactor scram is not impaired by the malfunctioning of any one

control rod. It can be concluded that no single failure in the reactor manual control system

can result in the prevention of a reactor scram and that the repair, adjustment, or

maintenance of reactor manual control system components does not affect the scram circuitry. Design criteria concerning the possibility of a failure to scram are covered under

Section 7.8, Anticipated Transient Without Scram (ATWS).

[7.7-9]

QUAD CITIES - UFSAR 7.7-8 7.7.1.4 Inspection and Testing

The reactor manual control system can be routinely checked for proper operation by

manipulating control rods using the various methods of control. The system allows for

detailed testing and calibration using standard test and calibration procedures for the

various components of the reactor manual control circuitry.

Routine inspection of the RPIS includes observation of the control rod display once per shift during power operation and during control rod withdrawal for proper control rod position

indication.

[7.7-10]

7.7.2 Rod Worth Minimizer

7.7.2.1 Design Basis

The design basis of the RWM is to serve as a backup to procedural control to limit control

rod worths during startup and low power operation so that in the event of a control rod drop

from the reactor core, the reactivity addition rate will not lead to damage of the primary

coolant system or to significant fuel damage. Operating procedures are the primary defense against high worth control rod patterns. Preplanned, normal rod patterns result in low

individual rod worths. The RWM is not intended to replace a nuclear engineer's selection of

control patterns, but is simply to monitor and reinforce good operating procedures to limit

deviations from these patterns. In performing this function, it should cause minimum

interference with desired operation.

[7.7-11]

7.7.2.2 Description and Definitions

7.7.2.2.1 Definitions

Sequence Step

Steps are the sequential subdivisions of an operating sequence. Each step consists of an

array of rods and a set of insert and withdraw limits that apply to each rod in the array.

The steps are numbered in the order they are to be followed when going up in power. The

withdraw limit of the array specified in a step is the same as the insert limit of that array

in the nearest higher step in the sequence containing that array.

Sequence Array

An array or group consists of a list of control rods. (Both "group" and "array" are used to

describe a unique list of control rods. The meanings are equivalent.) All control rods are

assigned to one and only one array. Rods can only be assigned to an array during the

sequence load procedure. An array can be moved any number of times within a sequence

and at any step. The sequence may optionally contain an array with rods which are to be

termed "out of service." Rods within this out of service array should be fully inserted and

are blocked from movement if selected.

QUAD CITIES - UFSAR Revision 4, April 1997 7.7-9 Operating Sequence

An operating sequence is a sequence of rod movements to be followed by the plant operator

when withdrawing or inserting control rods. The sequence can be printed out or viewed at

the operators RWM screen at any time. A sequence consists of an ordered list of sequence

steps each containing a list of rods (array) and the position the rods should be moved to, from the current position, at that step. The sequence is enforced in reverse order when

coming down in power.

Latched Step

The latched step is the step within the operating sequence compatible at a given time with

the existing distribution of control rod positions. The current control rod pattern is

compared to the loaded sequence and the total number of errors calculated at each step.

The latched step is the step with the least number of total errors. If this criteria yields

more than one step, then the lowest step within this list is defined as the latched step. The

RWM will latch at any other step within this list if that step contains the selected rod.

Notch Position

A notch position of a control rod is defined as any even number 00 - 48. Physically these

numbers correspond to notches located 6 inches apart on the control rod drive mechanism.

A control rod in movement passes through the odd numbers but can only be mechanically

latched at an even numbered position. An odd position is not even transmitted

electronically to the RWM. A control rod not latched at an even position, unless selected

and driving, will be considered to have an invalid position.

Shutdown Margin Test Sequence

The shutdown margin test sequence consists of any step of any two or more control rods.

One rod of the step may be fully withdrawn and the other will have a specified axial

position limit. A shutdown margin sequence may be loaded into the RWM, or the RWM

may be bypassed for the shutdown margin test.

Selection Error

A selection error is defined as the selection of a control rod inconsistent with the loaded

sequence in the RWM.

Insertion Error

An insertion error is defined as the insertion of a control rod inconsistent with the loaded

operation sequence in the RWM. For example, if the operator is withdrawing control rods

exactly according to procedures and has withdrawn several of the rods which are defined to

be in Group 4, the insertion of any withdrawn rod of Group 4 at that time is not considered

an insertion error even though it may be a deviation from planned procedures. However, if

the operator were to insert a rod which is defined in a lower numbered group, that action is

inconsistent with the operating sequence and is an insertion error. This definition is

independent of how far the rod is inserted.

QUAD CITIES - UFSAR Revision 3, December 1995 7.7-10 Withdrawal Error

A withdrawal error is defined similarly to an insertion error. For example, if several rods in Group 4 are not withdrawn, the withdrawal of a rod from any group higher than 4 is a withdrawal error, regardless of how far the rod is moved.

Low Power Setpoint

Above 10% power, the objectives of the RWM are satisfied with no constraints on rod

patterns. This is due largely to the advantageous effects of high initial power level on the consequences of a reactivity insertion accident. Therefore, core average power level derived

from feedwater and steam flow signals is used to remove RWM rod block constraints above

the low power setpoint (LPSP) (10% power), unless they have been manually enabled above

10% power by the operator.

[7.7-12]

Insert Block, Permissive

An insert block is interlocked with the reactor manual control system in such a manner as

to permit or inhibit the insertion of the selected control rod. An insert block is imposed

when a rod has moved one or more notches beyond the limits allowed in the sequence. The

following conditions will cause rod movement insert blocks:

A. Selection and driving in of a rod not within the currently latched step;

B. Selection of a rod deemed to be an insert error (It may be possible to remove this insert block by declaring the rod inoperable and inserting it fully using the "Out Of Service" function from the RWM screen. It also may be allowed to remove the

insert block by using the "Alternate Limit" function from the RWM screen if the

insert rod is at the correct alternate limit);

C. Selection of an improper rod when attempting to recover from an insert error;

D. Selection of any rod other than a withdraw error rod when attempting to recover from a withdraw error;

E. Various other rod selections when implementing special modes such as "Rod Test"; or

F. System initialization or hardware errors.

Withdraw Block, Permissive

A withdraw block is interlocked with the reactor manual control system in such a manner

as to permit or inhibit the withdrawal of the selected control rod. A withdraw block is

imposed when a rod has moved so as to violate the sequence. The following conditions will

cause rod movement withdraw blocks:

A. Selection and driving out of a rod not within the currently latched step;

B. Selection of any rod when attempting to recover from a withdraw error;

C. Selection of any rod except the rods with insert errors when attempting to recover from an insert error; QUAD CITIES - UFSAR 7.7-11 Revision 11, October 2011 D. Various other rod selections when implementing special modes such as "Rod Test"; or

E. System initialization or hardware errors.

Alternate Control Rod Limit

In addition to the insert and withdraw limits specified in the loaded sequence, an alternate

control rod limit may be selected for any rod. The alternate control rod limit for a rod is

defined as being one notch position less than the position limit for that rod at that step.

The only exception to this rule is that the alternate position to the limit of 00 is 02.

Out of Service Rod

An out of service (OOS) rod is a rod which is "pinned" at 00 with no movement or alternate

limits allowed. A control rod which can not be fully inserted may be declared OOS although

more restrictive rules apply to rods incapable of insertion. Placing a rod OOS effectively

removes the rod from its associated array. The rod is ignored during the latch procedure

and will not be considered as an insert or withdraw error during other rod movements.

Rods may be taken OOS in one of two ways: inclusion in a out of service array defined by

the sequence builder, or through use of the RWM screen function "Place Rod Out Of

Service." A control rod which has been declared OOS is not allowed to be moved in any

direction.

Substituted Rod Position

A substitute rod position can be entered through the RWM screen for rods whose positions

are undefined. A substitute value can not be entered for any rod with a "good" position (00, 02, 04, 46, 48, etc.,). A rod with a position that cannot be determined may have a substitute value entered if all attempts by the RWM fail in locating its position. When a substitute

rod's position becomes known, the substitute value is replaced automatically with the good

value and the operator is notified. A maximum of 10 rods may have substitute values

entered. If a substitute rod is selected and driven, the entered substitute value will be

discarded and a new substitute value entered if the new position is bad.

System Mode System mode is selected by a two-position switch in the control room. This switch is used by the operator to bypass the RWM system, if necessary, to remedy hardware problems.

The two positions are labeled NORMAL - BYPASS.

Operational State - Computer Ready

This status applies only when selected system mode is normal. The RWM program will

determine if it can latch and verify a sequence. If the RWM program is able to complete all of its diagnostics and has a valid sequence loaded, it signals a ready state through an

indicating light.

QUAD CITIES - UFSAR Revision 11, October 2011 7.7-12 Rod Test Function

The rod test function is a special case of the normal mode and is selected through the

operator interface. When in this mode, one rod may be fully withdrawn and reinserted ,

only if all other rods are fully inserted. Movement of a control rod is blocked when selected

if any other rod is not fully inserted. If placed in this mode with more than one rod

withdrawn past the fully inserted position, all rod movements are blocked until the rod test

mode is exited.

Control Rod Position

The control rod position is the axial position of a control rod in the core. Valid control rod

positions are 00 - 48, even numbers only.

Control Rod Condition

The condition of a control rod describes the validity of the control rod position. A control

rod may be one or more of the following:

A. Normal,

B. Bad,

C. Substituted,

D. Out of service,

E. Alternate enabled,

F. Selected,

G. Drifting,

H. Insert error, or

I. Withdraw error.

Rod Drift A rod drift is indicated if control rod odd-notched position is detected without being driven by the control rod drive (CRD) system. Rod drift is detected by the RPIS and sent to the

RWM as a digital input.

Analyzed Rod Position Sequence

The analyzed rod position sequence is a set of rules designed to minimize rod worth and

reduce peak fuel enthalpy below limits in the event of a rod drop accident. These rules are

to be followed to the LPSP of 10% rated core thermal power (RCTP).

[7.7-12a]

QUAD CITIES - UFSAR 7.7-13 Revision 11, October 2011 7.7.2.2.2 System Components The RWM function is provided by a computer program running on the redundant process

computer system as well as a dedicated redundant data acquisition system (DAS). The

component interconnections are shown on the block diagram, Figure 7.7-2A.

A. Redundant digital computers PPC-A and PPC-B.

B. Redundant DAS components.

C. Graphic display and control panel switch.

D. Relays interfacing with Reactor Manual Control System to provide rod blocks.

The block diagram illustrates the role of the digital computers in the RWM process.

Software to effect the RWM function resides both on the PPC components as well as on the

DAS components.

7.7.2.2.3 Arrangement

The RWM function consists of a computer program running on redundant process computers

as well as a computer program running on a redundant DAS system. The DAS and process

computer communicate using a redundant ethernet link dedicated to that application.

The color graphics monitor is located on the reactor controls section of the main control

board (901-5) in the control room. A touch screen system is used as the operator input

device. Touching certain areas of the screen enables certain actions. The only other control located on the main control board is a bypass switch used to disable the rod block ability of

the RWM.

The DAS obtains inputs from the Rod Position Indication System (RPIS), Reactor Manual

Control System (RMCS), and other plant instrumentation. Outputs from the DAS are used

to drive relays that interface with RMCS to provide insert and withdraw rod blocks when

required.

QUAD CITIES - UFSAR 7.7-14 Revision 11, October 2011 7.7.2.2.4 Features The operator is presented with a display on the graphics monitor to represent the following

conditions:

A. Rod step number, position, limits,

B. Insertion error rod identification,

C. Withdraw error rod identification, and

D. Current position of all control rods.

A two-position selector switch with normal and bypass positions on the operator's panel

determines the mode of operation. In the normal mode, the active PPC will perform the

function of the RWM. In the bypass mode, the rod blocks will be bypassed by a relay contact. The RWM will receive a signal that it is in bypass mode. The RWM program will

continue to display current rod positions and perform a subset of its normal functions, but

will not provide rod blocks or alarms when errors are detected.

The withdraw/insert permissive is achieved by sets of output relays driven by digital outputs from the DAS. The output relays are arranged in a one out of two taken twice logic

to provide reliability and redundancy. This logic is used in other plant logic including the

reactor trip system and will not be described here.

QUAD CITIES - UFSAR Revision 11, October 2011 7.7-15 7.7.2.3 Design Evaluation

During normal operation in any sequence, with the operator withdrawing and inserting

control rods according to the predetermined procedures, the RWM will neither block, nor

noticeably delay rod movement.

During such operation there will be no alarms except for equipment malfunctions, i.e., control

rod drift or input/output errors.

If the core power level exceeds the low power setpoint, the RWM will not inhibit the selection, insertion, or withdrawal of any control rod, but will only annunciate errors unless blocks have

been enabled to full power by the operator.

When the reactor is operating below the low power setpoint or with blocks enabled to full

power by the operator, the RWM will block movement of a selected control rod in the latched

step upon violation of either the insert of withdraw limit by one notch. The adherence to the

loaded sequence, when in the normal mode, can only be suspended when the operator selects

one of the special modes provided for testing conditions. Bypassing the RWM will also disable

the rod block functions of the RWM.

The control room operator interactions with the RWM program are primarily through the

touch screen. Any other PPC screen in the control room may also be used for this function, providing a means for the operator to control the RWM in the event of a failure of the provided

touch screen. All information necessary for rod movement will be available on the screen.

Different colors are used for quick recognition of an abnormal situation.

The primary screen will normally be displayed on the touch screen and will be the default

screen displayed when that screen is started. Other screens may be displayed at the

discretion of the operator.

7.7.2.4 Surveillance and Testing

Detailed on-demand system diagnostic routines are provided to test the computer and the

control rod interlock networks.

The Technical Specifications, through surveillance requirements, impose the following

verifications and testing to be performed on the RWM: verifications to ensure the correct

control rod sequence is loaded into RWM; verifications on the bypassing of control rods and the

position of those rods to be bypassed; functional testing to verify the rod block and selection error functions, and the verification of the automatic bypass setpoint. Consult the Technical

Specifications for the frequency and details on the RWM surveillance requirements.

[7.7-13]

QUAD CITIES - UFSAR 7.7-16 Revision 11, October 2011 7.7.3 Load Control Design

Load control of a BWR power plant differs from a conventional fossil fuel power plant due

primarily to the sensitivity of boiling to pressure variations. In the conventional plant, the turbine control valves are controlled by the speed/load governor responding directly to system frequency and load demand via the governor setpoint. The resulting pressure

changes in the boiler cause a pressure regulator to adjust the firing rate of the boiler

furnace to match the steaming rate with the turbine steam flow.

[7.7-14]

In the nuclear boiler, power, hence steaming rate, is directly affected by the steam volume

in the reactor core. In turn, the steam volume is sensitive to pressure variations. If the

BWR turbine were controlled as in the conventional plant, opening the control valves would cause reactor vessel pressure to decrease, which would cause the steam volume in the core to increase, which in turn would cause the neutron flux (fission power) to decrease; exactly

the opposite effect desired. Conversely, closing the control valves would cause the reactor

power to increase rather than decrease. The greater the rate of change of pressure, the greater the short-term change in neutron flux. However, the difference in the neutron flux

between two steady-state pressure levels (e.g., 1000 and 1020 psia) is small, providing only

the operating pressure is changed.

The heat addition rate of the BWR boiler can be changed much faster than that of a

conventional boiler, but even so, it cannot be changed fast enough to cope with the effect of

a rapid pressure change on reactor power. A control scheme was adopted which placed the

turbine control valves under control of a high performance pressure regulator (refer to

Section 7.7.4). The steam generation rate in the reactor must first be changed before the

pressure regulator will react to change the turbine steam flow.

This load control scheme is made up of two control systems, a turbine control system which is supplied with the turbine, and a recirculation flow control system which is supplied with the reactor. Figure 7.7-3B is a diagram of the plant load control scheme, and shows the basic features in the power operating mode. Reactor pressure and turbine-generator

controls are addressed in Section 7.7.4. Additional turbine controls are addressed in

Section 10.2.

In addition to the two control systems named above, an economic generation control (EGC) system was originally included in the control scheme for load control. This system is

abandoned.

QUAD CITIES - UFSAR 7.7-17 Revision 11, October 2011 7.7.3.1 Recirculation Flow Control System

Reactor power may be varied over a range of approximately 40% by varying recirculation flow rate. As recirculation flow rate is increased, steam voids are removed from the core faster, thus reducing the existing void accumulation. A positive reactivity insertion occurs

by increasing the moderation of neutrons, resulting in a reactor power increase. The

positive reactivity input is balanced by the negative reactivity effects of higher fuel

temperature and new void formation.

[7.7-15] Speed of the reactor recirculation pumps is varied to change the recirculation flow. A block diagram of the recirculation flow control system is shown on Figure 7.7-3B. An Adjustable Speed Drive (ASD) varies the frequency of the voltage supply to the pump motors to give

the desired pump speed (see Section 5.4.1.2). To change reactor power, a demand signal

from the operator is applied to the master controller. A signal from the master controller

adjusts the setpoint of the controller for each ASD. The recirculating pump motor adjusts its speed in accordance with the frequency of the ASD output voltage. Individual loop

controllers can be placed in manual so that individual speed setpoints can be sent to the

respective ASD. The speed demand from the master or individual controls is used directly

at the ASD and actual speed is not used as a bias to the demand signal. This is considered

"open loop"control.

The ASD includes programmed settings to limit the recirculation pump speed to ensure the MCPR limit is not exceeded during a transient. The ASD setpoints are specified in the

Core Operating Limits Report. The Technical Requirements Manual requires these

setpoints to be verified every 24 months.

[7.7-16]

7.7.3.1.1 Reactor Recirculation Control System (RRCS)

The RRCS digital control system (DCS) provides system control and information to the

operator. It monitors and determines jet pump flows, loop flows and total core flow. Key

parameters related to core flow and the Reactor Recirculation System's operation are

processed and displayed at the Operator Station in the Main Control Room. All recirculation pump speed control logic and operator interfaces are provided by the RRCS.

Included in the RRCS logic are the control interlocks, core flow runbacks, alarms, and

trending.

A digital controller for RRCS is within the common feedwater (FWLC) and recirculation

control (RRCS) cabinet are used for communication with the common FWLC and RRFC

equipment, such as the Operator Station (OS) and the Engineering Workplace (EW).

Separate gateway computers for the FWLC and RRCS systems are used for supporting data

transfer to a local area network (LAN) and transient recording of data.

7.7.3.2 Economic Generation Control System - Abandoned

System is abandoned in place.

[7.7-17] [7.7-18]

QUAD CITIES - UFSAR 7.7-18 Revision 11, October 2011 7.7.3.3. Failure Mode and Effects Analyses

The failure Modes and Affects Analysis (FMEA) and reliability analysis of the digital

Reactor Recirculation Control System (RRCS) are provided in Westinghouse report P03-

342, Revision 2. The effects of the original plant design failures bound any possible failures

existing in the RRCS. The digital RRCS has self checking ability and designed failure

responses are programmed for loss of input signals, parameters out of specified range, failure of internal self-checks, power supply failures, and other failures to minimize the

affect of these failures and to prevent plant transients. The parameters that determine the

worst case recirculation flow related accident are based on the settings, limits, and rate of

change limits of the MG Set scoop tube positioner, not the control functions and algorithms

in the RRCS controller.

7.7.3.3.1 Section Deleted

7.7.3.3.2. Load Demand Error Signal Failures - Function Not Available

Load following and Automatic Flow Control are no longer a plant control option.

7.7.3.3.3. Section Deleted

QUAD CITIES - UFSAR 7.7-19 Revision 11, October 2011 7.7.3.4. Design Evaluation

The recirculation flow control arrangement contributes to the stable response of the reactor.

The stability of the unit is discussed in Section 4.3. Chapter 4 describes reactor thermal

margins under the flow control mode. Figure 4.4-1 depicts typical reactor power-flow

behavior lines: with flow and power initially at any point on the curve, a flow change will

cause the power to change along the path indicated by the curve. Malfunction of the flow controller can cause either a recirculation flow increase (insertion of positive reactivity) or a decrease (high power to flow ratio). Inadvertent recirculation flow increases are less severe

than the transient caused by starting a recirculation pump in a cold loop, and inadvertent

recirculation flow decreases are less severe than a trip of one or two recirculation pumps.

These malfunctions are discussed in Chapter 15.

[7.7-19]

The recirculation flow control system has a loop selection network which is controlled by

differential pressure(p) instrumentation in the low pressure coolant injection (LPCI) break detection system. See Section 6.3.

The p instrument trip points are selected such that the instruments null (essentially zero differential) when the reactor recirculation pumps are delivering rated flow. This will optimize the setting of the instruments should there be even a slight difference in the loss

coefficient of the jet pump assemblies.

The trip setpoints for these instruments will remain the same regardless of the number of

recirculation loops in operation. During one pump operation, a reactor pressure permissive

will prevent the loop selection network until reactor pressure has dropped to approximately

900 psig (allowable value is specified in the Technical Specifications). This requirement

adjusts the selection time to allow for pump coastdown and thus optimize sensitivity and

still ensure that the network is not delayed unnecessarily. Stopping the recirculation pump

is necessary to eliminate the possibility of breaks being masked by the operating

recirculation pump pressure. Thus, the low reactor pressure permissive allows the same

trip point setting regardless of the number of recirculation loops in operation.

The trip setpoint is set at about 0.75 psi (allowable value is specified in the Technical

Specifications). The only requirement is that any positive p would result in the selection of Loop A, any negative p would result in the selection of Loop B.

7.7.3.5 Other Reactivity Control Systems

The standby liquid control system is discussed in Section 9.3.5.

QUAD CITIES - UFSAR 7.7-20 Revision 11, October 2011 7.7.4 Pressure Regulator and Turbine-Generator Controls

7.7.4.1 Design Basis

The pressure regulator and turbine-generator controls are integrally connected to

accomplish the functions of controlling reactor pressure and turbine speed. Specifically, reactor pressure must be prevented from increasing to too high a value during load

maneuvers, and turbine speed must be maintained below design limitations. The system

must result in stable response for all anticipated maneuvering rates.

[7.7-20]

7.7.4.2 System Description

Control and supervisory equipment for the turbine-generator are arranged for remote

operation from the turbine-generator control panel board or console in the control room. In

addition, turbine oil pressure and steam extraction pressure are transmitted to receivers on the panel board. Normally, the pressure regulator controls turbine control valve position to

maintain constant reactor pressure. The ability of the plant to follow system load is

accomplished by adjusting the reactor power level, either by regulating the reactor coolant recirculation system flow or by moving the control rods. A block diagram of the turbine

control system is shown on Figure 7.7-3B.

However, the turbine speed control can override the pressure regulator, and the turbine control valves will close when an increase in system frequency or a loss of generator load

causes the speed of the turbine to increase. In the event that the reactor is delivering more

steam than the turbine control valves will pass, the excess steam will be bypassed directly

to the main condenser automatically by pressure-controlled bypass valves.

The total capacity of the bypass valves is equal to 33.3% of the rated reactor flow. Load

rejection in excess of the bypass valves' capacity, which occurs due to generator or tie line

breaker trips, will cause the reactor to scram.

The pressure regulator and turbine-generator controls utilize a triple modular redundant (TMR) design with a separate turbine controller, pressure controller and overspeed

protection module. Each controller / module consists of three (3) separate processors, utilizing a software-implemented fault-tolerance (SIFT) technology that allows the

controller to remain on-line if one of the processors fails.

The TMR turbine controller is tasked with turbine control and protection, the TMR

pressure controller performs the steam bypass and pressure control functions and the TMR

protection module provides a second level of overspeed protection. The turbine controller

and pressure controller communicate over redundant unit data highways to coordinate turbine and pressure control requirements. The protection module functions independent

from the turbine and pressure controllers with dedicated speed sensor inputs.

The separate TMR system for control of the turbine bypass valves and control of the turbine

allows the two functions to maintain independence from a control hardware and software

standpoint. For critical functions, the controllers utilize triple-redundant process sensors and will continue operation if one of the process sensors fail. The pressure controller is

designed to continue operation even if two (2) of the three (3) sensors fail.

QUAD CITIES - UFSAR 7.7-21 Revision 11, October 2011 The maximum combined flow limit (MCFL) function of the control system limits the combined steam flow through the turbine control and bypass valves to a value of at least 110% of rated reactor steam flow but not more than 125%. The low MCFL value is

important for slow power increase events and defines the amount that steam flow can

increase before the plant will begin to pressurize. The upper MCFL value is intended to

prevent a Group I isolation on main steam line high flow.

Normally, the bypass valves are held closed and the pressure regulator controls the turbine

control valves. All the steam production is normally used to make electrical power. If the

speed control or load limit reduces the steam flow to the turbine, the bypass valves will

open to pass steam directly to the main condenser, to maintain a constant system pressure.

If steam flow exceeds the combined capacity of the turbine control valves and the bypass valves, system pressure will rise and scram the reactor. A rapid reduction of electrical load

will initiate a reactor scram as described in Section 7.2.

The turbine stop valves are equipped with limit switches which open when the valve has moved from its fully opened position. These switches provide a scram signal to the reactor

protection system, anticipating the resulting reactor high pressure condition. The turbine

stop valve scram signal is discussed in Section 7.2.2.5.

To protect the turbine, the following conditions initiate closure of the four turbine stop

valves (see Section 10.2):

A. High reactor vessel water level,

B. Low lube oil or bearing oil pressure,

C. Overspeed,

D. Excessive thrust bearing wear,

E. Generator electrical faults,

F. Remote and local manual trips,

G. Vacuum trip,

H. Low EHC hydraulic pressure,

I. Loss of feedback signal trip,

J. High water level in moisture separator, K. Loss of stator cooling without runback, and

L. High vibration trip when enabled.

7.7.4.3 Design Evaluation

The pressure regulator and turbine-generator design is such that the system provides a

stable response to normal maneuvering transients. Section 4.3 evaluates the stability of

the overall boiling water reactor cycle, including the pressure and turbine control.

QUAD CITIES - UFSAR 7.7-22 Revision 11, October 2011 The bypass valves are capable of responding to the maximum closure rate of the turbine admission valves such that reactor steam flow is not significantly affected until the magnitude of the load rejection exceeds the capacity of the bypass valves. Load rejections in excess of bypass valve capacity may cause the reactor to scram due to high pressure, high

neutron flux, or rapid electrical load reduction. If power is greater than the bypass capability, any condition causing the turbine stop valves to close, will directly initiate a

scram before reactor pressure or neutron flux have risen to the trip level.

The pressure regulator or controller can be assumed to fail in either of two ways: opening

the turbine control valves or the bypass valves, or closing them. These malfunctions are

discussed in Chapter 15; in either case, fuel damage does not occur. The triple modular

redundant design reduces the probability that pressure regulator malfunction will cause

operational problems.

7.7.5 Feedwater Level Control System

7.7.5.1 Design Basis

The feedwater control system is designed to regulate feedwater flow to the reactor vessel

such that reactor vessel water level is maintained to an operator controlled setpoint. There

are two basic modes of operation: single-element and three-element control.

[7.7-21]

7.7.5.2 System Description

7.7.5.2.1 Description of Single-Element Control

Single-element control is a mode of operation which controls feedwater flow based only on

reactor water level deviations. The actual measured level is compared to the level set on the controller. The regulating valve is adjusted by a signal proportional to the level error

signal. Feedwater and steam flow signals have no effect under single-element control.

Single-element control is used during plant start-up conditions or when at low reactor

power. The operator can select single-element control at anytime. The feedwater level control digital control system (DCS) can automatically choose single-element control when

appropriate.

7.7.5.2.2 Description of Three-Element Control

Another feedwater control mode is three-element control. In this control mode, the level of

the water in the reactor is controlled by a feedwater controller which receives inputs from

reactor vessel water level, feedwater flow, and steam flow transmitters.

During steady-state operation, feedwater flow exactly matches steam flow and the water

level is maintained. A change in steam flow is immediately sensed and the system adjusts

the opening of the feedwater control valves to balance the two flows and maintain level.

QUAD CITIES - UFSAR 7.7-23 Revision 11, October 2011 7.7.5.2.3 Control Signal Inputs

Reactor vessel level signals used by the feedwater level control system are indicated and/or recorded in the control room. Level sensors are described in Section 7.6.2.2.3.

Feedwater flow is monitored by flow transmitters coupled to flow nozzles in the feedwater

lines. The total feedwater flow is the summation of the signals from the three feedwater

lines. [7.7-22]

Steam flow is monitored by four flow transmitters coupled to four flow restrictors in the steam lines. The level control system calculates total steam flow by using the average of

the valid input signals and multiplying by four. A straight sum of the flows method can be

selected by the operator for testing purposes.

Reactor vessel majority water level, total feedwater flow, and total steam flow are displayed and recorded in the control room. High and low reactor vessel water level are annunciated

in the control room. High water level will cause the feedwater pumps to trip, to prevent

overfill. A low water level can cause initiation of the level scram function by RPS.

[7.7-23]

Three level signal inputs are used by the control system and a majority based value is used

to control feedwater flow. The feedwater valves fail "as is," and the valves may be switched

to manual control in the event of failure.

Each reactor feedwater pump has recirculation controls which pass feedwater back to the

condenser when individual feed pump flow is below minimum flow required to cool the

pumps. A staggered pump tripping logic is used for low suction based trips. A low-low

suction pressure will trip all feed pumps simultaneously.

To enable the feedwater system to make maximum contribution to reactor core cooling in

the event of small breaks, the reactor feedwater pumps are flow limited to protect against a

potential pump runout when the rated capacity of the pumps is exceeded. The level control

system limits total feedwater flow to a value dependent on the number of feed pumps running. This protection is referred to as feed pump runout protection (FPRP). See Section

10.4 for further discussion of the feedwater system.

7.7.5.2.4 Digital Control System

All inputs and outputs to the feedwater level control are processed by a digital control system (DCS). The digital control system provides the analog signal filtering, conversions, and setpoints. The digital control logic and control algorithms are contained in the DCS

software.

QUAD CITIES - UFSAR 7.7-24 Revision 11, October 2011 Manual pushbotton stations are provided on the main control panels for controlling the

level setpoint, mode of control, and for taking manual control of individual flow regulating

valves. A DCS Operator Station is provided in the Main Control Room to provide feedwater

level DCS graphic displays and operator interface. The operator can acknowledge system

alarms, control the system logic, adjust the level setpoint, change control options, and

position the regulating valves from the DCS Operator Station.

7.7.5.2.5 Supported System Requirements

The following are output functions of the feedwater level control to other plant systems:

  • Reactor Recirculation system runback logic
  • RWM LPSP permissive logic (Section 7.7.2)
  • Feed Pump logic for low suction pressure conditions (Section 10.4.7)
  • Condensate Booster Pump Minimum Flow control
  • Stand-by Condensate Booster Pump Auto Start logic (Section 7.7.6.2)
  • Hydrogen Addition (total steam flow signal)
  • Plant Process Computer

7.7.5.3 Design Evaluation

Key feedwater system parameters are recorded and, upon abnormal conditions, annunciated in the control room; the operator can monitor system operation continuously.

Feedwater level control signals are redundant, and equipment design is reliable, minimizing the possibility that malfunctions will result in level control difficulties.

The feedwater level control system is designed to maintain water level at an operator

controlled setpoint which is typically at the mid-point of the feedwater level control instrument range of 0 to 60". Proper control of reactor water level will prevent inadvertent

RPS trips and main feed pump trips from a level that is too low or too high.

Feedwater control system malfunctions could result in maximum or zero feedwater flow.

These malfunctions are discussed in Sections 15.5 and 15.6. In either case, fuel failure does

not occur.

The instrumentation for control of the feedwater system is separate from reactor protection

system instrumentation, thereby limiting the consequences of sensor malfunctions. Reactor

overfill protection will trip the feedwater pumps. This function is not performed by the

feedwater level control system.

QUAD CITIES - UFSAR Revision 11, October 2011 7.7-25 7.7.6 Main Condenser, Condensate, and Condensate Demineralizer

7.7.6.1 Design Bases

The main condenser, condensate, and condensate demineralizer systems' control is

designed to provide indications of major system trouble. Main condenser sensors must

provide inputs to the reactor protection system to anticipate loss of the main heat sink and to protect against condenser overpressure. The condensate system controls must ensure

adequate cooling to the condensate pumps.

[7.7-24]

7.7.6.2 System Description

The condensate pumps take suction from the main condenser hotwell. The discharge

passes through the steam jet air ejector inter- and aftercondensers, the gland seal

condensers, and the off-gas condensers. The flow then passes through the condensate

demineralizers and then to the suction of the condensate booster pumps. The condensate

and condensate booster pumps are run with a common motor. The discharge of the booster

pump passes through the low pressure feedwater heater strings and then to the suction of

the feedwater pumps.

When a condensate/condensate booster pump is in standby, detection of low pressure at the

condensate booster pump discharge header starts the standby condensate/condensate

booster pump. In addition, if any of the running pumps trip, a pump in standby will

autostart. An air-operated control valve, located on the discharge header of the condensate

booster pump recirculates condensate back to the condenser during plant startup.

Minimum cooling flow through the condensate pumps, air ejector condensers, gland seal

condenser, and off-gas condenser is maintained by the feedwater pump minimum flow

valves. [7.7-25]

Conductivity of condensate both upstream and downstream of the demineralizers is

measured, recorded, and actuates an alarm on high conductivity. The upstream

conductivity sample point is on the influent header common to all of the demineralizers.

[7.7-26]

Main condenser hotwell level is indicated locally, recorded in the control room, and is

automatically or manually controlled by either making up to or returning condensate from, the condensate storage tank. Vacuum switches monitoring condenser vacuum provide

scram signals to protect the reactor from loss of the main heat sink; protection for the

condenser itself is assured by closure of the turbine stop and bypass valves as condenser

absolute pressure increases above a preset value.

7.7.6.3 Design Evaluation

Indication of key parameters from the main condenser, condensate system, and condensate

demineralizer system are provided in the control room. The operator is kept cognizant of

the conditions of the systems. Abnormal conditions are annunciated, so that the operator may take appropriate action. The reactor is protected from loss of the main heat sink by

main condenser low vacuum scram signals; the vacuum sensors meet the design

requirements established for all reactor protection system functions (Section 7.2). To

protect the condenser from overpressure, a decrease of condenser vacuum below the scram

set point will initiate closure of the turbine stop valves and bypass valves.

(Sheet 1 of 1)

Revision 8, October 2005 QUAD CITIES - UFSAR

TABLE 7.7-1

EGC CONSOLE TOP PLATE FUNCTIONS - ABANDONED EQUIPMENT

Pushbutton Switches Purpose TRIP Used to remove unit from local program control or remote automatic control. Flashing light in switch

indicates control is automatically tripped. Depressing

TRIP pushbutton will change to steady light. Manual trip will cause steady light only. AUTO Used to permit remote automatic control by Raise-Lower impulses from the System Power Supply Office.

Light in switch indicates AUTO control is selected. LOWER PROGRAM Used to lower generation under local program control.

Light in switch indicates selection. RAISE PROGRAM Used to raise generation under local program control.

Light in switch indicates selection. PRIMARY PULSE Used to select Raise-Lower control impulses from primary telemetering channel. Light in upper half

indicates selection. Light in lower half indicates

pulsing. BACKUP PULSE Used to select Raise-Lower control impulses from backup telemetering channel. Light in upper half

indicates selection. Light in lower half indicates

incoming pulsing. LAMP TEST Used to illuminate lamps to test for defective ones.

Setters HIGH LIMIT Establishes unit MW generation high (raise) regulating limit. LOW LIMIT Establishes unit MW generation low (lower) regulating limit. RATE OF CHANGE LIMIT Establishes maximum ramp rate in MW/min for unit.

(Sheet 1 of 1)

Revision 8, October 2005 QUAD CITIES - UFSAR

TABLE 7.7-2

EGC STATUS INDICATORS (ANNUNCIATORS) - ABANDONED EQUIPMENT

Indicator Description ACT FAIL Electrohydraulic control system interface unit or governor motor actuator failure. HIGH LIMIT Unit generation equals or exceeds HIGH LIMIT setting.LOW LIMIT Unit generation equals or exceeds LOW LIMIT setting.

RAISE OUTPUT Raise impulse from electrohydraulic control system interface unit or governor motor actuator to generating

unit control system. LOWER OUTPUT Lower impulse from electrohydraulic control system interface unit or governor motor actuator to generating

unit control system. RAISE INPUT Raise input pulse to controller.

LOWER INPUT Lower input pulse to controller.

EXT TRIP Interlocks in trip circuits from contacts provided elsewhere in the boiler, turbine or generator control

system. EXT BLOCK Control action suspended by contacts provided elsewhere in the boiler, turbine or generator control

systems. DECREASE RATE LIMIT Control action to decrease generation limited at second rate by controller (not used). INCREASE RATE LIMIT Control action to increase generation limited at second rate by controller (not used). Signal Light RESERVE EMERGENCY Signal light initiated manually by Load Dispatcher in System Power Supply Office to indicate system

generation deficiency.

QUAD CITIES - UFSAR Revision 13, October 2015 7.8-1 7.8 ANTICIPATED TRANSIENT WITHOUT SCRAM MITIGATION SYSTEM

7.8.1 Introduction

This section discusses the anticipated transient without scram (ATWS) mitigation system.

Related topics and systems include the standby liquid control system (SBLC), discussed in Section 9.3.5; the control rod drive (CRD) system, Section 4.6; the reactor recirculation

system, Section 5.4; the reactor protection system (RPS), Section 7.2; the residual heat

removal (RHR) system (suppression pool cooling mode), Section 5.4; and the ATWS accident

analyses, Section 15.8. For diagram of Nuclear Boiler Recirculation Pump Trip ATWS

piping refer to P&IDs M-35 and M-77.

An anticipated transient without scram is a postulated operational transient (such as loss

of feedwater, loss of condenser vacuum, or loss of offsite power) accompanied by a failure of the reactor protection system to shut down the reactor. Even though the reactor protection

system has been shown to be highly reliable, it is postulated that a common mode failure in

either the electrical or mechanical portion of the system is possible.

[7.8-1]

Since a normal scram is assumed to be unavailable for reducing the reactor power, and

since the transient event is one in which power reduction is necessary, another method of

reducing power is needed. Two automatic ATWS functions are provided: recirculation

pump trip (RPT), which mitigates the short-term effects, and alternate rod insertion (ARI),

which mitigates long-term effects. Should both the RPS and ARI fail to insert the control

rods, the standby liquid control system would be manually initiated to control reactivity.

[7.8-2]

The trip of the reactor recirculation pumps causes a quick reduction in core flow which

increases core void generation, thus introducing negative reactivity and decreasing reactor power. The quick power reduction brings reactor pressure, neutron flux, and fuel surface

heat flux down rapidly enough to limit the peak pressure, clad oxidation and peak fuel

enthalpy so that neither reactor coolant pressure boundary breach nor fuel failure occur.

An analysis was performed which considered the trip of the adjustable speed drive (ASD)

controller and feed breaker trip.

Alternate rod insertion (ARI) is a means of control rod insertion which is motivated

mechanically by the normal hydraulic control units and control rod drives, but which

utilizes totally separate and diverse logic from RPS. Alternate rod insertion energizes

valves which cause the scram valve pilot air header to bleed down. Although this type of

rod insertion does not eliminate the short-term consequences of the assumed failure of

normal scram action, it does reduce the long-term consequences. The most significant

long-term consequences involve containment limits, particularly suppression pool

temperature.

[7.8-3]

7.8.2 Design Requirements

The ATWS rule (10 CFR 50.62) requires the following three elements to mitigate ATWS

events: [7.8-4]

1. Recirculation pump automatic trip equipment;
2. An alternate rod insertion system, diverse from RPS, with redundant scram air header exhaust valves; and QUAD CITIES - UFSAR Revision 14, October 2017 7.8-2 3. A standby liquid control system that meets minimum flow and concentration requirements.

The RPT portion of the ATWS mitigation system is designed to perform its function in a

reliable manner, and to conform to the standard NRC approved Monticello tripping logic

design[1]. [7.8-5]

The overall requirements for the ARI portion of the ATWS mitigation system are:

[7.8-6]

A. The system should be diverse from RPS;

B. The system shall be designed so that any component whose single failure can cause insertion of all control rods shall be highly reliable;

C. The system should be testable in service;

D. The system should be designed so that, as much as possible, no single component failure can prevent total mitigation action; and

E. All hardware should be of high quality and environmentally qualified.

For an ATWS (per 10 CFR 50.62), the standby liquid control system must be capable of

injecting into the reactor pressure vessel a borated water solution equivalent in reactivity control to injecting 86 gal/min of 13 Wt. % sodium pentaborate at natural B 10 concentration into a 251-inch ID reactor vessel for a given core design. The specific requirements of flow

rate and concentration for Quad Cities Station are addressed in Section 9.3.5.

[7.8-7]

7.8.3 Mitigation System Description

All of the anticipated transients, which require mitigation in the unlikely event of an ATWS, quickly reach at least one of two conditions which are readily sensed and from

which mitigating actions may be initiated. These conditions are high reactor vessel

pressure and low-low reactor water level.

[7.8-8]

The ATWS mitigation system consists of reactor pressure and reactor water level sensors

and trip units, logic, power supplies, and instrumentation to automatically initiate RPT and

ARI. The reactor dome pressure automatic actuation setpoints of 1250 psig for Unit 2 and 1200 psig for Unit 1 (analytical limit) were chosen to be slightly above the relief valve setpoint. The low-low reactor water level automatic actuation point of -59 inches (analytical limit) is consistent with that level at which the recirculation pumps trip, and

high pressure coolant injection and reactor core isolation cooling are initiated. The

allowable values for the pressure and level actuations are included in the Technical

Specifications.

Certain manual actions are required of the operator. Suppression pool cooling and standby

liquid control must be initiated manually as required by emergency operating procedures.

The following subsections describe the capability and requirements for manual initiation of

RPT and ARI. Alarms and indications are available to the operator to allow manual actions

within the time limits. In addition to the alarms and indications which are initiated by

RPS scram logic, other annunciator windows actuate when the reactor water level or

reactor pressure reach the ATWS setpoints. Therefore, during an ATWS event, the

operator is alerted that an ATWS event has occurred and then has sufficient time to

perform the required manual actions.

[7.8-9]

QUAD CITIES - UFSAR Revision 13, October 2015 7.8-3 7.8.3.1 Recirculation Pump Trip The ATWS mitigation system automatically initiates a RPT of both recirculation pump ASD

controllers and feed breakers on a two-out-of-two trip logic in either of two channels upon

either continuous low-low reactor water level for approximately 9 seconds or high reactor

pressure. The performance characteristics are:

[7.8-10]

Logic Delay for Trip (Sec) (Including dynamic response <= 0.53 of the sensors and trip logic action of the ASD units.)

Pump Inertial Constant (JN/ft, Sec) <= 3.0

Manual RPT is achieved by a manual trip of either the ASD emergency stops or ASD feed breakers. The breaker control switches are located at panel 901(2)-4 and at the switchgear

breakers, and the emergency stop pushbuttons are on 901(2)-4, at the ASD control panel, and at the 1(2)-2201(2)-25A/B panels. Manual RPT should be performed following receipt of alarms indicating an ATWS has occurred if automatic RPT does not occur:

[7.8-11]

High Torus Water Average Temperature Alarm

High Reactor Dome Pressure Alarm

Reactor Low-Low Water Level Alarm

7.8.3.2 Alternate Rod Insertion

The ATWS mitigation system logic automatically energizes the ARI valves when the ATWS

reactor vessel high pressure trip setpoint is reached, the ATWS low-low reactor water level trip setpoint is reached, or the manual switches are actuated.

[7.8-12]

Two manual initiation pushbutton switches are provided in the control room at panel

901(2)-5 for each division of ARI logic. Failure of automatic initiation cannot prevent

manual initiation. In order to avoid an inadvertent manual initiation of ARI , the two

initiation switches per division must first be armed by rotating a collar integral to each

pushbutton. Once armed and then depressed, the pair of switches associated with a

division will activate the ARI trip function.

[7.8-13]

Manual ARI should be initiated upon reaching any of the following alarm conditions:

High Torus Water Average Temperature Alarm

[7.8-14] High Reactor Dome Pressure Alarm

Reactor Low-Low Water Level Alarm

Control Rod Drive Position Indication - Not inserted after scram annunciation QUAD CITIES - UFSAR Revision 13, October 2015 7.8-4 7.8.3.3 Alternate Rod Insertion Valves Upon ATWS initiation (automatic or manual), the ARI solenoid valves as shown in P&ID

M-41 are energized to block the instrument air supply to the scram air header and to depressurize the scram air header by venting air to atmosphere. Depressurization of the

scram air header causes the scram valves to open resulting in the drives scramming. All

ARI valves are normally de-energized. The ARI valving system operates as follows:

[7.8-15]

A. There are two sets (2 divisions) of valves installed on the scram air header. Each division has sufficient capacity to accomplish rod insertion. Each division of

valves consists of the following three valves:

1. Two ARI valves are simply normally closed valves that open when energized to depressurize the scram air header.
2. One ARI valve is a three-way valve installed in the scram air header supply line. This valve is normally positioned to allow air to be supplied to the

scram air header. When energized, this valve repositions to close off the

supply air and vent the scram air header to the atmosphere.

B. Once actuated, the ARI valves remain energized between 35.9 and 37.8 seconds to ensure the scram air header is adequately depressurized. The timer setting

for the seal-in is based on the fact that full rod insertion could be prevented if the

ARI automatic reset occurs in less than 35.9 seconds. After this delay, if the

initiation signal has cleared, the ARI valves are de-energized. If the initiation

signal is still present after the delay, the ARI valves remain energized until the

initiation signal clears.

[7.8-16]

7.8.4 Design Evaluation

The sensors, trip units, and actuation relays (with the exception of the RPT reactor low-low

water level trip time delay and the ARI reset circuitry) are common to both RPT and ARI.

Thus, the automatic initiations occur concurrently (except for the RPT low-low water level

time delay) at identical setpoints. Therefore, the following design analyses dealing with the

inputs, the logic, and logic power supply apply equally to ARI and RPT.

[7.8-17]

The RPT is modeled after the NRC-approved Monticello tripping logic design with the addition of a time delay (of approximately 9 seconds) for the low-low water level trip. The time delay for RPT on low-low water level has an insignificant affect on ATWS

consequences and is desirable to avoid making the consequences of a postulated loss-of-coolant accident (LOCA) more severe. The final tripping devices are the ASD emergency

stop and feed breaker.

[7.8-18]

The ARI function requires control rod start of motion within 34.6 seconds and full insertion

within 38.6 seconds of ARI actuation. Test results indicate that all scram inlet and outlet valves are open within 30 seconds. Section 7.8.3.3 describes the seal-in and reset time

delay of the ARI values. Based on the NRC-approved General Electric Company Topical

Report NEDE-31096-P-A

[1], ARI achieves the design objectives. The most limiting of these objectives (pressure suppression pool temperature) requires full rod insertion within

approximately 60 seconds.

[7.8-19]

The ARI design is safety-related and segregated into two electrical divisions: namely

Division I and Division II which are maintained separate.

[7.8-20]

QUAD CITIES - UFSAR Revision 14, October 2017 7.8-5 The ARI system utilizes valves which are normally de-energized but which are energized to

perform their safety functions. The ARI valves are powered from dc sources. This is in

contrast to the RPS which employs ac-powered valves which are de-energized to initiate a

scram.

The ARI system uses an analog transmitter/trip unit configuration. The transmitters are

separate from sensors used for the RPS. In addition, the trip units utilized are separate

from the process instruments used for the RPS.

[7.8-21]

The ARI trip setting for reactor pressure is 1250 psig for Unit 2 and 1200 psig for Unit 1 (analytical limit) and for reactor vessel water level is -59 inches (analytical limit) with

respect to reactor level instrument zero. The RPS trip setting for reactor pressure is 1060

psig (analytical limit) and for vessel level is 0 inches (analytical limit) with respect to

reactor level instrument zero. Therefore, the automatic setpoints for ARI actuation have

been selected such that they will not pre-empt the RPS scram function. The allowable

values for the pressure and level actuations are included in the Technical Specifications.

[7.8-22]

For each actuation parameter (e.g., low-low water level) the logic is arranged in a two-out-

of-two configuration per division. This allows individual sensors, trip units, etc., to be

tested or calibrated during plant operation without initiating the ARI system.

[7.8-23] Reactor vessel water level sensors that drive the ATWS functions (ARI and RPT) are shared and also drive various actuation and trip functions that receive level signals. See sections

7.6.2.2.3, "Reactor Vessel Water Level" and 7.6.2.5, "Analog Trip Instrumentation". In

addition, the ATWS channel A and B sensors provide input to the plant process computer

and to the main control room narrow range level instrumentation.

QUAD CITIES - UFSAR 7.8-6 7.8.5 References

1. General Electric Licensing Topical Report, "Anticipated Transients Without Scram; Responses to NRC ATWS Rule 10 CFR 50.62," NEDE-31096-P-A, February 1987.

480 VAC AT TURBINE BLDG MCC 18-2 (28-2)C M APS MG A G iA-i (2A-i)EPA 1A-2 (2A-2)EPA r~(RPS 1 A 480 VAC ATTURBINEBLDG MCC 19-2 (29-2)0)M RPSMGB G 10-1 (20-1)EPA 1B-2 (2B-2)EPA B MECH.NTL RPS QUAD CITIES STATION UNITS 1&2 REACTOR PROTECTIONSYSTEMPOWERSUPPLY

.S.(2AB-2)1 AB-1 EPA (2AB-1)1AB-3 REG (2AB-3)RESERVE INSTR.&RPS TRANSFORMER RESERVE INSTRUMENT

&RPSBUS________120/240-15.2 (25-2)FIGURE 7.2-1 TRIP SYSTEM TRIP SYSTEM LOGIC TYPICAL PROTECTION SYSTEM (CONTROL AND INSTRUMENTATION PORTIONS)ELECTRICAL RELAY/SOLENOID MECHANICAL CONNECTION ELECTRICAL CONNECTION ELECTRICAL CONTACTS (SHOWN CLOSED)VALVE~~~1 CHANNEL t~HANNEL PROTECTIVE 0 ru~rl C z (f2 ru 0~r1 0 z H-4 z H H H-4 0 z'C 1-CI,~t)H 0 z DEVICE SENSOR~~_-1 TRIP SIGNAL'-TRIP SIGNAL TRIP SIGNAL LEGEND:

BEgf!lR pROifCTION sysifll M/G SfI A "88V, 6fJCY, MCC A *525 ' *42:!:----I I I I I t

' ' ' ' --

=1 . ------; .. -!:------: I _J CR ;t,CR; ----2 _ l_______ ... "'-""" ----------------------. I'""'"-..

g "----------------------------1-* ...

_:_:._ __ _____, I I )"'I } L-**---****--

1 )

Lt---T---1 _J L --t--.. .,, ; --------NOlEfR I . I I Cl1 §i. ' ".........

"*------, i!I

_ -----

  • * : * "
  • _ _c ___ , ------*-* : *----, ' I -, I . EXHAUST ............

,!! ____ -::.._ "" : ! .l , '"' I : * -----_, . : : '---.-':Ill ' * .. ... ,, :': . : """-" ! -: .. J . *----'. . .. " .. """' .. ::

--4-4-: . **-" . '""" .... ,

_:_: ___ -----I I IC. ------.. .. :: ..

....., --------= ===------j-.,-:; .. """. "T-T ::.,. :* :: .. .. --------_:-_________ " .. .. ' ' .... ' '" ,1, '" '" ....., ......, """"' __ --*-_ u -:.'-'-..:;'r--" '------, , ] ' I


=----_-_-_-_

.:: =::-::.*-rrr--::--" """' : . h<Il>y c:. I :

_,L__,J_ -'-'--*'----:::_-_________

_l_tL ..** ___

I I PILOT I I I '"'"' JL_J, ____ _:-.:::-c:_________

'l "-----:m11 1 * '1-I NE l Dill=====

--:; ---i _______ ----=-------=----

--_-..:.: = =-:*..: f tT

,' , ----*-., -----, -D G3"'h\. ___ --' L_ ;.L.: .::::: :::: :.t:.: --------I I. -

I Mmw!IC'L:

I I 4----------n I ____ ---.:X:Jt ___ --'!![ L ' INDICATORJ 1 J D"f.b-____ ]

.: =-i-===

=== -T--:_-_::.., . __ ------I ISO.


..J L.

-

ARI EXIWJST

  • tJ* I SCRAM VN.VE PILOT AIR Hf.ADER I I I OlHER CONTROL ROD ORM SCRAM VN.VES 1YP. G1 > --ff----*lDNEl1lRAI.

I BUS A I I Jui ROD SCRMI J.TI 1rn SWITCH I I I 4----i.TO NEl1lRAl BUS B l I PARTOFHYD.MOO.

I VENT : .y

Gl SIDE2 L__ __ _____ _J t

SCRAM HEADER lttRU G4 SIDE 1 .t 2 TO Cl.£AN RADIOACTIVE WAS1E TO Cl.£AN RADIOACTIVE WASTE tii :c OUAO CffiES STATION UNITS 1 &: 2 TYPICAL LOGIC ARRANGEMENT FIGURE 7.2-3 REVISION 8, OCTOBER 2005 DISCHARGE VOL HIGH WATER LEVEL TRIP PRIMARY CONTAINMENT HIGH PRESS TRIP CONDENSER LOW VACUUM REACTOR VESSEL HIGH PRESS TRIP A1 TRIP CHANNELS (AUTO) NOTE 4 MAIN ST LINE ISOLATION VALVE CLOSURE LOCAL TRIP LOGIC A1 f"ROM BUS "K DISCHARGE VOLUME HIGH WATER LEVEL C11-N00A LOCAL f TURBINE STOP VALVE CLOSURE CONDENSER L{ TURBINE CONTROL VALVE FAST CLOSURE REACTOR VESEL LOW LEVEL LOCAL LOCAL LOW VACUUM LOCAL \ f \ \ TEST SWITCH {OPEN TO TEST) CR PERMISSIVE WHEN ENERGIZED PERMISSIVE WHEN ENERGIZED I A1 TRIP LOGIC (AUTO) PERMISSIVE WHEN ) DISCH VOL HIGH LEVEL TRIP BYPASSED \ CR I ) ( MODE SWITCH PERMISSIVE IN "REFUEL" AND °SHUTDOWN" I \ CR I ) 0 MODE SWITCH PERMISSIVE IN "REFUEL" "sTARTUP" & "SHUTDOWll' I \ CR I B21 N024A LOCAL PERMISSIVE ) MAIN STEAM LINE HIGH RADIATION D11 K603A CR NEUTRON MONITORING SYSTEM TRIP PRIMARY CONTAINMENT HIGH PRESSURE N002A LOCAL REACTOR VESSEL HIGH PRESSURE B21 N023A LOCAL CR A MAIN ST LINE ISOLATION VALVE CLOSURE TRIP REACTOR VESSEL LOW WATER LEVEL TRIP f CR A,.;t PRIMARY CONT PRESS TRIP APPROACH ALARM MAIN STEAM LINE HIGH RAD TRIP CR DISCHARGE VOL HIGH LEVEL TRIP BYPASSED NEUTRON MONITORING SYSTEM TRIP f CR A MODE SWITCH "SHUTDOWN" TRIP BYPASSED TURBINE STOP VALVE CLOSURE TRIP CR A MAIN ST LINE VALVE TRIP BYPASSED TURBINE CONTROL VALVE FAST CLOSURE TRIP CR TRIP SYSTEM "8" TRIPPED TRIP SYSTEM "K TRIPPED CR A TURBINE STOP VALVE & CONTROL VALVE TRIPS BYPASSED WHEN ENERGIZED

\ I PERMISSIVE ) WHEN \ ENERGIZED

\ I PERMISSIVE ) ( PERMISSIVE IF :) TURBINE 1 ST STAGE WHEN PRESSURE 45% \ ENERGIZED OF RATED \ I \ N003A I LOCAL I PERMISSIVE ) WHEN \ ENERGIZED

\ I PERMISSIVE ) WHEN ENERGIZED I I

\

  • PERMISSIVE ) WHEN ENERGIZED I I \\-----....---f

\

  • PERMISSIVE ) WHEN ENERGIZED. \ \ I I
  • I I I I I TRIP LOGIC A2 TRIP LOGIC A3 A3 TRIP LOGIC (MANUAL) J_ ( NEUTRON ;) MONITORING SYSTEM INTERLOCK INITIAL FUEL LOADING ONLY -A 1 TRIP ACTUATOR RESET SWITCH NOTE 3 CR RESET SWITCH NOTE 3 CR '"----1 SEAL "2. TRIP ACTUATOR SAME AS Al TRIP ACTUATOR I !Nit SCRAM CONTACTOR I ll--s""cR"'A"'M"'"""c"=o'"'NT"'A"=c"'TO"'R:--11

.................................

...... I : I : ! i' '111 ! ii I I : : 1 I I I I I I I I I I : I : : : : : : f : I : : : : 111::: I I t : : ; t : : : f : t 1/3 TRIP LOGIC -SCRAM PILOT VALVES A -GROUP Gl t I I t I I I I 1 /3 TRIP LOGIC -SCRAM PILOT VALVES A GROUP G2 1/3 TRIP LOGIC -SCRAM PILOT VALVES A t I I I I I I GROUP G3 I I I I 1/3 TRIP LOGIC -SCRAM PILOT VALVES A -GROUP G4 \ I MODE SWITCH IN SHUTDOWN CR MANUAL TRIP SWITCH CR A3 TRIP ACTUATOR SAME AS Al TRIP ACTUATOR I::::: 11111: I t i I i I I I I I I I : : I i I I I I t I I I I : t I t I 1/3 CLOSE LOGIC BACK UP SCRAM VALVE A * ( AUTOMATIC BYPASS AFTER 10 SEC \ I ' c DC SUPPLY (+) ) I I I I I I t t t rl __ ...!.., __ _!. __

SUPPLY(+)

f ZA f M QUAD CITIES STATION UNITS 1 TYPICAL LOGIC ARRANGEMENT FIGURE 7.2-4 REVISION 7, JANUARY 2003 & 2

  • * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * ... * * -TO A2 TRIP CHANNELS (SH.2)

...... TO A1 TRIP CHANNELS (SH.2) . --------------


  • * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * ****** .........

TO A1 TRIP CHANNELS-**

              • --------------

TO A2 TRIP CHANNELS . . r-----------, . 1------.., I : I I
  • I I : I I
  • I I : I I
  • I I : I I
I PRIMARY CONTAINMENT HIGH PRESSURE PRIMARY CONTAINMENT HIGH PRESSURE I I TO ANNUNCIATOR RAM RAM IRM TYP LPRM RAM RAM SRM APRM DISCHARGE VOLUMES HIGH WATER LEVEL VENT NEUTRON MONITORING SYSTEM I I I r-----------------------I ......................... . I
  • J I: L SH I: I. I: I . I: I. I: I. I: I .
  • 1 DISCHARGE DISCHARGE 1 : : I HEADER HEADER 1 * . I I : : I I . I . I I: I : I DRAIN I . I . I I: I :

4 LI...-

I 81 TRIP ACTUATOR SAME AS A1 TRIP ACTUATOR 8 Oci 1-z ::>z .::><( :c (/)(_) _J Wo_ z....., za::: <( 1-:c u:;;: 0... (/) Q2<t 1-w N::2 CD<( (/) .___,. 82 TRIP ACTUATOR SAME AS A1 TRIP ACTUATOR BUS HBU 83 TRIP ACTUATOR SAME AS A1 TRIP ACTUATOR L_ -----------------------------------------------------------------------DC SUPPLY (-) BACK-UP SCRAM VALVE-A TRIP & CLOSE LOGIC SAME AS FOR TRIP SYSTEM "A.' EXCEPT AS SHOWN QUAD CITIES STATION UNITS 1 & 2 DC SUPPLY ( -) TYPICAL LOGIC ARRANGEMENT BACK-UP SCRAM VALVE-8 6B 58 48 38 2B 18 FIGURE 7.2-5 REVISION 7, JANUARY 2003 GROUP 1 WJN STEAM LINE MODE SWITCH MAIN STEAM LINE REACTOR WATER MANUAL TUNNEL WJN STEAM LINE (lYP. EACH PRESSURE -IN TEMPERATURE FLOW LEVEL LOW "!Mt' HIGH LOW-LOW VALVE) I I I ........ CLOSE MAIN STEAM LINE ISOLATION VALVES ..__ CLOSE MAIN STEAM DRAIN ISOLATION VALVES .__ CLOSE RECIRCULATION LOOP SAMPLE ISOLATION VALVES GROUP 2 DRYWEl.l.

REACTOR WATER DRYWEU. MANUAL HIGH LEVEL PRESSURE (lYP. EACH RADIATION LOW HIGH VALVE) I -CLOSE DRYWEU. TORUS VE.NT, PURGE. AND SUMP ISOLATION VALVES -PROVIDE TIP WITHDRAWAL COMMAND i--RHR TO RAOWASTE, AND RHR SHUTDOWN COOL.ING .__OXYGEN ANALyzER GROUP 3 REACTOR WATER MANUAL

  • SBLC RWCU AREA MAIN STEAM LINE TUNNEL l.E\IEL (TYP. EACH JNffiATION TEMPERATURE TEMPERATURE LOW VALVE) SW INTERLOCK HIGH HIGH I I I -* -CLOSE CLEANUP SYSTEM ISOLATION VALVES GROUP 4 HPCI HPCI TURBINE REACTOR DRYWEU. REACTOR STEAM FLOW SPACE PRESSURE PRESSURE PRESSURE TEMPERATURE HIGH HIGH LOW HIGH LOW I I I I CLOSE HPCI TURBINE CLOSE HPCI STEAM SUPPLY VALVES EXHAUST VACUUM BREAKER VALVES GROUP 5 RCIC RCIC TURBINE REACTOR STEAM FLOW SPACE PRESSURE TEMPERATURE HIGH HIGH LOW I I CLOSE RCIC ISOLATION VALVES RHR SHUTDOWN REACTOR REACTOR WATER -CLOSE RHR COOLING ISOLATION PRESSURE LEVEL LOW COOLING INJE ON VALVES HIGH (GROUP 2) (WHEN IN SHUTDOWN COOLING MOOE) I I I I QUAD cmES STATION UNITS , & 2 CLOSE RHR SHUTDOWN BLOCK DIAGRAM COOLING SUCTION VALVES PRIMARY CONTAINMENT ISOLATION
  • ALSO ISOLATES ON NON-REGEN HEAT EXCHANGER OUTLET TEMPERATURE HIGH. SBLC INTERLOCK AND NON-REGEN FIGURE 7.3-1 OUTLET TEMPERATURE TRIPS ARE NOT CONSIDERED PRIMARY CONTAINMENT ISOLATION SIGNALS. REVISION 10, OCTOBER 2009 S.S a,AVERAGETHERMAL NEUTRON FLUX (nv)03 CO-~cD (4.I I FULLY INSERTED RETRACTI0 N-~F 1-F STARTUP I HEATING POWER.IIIIIIII R03C, PERCENT POWER C,,:~~>-U~~~,~~I~-U~~D-U m~,>-4 c~
  • ------------**
  • --------------------------.--

.. --..

IRM SRM lRM ' ROD BLDCK SCRAM, ROD BLOCK & >.LARM &: >.LARM SOURCE RANGE lHTERMEOrArE RANGE TYPICAL OF 164 LPRMs SWITCH i------. MATRIX LPRM TRIP AUXILIARIES Al.ARMS POWER RANGE APRM TRIP AUX[UARIES ROD BLOCK SCRAM, ROD BLOCK &: >.LARM RECIRCUt.AHOH FLOW FLOW UNIT CITIES STATION UNITS 1 & 2 BLOCK DIAGRAM INSTRUMENTAT[ON SYSTEM 7.6-2

.~}I+/-I+I+I+k+I+I+I+I+I+~~

J+i+i+i+i+i+i+i+i+i+i+i+i~

~+I+I+I+I+I+I+I~I+I+I+I+I+I+i+

+I+I+I+I+I+I+I~I+I+I+I+i+I+i+

+I+/-I+/-I-F~+I+I+I+I+I+i+I+k-F-I+/-I+

+I+I+I+I+I+I+f+I+I+I+I+I+

+I+/-I~I+I~I~I+I+I+I+I+

+I+l+I+I+I+I+f+I+

+I+I~I+I+X~SOURCERANGEMONITOR DETECTORS A-NEUTRON-EMITTINGSOURCES (NO LONGER INSTALLED)

.QUAD CITIES STATION UNITS 1&2 SRM-DETECTOR AND SOURCE LOCATIONSFIGURE7.6-3

+1+1+1+1++I+I~I~I+I+I~I+I+

-x-$+i+I+I+I+I+I+I+I+I+

+I+I+I+I~I~I+I~I~I+I+I+I+

+I+I+I+I~I~I+I+I~I+I+I+I+

+I+I+I+I+I~I~I+I+I~I+I+I+I+I+

+I+I+I+I+I+I+I+I+I+I+I+I+I+I+

+I+I+I+I+I+I~I+I+I~I+I+I+I+I+

+I+I+I+I+I+I+I+I+I+I+I+I+I+I+

+I+I+'+I+I~I~I+I+I~I+I+I+I+I+

+I+I+I+I+I+I+I+I+I+I+I+l+

+I+I+I+I+I+I+I+I+I+I+I+I

-x-+I+I+I+I+I+I+I+I+I+I+

+I+I~I~I+l+I+I+I+

+1+1+1+1+*-INTERMEDIATE RANGE MONITORING CHANNELS 4 REACTOR PROTECTION SYSTEM LOGIC CHANNEL A*-INTERMEDIATE RANGE MONITORING CHANNELS 4 REACTOR PROTECTION SYSTEM LOGIC CHANNEL B QUADCITIESSTATION UNITS 1&2 IRM-DETECTOR LOCATIONS S SFIGURE7.6-4

+1+1+1+1+1+1+1+1+1+1.+1+1+1-+1+1-______+1+1-+/-~+1+1+/-1-H:~g~--5.-:-;~:~

-+-+1+1+1+1+1+1 I~+?~+~+1+1+1+1+1+1+1+1+1+1+1+lIIfj~I+)+)~+1+1 QUAD CITIES STATiON IJNITSI&2 IRM-RESPONSE TO ROD WITHDRAWAL ERROR FIGURE 7.6-5 i~fli 2 flFC.1993-+k~i+1+1+1+1+/-~j+~+I+1+1+1+1+~1-.+1+1++1+1+1+1++M~1+/-YW11HORAWN cONTROl.11005 CONOINON 1)REACTO11JIJST SUBCR~UCAL 2~O$E~RM RYPASSE)IN EACH REACTOR PROTECfl0t4 SYSTEM LO(~C CHANNEL OUT OcSEOUENCE FULLY WITHDRAWN 1RM BYPASSED COREAVERAGEFLUX

.100.0 10.0>(110 T 0.1 1 001 C)C-4 0,001_____________________________________________________________________________________468 10 12 16 DISTANCE (feet)S QUAD CITIES STATION UNITS 1&2 IRM-POWER DISTRIBUTION DURING ROD WITHDRAWAL ERROR FIGURE 7~6-6

+.-+1++1+-.----+1++1+-.-+1+

+1+-.--+1+

+1+-.-+1+++1+14-1+1+1+1+1+-.-1+1+1+1+-.-1+1+1+1+-.-1+1+1+1+-.-1+1+

1+1+-.-+1++1+1-.-I+1+1 1+1+1 1+1+1 1+1+1-.-1+1+1 1+1+1-.-1+1+1 1+1+1-.-I+1+1 I+1+1-.-I+1+1 1+1+1 I+1+1+1+1+1

+1+1+1+1.S.+1+1-.-+1+1+1+1-.----+1+1+1+1-.-+1+1

+1+1-.---+1+1+1+1-.-+1+1+1+1-.--+1+1+1+1-.-+1+1

+1+1+-.+1++1+-.-+1++1++1++1++1++1+-.-+1+

+1+-.-+1+

+1+-.-+1++++1+-.-+1++1++1+

+1++1++1+-.-+1+

+1+-.-+1+

+1+-.+++-.+1+1-.+1+1-I+1+

+++++

+NOTE: EACH LOCATION REPRESENTS ASTRINGOF FOUR DETECTORS SPACED 3FEETAPART.

QUADCITIESSTATION UNITS 1&2 LPRM-DETECTORLOCATIONSFIGURE7.6-7

~oooooc D00000C D00000c D00000C 000000C 000000C D00000C J00000t D00000c)00000C)00000c)00000c)00000C)00000C 000000 TUBE CHAMBER~ooo~IP CALIBRATION TUBE DO~OO)00000C)00000C D00000C)00000C)00000C D00000C D00000C QUAD CITIES STATION UNITS 1&2 LPRM~LOCAL DETECTOR LOCATIONS CONTROL ROD BLADES N I...FIGURE7.6-8

.I QUADRANT 1 1 QUADRANT 2~I+/-I+/-I+l+I+I+I--I+I+I+I+I+I+f

+I+I+I+I+I+I+I--l+I+I+l+i+l+i+

+I+'+'+I+I+I+I+I+I+'+I+I+I+I+

--.--.--.-4-.--.--.--.-I II 1+111111111

~'21~1 12 111 112111 112+0-0-0-0-0-0 ~+l+I+I+I+I+I+l+/-I+l3+l+I3+I+I3+I~

--.--.--.---.-0-*-0-.-0-._

~~+I+I+l+l+l+I-~l+l3+I+I3+I+I3

-.--.--.---.-0---.-0-.-0

+I+I+I+I+l+I~~I2+l1+I2+ll+l2+I1

--0-0-0-0-0-0 H+I+I+I+I+l-1+13+1+13+1

~-.--.---.-0-.-0-.

~+l+I+I+l-12+11+12 II----0-0-0/"/QUADRANT 3 QUADRANT 4ILLUSTRATIONOF MONITORINGCOVERAGEASSUMINGQUADRANT SYMMETRIC OPERATION 02EQUIVALENTDATAROTATEDFROMQUADRANT 1 03EQUIVALENTDATAROTATEDFROMQUADRANT2 0 EQUIVALENT DATAROTATEDFROMQUADRANT3UNMONITOREDPERIPHERALASSEMBLIES

.QUAD CITIES STATIONUNITSI&2 LPRM-QUADRANT SYMMETRYFIGURE7.6-9 I+I+I+i~+I+I+

~+Ij~fI+L+I+/-

8~+.-+I+I+T+I+I+I+/-1D+/-+/-~1 I+T+1+13+1+/-I+o~a+1+/-18+I I+/-~2+I+I2+I+~2+I+/-L+

+8Ic+l+I+I+1A+I+I+

I+~d-I+BIC+l+~+1+1+I I I I+/-1l+/-Ij~2O+I+/-I 2+I+/-D~2+/-J+L+I+/-B~24++/-~+/-I+T+/-l+~+/-I+/-AI8+/-I+I+I+/-T+/-

I I I I+~~I+/-I1+/-'+~4+/-l+/-L+/-l+/-fI+/-I~++DIA+l+I+I+i0+I+I+I+01A+I+I+

I I+15+I+/-0I7+/-I+/-Ia+/-I+/-BI9+/-I+/-I1o+I+/-DII+

-.-~+/-I+/-I 1+/-I+~I 2+I+I 3+I+j 4+I+L+/--.-S*+I+l+1A+l+I+l+ic+l+

~I~5-LPRMSTRINGS PROVIDING INPUT TO CHANNELS 1, 2,4 C 2 UPPER RIGHT NUMBER LPRM STRING IDENTIFICATION S-UPPER LEFT LETTER LPRM CHAMBER USED AS INPUT FOR CHANNEL 1 LOWER LEFT LETTER LPRM CHAMBER USED ASINPUTFOR CHANNEL 4 A LOWER RIGHT LETTER LPRM CHAMBER USED ASINPUTFOR CHANNEL 2QUADCITIESSTATIONUNITSI&2 APRM~LPRM ASSIGNMENTS, CHANNELS 1,2.4.I+

+++

++12++1++DI18+-5-+~iB++

+SFIGURE7.6-10

+12+-.---1+1+I+/-D~+I+~F-1+112+-.-1+1++B16+-,-+1+~-.--+139+1-.-1+/-1+1~+Al4+1-.-+T+I 1+I+1+I+i~+I+I+l+~+I+l+

I I I I~++BIC++/-120+-.-+1+l+L8+I 1+/-1+1 I+B~2+I I+/-i~I+~2+I+/-L+I+A~3+

+DV+I+I+l+~f+/-I2+I+/-D~+I+/-L4+

+I+I~I~I+l+

I I I I'b~+/-+DIA++/-17+/--.-+1++AI1+-.-+BIC+'+/-~+/-'1+1+1l+/-D18+/-I-,-I+T+I1+12+/-1-.-1+1+1+AL5+l+I1+I+CL++~IC+I+/-I+/-I+/-DlA++/-I9+I+BI10+I+I11+

-.-+/-l+/-IPID+/-l+/-I+/-

+C13+1+14+I-.+/-DIA+I+I+I

~'+I~-LPRMSTRINGS PROVIDING INPUT TO CHANNELS 3,5,6 2 UPPER RIGHT NUMBER LPRMSTRINGIDENTIFICATION C.-UPPERLEFTLETTER LPRM CHAMBER USED ASINPUTFOR CHANNEL 5 LOWERLEFTLETTER LPRM CHAMBER USED AS INPUT FOR CHANNEL 3 D A LOWER RIGHT LETTER LPRM CHAMBER USED ASINPUTFOR CHANNEL 6++32-.+1++T++118+-.-+1+++QUADCITIES STATION UNITS 1&2 APRM-LPRMASSIGNMENTS,CHANNELS 3,5,6++++

++...+1+1 FIGURE 7.6-11 00%100%110%REVISION 7 JANUARY 2003 QUAD CITIES STATION UNITS I&2 ILLUSTRATIVE APRM SCRAM AND ROD BLOCK TRIPS VS.RECIRCULATION FLOW 130%120%110%100%00%80%10%60%

50%

40%30%20%C I Scram Trip (AL)10%-R-Rod Block Trip (AL)Core Power vs Flow Response (Typical)0%20%30%40%50%60%10%80%

RECIRCULATION FLOW (%of rated)FIGURE 7.6-12 0 20 40 60 80 TOTAL FLOW[%RATED]100QUADCITIES STATION UNITS 1&2APRMRESPONSEDURINGFLOW-INDUCED POWER LEVEL MANEUVERING S S S w a: w a: a: a: w C a-LU a: C 100 80 60 40 20 0 100 80 60 40 20 0FIGURE7.6-13

-J I-2 CD 2 w~40 0~02040 60 80 100 CORE POWER[%RATED]QUAD CITIES STATION UNITS 1&2 APRM RESPONSE DURING CONTROL ROD4NDUCED POWER LEVEL MANEUVERING 100 80.S.100 80 60 40 20 0 20 0FIGURE7.6-14 NOTE: ASSIGNMENT IS AUTOMATICALLY INIATIATED UPON ROD SELECTION+1++1+1+1+-.--.-1+1+1+1+1+1+1+1+-.--.-+1+1+1+1+1+

~I~I+I+I+I+

-.--.---.-+1+1+1+1+1+

~I~I+I+I+I+-.--.--0-+1+1+1+1+1+

~I+I+I+I+~--0---0-+1+1+1+1++1+1-0-+1+1+r1+/-~-'*)---+/-T+/-I-id*I+LT+/-t+1+1-.-+1+1+1+1-0-+1+1+1+-0-+1++/-1+/-+1*-_~0)L_+1+-0-+1++1+-0-+1++/-1+1+

+1+-.-+1++1+-0-+1+~T+/-I-RBM AUTOMATICALLY BYPASSED (READING ZERO)0-TYPICAL RODYIELDINGTWO LPRM STRINGS AS INPUTS 0-TYPICAL ROD YIELDING THREE LPRM STRINGSASINPUTS*-TYPICAL ROD YIELDING FOUR LPRM STRINGSASINPUTSQUADCITIES STATION UNITS 1&2 RBM-LPRM INPUT ASSIGNMENT 900++.S.FIGURE7.6-15 Revision 14, October 2017

Figure 7.6.16 has been deleted.

,---, ,---....., I I l I I ANNUNC I I ANNUNC I WINDOWS I I SER 1 PHL 9111(2)-5 j lpNI. 9i1(2)-J+I L---......1 L ___ ..J ,---, I I r---------r---------r---------,----------,----

-1 .a. . -*-*r l ----I I ljLPli l I 21 I I I ... I I I l I ..... NUNC I I AHNUNC IWIHCH:l\WSI I SER(:: I ---1 ;;o r'l < (Jl -ri r-. r--1 OC) zc u::i :;;()

  • fTI 0 Cl'-.J --1 . om

'-.J f'0 <'.Sl <'.Sl -.,J I PNL !IQ 1 (2)-5 l L---......1 0 c Orn }> ]r cO s:g z ..... o .. -.-1 :A: Lf) ;:! (/) Co fTl CIJr**i IJ) [/) }> -< C) (/) :;;() ---1 }> 2s: ,___, 0 z A -BARRIER (BEIWEEN BAYS} I I I l I I ------------i I I I I I I I I

.... I I I I I I I I : I l I .. I I I I I I LUI L I I I ' -------L I I . I -

BAY 1 -----------L I r,;;.;.e

... P 1 BAY 2 ---------' -BA y 3 ---------I flO XMTR I BA y 4 ------XMTR I BAY 5 -

ROD I I RODPOllTIOll l I D!SCllUGE I UlNIUIZER INrtll!WA

!ION VOLUME IN LIMlllNG mmi ta*mllON MALFUNCTlllH JATER LEVEL fllHER f\01 CON* VERIER UPSCALE 1uorrnrnvE DR COMPARAlaR ALARM l ROOWIVEMENT SWITCH llALFUHCTl!lll DllRING'flrKOUr ROO SfLECT fl,)tU\tlTCH IN '"Off" NlSl\IC!ll SCRAM DISOURG£ VOl..UMEHIGH' "ATER LEVEL' SCRAM TRIP BYPASSED --.--AfAV UHC*U nv1 I 1ucre* ) I EITllU 100 MOOE SWJTCH lt.DtK lilDNJl'OI IN "ltUN" Ul'SCAtf () R 1*0P-P8SlllON lftATIYI I IEFUELllODt ONUooGUt PEllMSSIVI lltflUOCIC . llACTDlllOOE S"M'TtHtM "SHUTOOWll" l 01 "1lEfUfL"" ' PDllllOH REFUU avu 11.AtTOI CO Rf ftEACTU llCOE!WlfCH IH"'ITl\ATUr POSITION \ MFUU PU.TFOH OVfll"EAtTCll CDftt REACTOR Moot $WllCHIN 'SHIJTDQWlt PO$ff!OH l 014,,ll aOTfUllUP 1

'IOOi Iv.ITCH IN "REFUEL' OR ".!TAl!TUl'tliOTSTANDl't", POSl111Jll 1 AMY !RM RANtlE SWITtHl"EITllU OF'l'l'O Lll'IESTIWl6El ANY IRMUl'$CAlE OR IHOPERAllYE OR DETECTOR HQ I f!Jll W IN!lATED AllY 51!11 UP:ICAL£ ANY !RV 00Yk!CAL£ IN bu! tQWl!l CR INOPERATIVE -i-i. J. QUAD CITIES STATION UNITS 1 & 2 CONDITIONS WHICH PREVENT CONTROL ROD WITHDRAWAL FIGURE 7.7-1 REVISION 8, OCTOBER 2005 CONTROL ROD (/) _J POSITION z 1--j CONTROL ROD 2 ID n:: w LOW POWER f-SETPOINT f-:J o_ BYPASS f-:J 0 RMC ""'-DATA f-:J ROD BLOCK o_ z RELAYS 1--j DAS C DAS D ETHERNET ETHERNET ETHERNET PSS SEQUENCE TOUCH SCREEN ETHERNET 1------PPC-A PPC-B QUAD CITIES STATION UNITS 1 &: 2 BLOCK DIAGRAM ROD WORTH MINIMIZER FIGURE 7.7:__2A REVISION 11 I OCTOBER 2011

PRES S U RE SETP OfN T MCF L S ETP OJN T BWR SPEED E RROR SP EE D/L OAD OE M A N O TR I P CLOS E BI A S T C V R EFERENCE M !----------<

I TCV POS I T I O N-------. CONTRO L 1-----------------------N P RESS UR E ER ROR ADJUSTABLE SPEED DRIVE TCV L I M l TE R t-------------

(-( REA C T OR C O O L-D OWN FU NCT ION PR ESSURE DEMA ND ,_, SUM BPV R E F EREN CE RO T O R W A RM REF E RENCE TC V R EFERENCE 1-------B PV I B P V SPEED SIGNAL POS I TION BPV 2 CONTROL ETC PUMP (TYPICAL OF TWO) MOTOR SPEED (INDICATION ONLY) RRCS -----< § D RAISE tJ D LOWER INDIVIDUAL SPEED CONTROL STATIONS (MANUAL) MASTER SPEED CONTROL STATION (MANUAL) QUAD CITIES STATION UNIT 1 & 2 REACTOR PRESSURE, TURBINE SPEED, AND RECIRCULATION FLOW CONTROL SYSTEMS FIGURE 7.7-38 R E VI SIO N 11, OC TOBER 2011