chapter3_1

3 Maintenance an Efficiency Control of Thermal Power Plants 3.1 Maintenance management of aged thermal power plants Introduction The time (nominal life) when degradation in function and performance of the major equipment constituting a thermal power plant appears remarkably and recovery of function and performance requires costs overstepping the bounds of maintenance for ordinary operation is about 30 to 40 years. For small- to medium-capacity machines constructed from the 30s to the 40s of the Showa period, almost 30 years have elapsed since they begun commercial operation, and, with respect to large-capacity machines constructed and brought into operation from the middle of the 40s, 20 years or more have elapsed, whereby aging of thermal power plants as a whole is proceeding. On the other hand, although we are pushing ahead with the development of new sources responding to the growth of electricity demand (maximum electricity demand), we have problems such as difficulty of site selection, more distant locations, longer periods for development, etc. For this reason, renovating and renewing these aged thermal power plants efficiently and having them show continuous activity as valuable leading sources in areas of high demand constitute important issues from the viewpoint of securing supply capacity and cost reduction. 3.1.1 How to carry out maintenance of aged thermal power plants (increasing longevity) How to carry out maintenance of aged thermal power plants leads to studies and judgments being carried out from comprehensive viewpoints such as a source program into the future including nuclear, hydro, etc., positioning and role in terms of a power generation program, outlook regarding the renovation costs required to upkeep function and soundness, trends in fuel costs according to power generation efficiency and fuel class, outlook for the introduction of alternative sources, trends in technological development, etc. to define the direction. However, the current course of action is broadly divided into the following two items (Fig. 3.1.1):

(1) Maintaining operation through the renewal of deteriorated plant equipment (increasing longevity) This is a direct extension of matters that we have implemented conventionally as measures against age deterioration, and plans continuance of operation (increasing longevity) for about 60 years (within the range where substantial renovation of civil and construction equipment such as piles and foundations is not required).

(2) Maintaining operation through repowering, replacement, etc. (increasing longevity) (3) Planning the continuance of operation (increasing longevity) through repowering, replacement, etc.

allowing an increase in the scale of output, improvement of power generation efficiency, improvement of operational function, environmental betterment, etc. is to be carried out in conjunction with the renewal of deteriorated plants (The target is mainly small- to medium-capacity plants).

3.1.2 Maintaining operation through the renewal of deteriorated plant equipment (increasing

longevity) The actual service life of plant equipment differs from its nominal design life and it is significantly dependent on good or poor operation and maintenance. With respect to thermal power plants, for the purpose of keeping their function and performance at an established level, the scope of inspection, method and frequency are defined on an equipment-by-equipment basis as a standard, and patrols, routine checks, periodic inspections, and service and maintenance (repair, replacement, etc.) are performed according to such standard. Further, as to aged thermal power plants, in addition to these items, precise inspection for the pressure part of the boiler, turbine rotor, casing, major valves and rotor power generator, remaining life assessment, renewal of deteriorated equipment and portions, addition of equipment function to respond to demand-supply operation, and strengthening and enhancement of durability are planned simultaneously.

Developing program Power generation

[Background]

Long-term operation program

increasing longevity (prolonging life)

[Destination][Needs] [Measures to increase longevity]

[Measures to increase longevity]

Repowering Replacement (addition)

Maintaining operation through the renewal of deteriorated equipment

Stable supply Cost reduction

[Background]

Fig. 3.1.1: How to carry out the maintenance of aged thermal power plant (increasing longevity)

142

Inspection & repair program

[Example] • Monitoring of life consumption of high-

temperature thick part of boiler and turbine• Diagnosis of boiler combustion • Monitoring of vibration of large rotating

machine • Trip test of safety device • Routine replacement of auxiliary machine • Vibration of auxiliary rotating machine • Operation status of auxiliary machine • Opening of control valve

Precise inspection, diagnosis of plant

Improvement of function

Track record of operation

Enhancement of operation control and supervisory function Reduction of load on equipment Early detection of problems

Securing of soundness of plant equipment Upkeep of function and reliability level Securing of economics

[Example] • High-temperature pressure-

resistant part of boiler and turbine

• High-speed rotor of turbine • Insulation of generator

[Example] • Renewal and renovation of

deteriorated equipment and parts [Example] • Improvement of control

performance, strengthening of supervisory function

• Automation, improvement of control performance

Experience of and information on maintenance for aged thermal power plant (Problems, renewal of equipment, renovation) Renovation technology (modernization technology) Inspection technology, remaining life assessment technology Increasing longevity program administration support system

Operation

Periodic inspection, service, maintenance

Daily simplified maintenance

Patrol

Routine inspection/test

Monitoring of operation status

Operation control

Improvement of proof stress

Renovation program

Maintenance

Long-term operation program

Maintenance of aged thermal power plant

Fig. 3.1.2: Maintenance of aged thermal power plant (continuance of operation through renewal of aged equipment (increasing longevity))

For this reason, even if the plant reaches its nominal design life, there is still considerable practical operation life of a major plant, and, as to the reliability of the entire plant as well, it is understood from long-term experience of operation and maintenance that continuance of operation (increasing longevity) may be possible at relatively low cost. Basically, although this is a direct extension of matters that we have implemented conventionally as measures against aged deterioration, for the purpose of maintaining stable operation while securing the economics, it is necessary to push ahead with understanding and preserving (recovering from deterioration) the function, performance and soundness of plant equipment more efficiently than conventionally, and it is requested that plant diagnosis (deterioration diagnosis, remaining life assessment) technology, renovation technology, trouble information, etc. be used to further push ahead with critical classified renovation according to the operation period and operational method. [Fig. 3.1.2] From here, we introduce developments in operation and maintenance of aged thermal power plants in the past, renewal conditions of aged machines, devices and portions, inspection technology and remaining life assessment technology, examples of large renovation for maintaining operation (increasing longevity), acceleration of the time necessary for periodic inspection that tends to become longer with aging, a work program support system for critical classified renovation, status of efforts to cope with the increase in longevity of aged thermal power plants in U.S.A., etc. 3.1.2.1 Operation & maintenance status of aged thermal power plants (from the 50s to the early 60s of

the Showa period) Aged thermal power plants were originally designed to operate continuously (operation to cover the base load).

1992 (September) 50 Showa period = 1975

10 th

ousa

nd k

W

1970 (September)

1960 (January 36)

Fig. 3.1.2.1-1: How electricie

14

Tim
ty is used in a day (example)
3

Fig. 3.1.2.1-2 How electricity is used in a day

(representative example)

Table 3.1.2.1-1: Precise inspection (representative

example) Plant Portion to be

inspected Inspection method

Rotor Visual inspection with bore scope Ultrasonic testing Magnetic particle testing

Moving blade Ultrasonic testing Measurement of lifting amount of stud part

Steam turbine

Casing Structure examination of representative point (macro)

Superheater and reheater tube

Ultrasonic testing (weld point) Tube removal examination from a representative point

Boiler

Drum Main steam pipe Reheat steam pipe

Radiographic test

Ultrasonic testing

Generator Rotor Visual inspection with bore scope Ultrasonic testing Magnetic particle testing

Transformer Main body Oil leak test Dissolved gas analysis

Electric motor

Rotor Liquid penetrant detection test Insulation diagnosis

Equalizing pool type Water reservoir-type hydro Pumped-storage

hydroelectric power

Pumped-storage power

Oil

LNG, LPG and other gases

Coal

Nuclear power

Run-off river-type hydro

(Time)

Base load operation was carried out at the initial stage of construction. However, because of subsequent changes in the demand-supply structure, that is to say, an increase in demand (maximum electricity demand), a widening of the gap in demand-supply between day and night, and an increase of the segment share of nuclear power generation, base load operation handed over its role to nuclear power and large-capacity thermal machines. As a result, the operation pattern has changed to the operation of a middle-sized thermal power plant positively bearing adjustment between demand and supply, i.e., the operation pattern under which load change, reduction of minimum load, frequent start up and shut down, etc. are performed.

Table 3.1.2.1-2 Remaining life diagnosis

Plant Portion to be inspected Diagnosis portion Remaining life diagnosis

technique Boiler tube • Furnace evaporation tube • Superheater tube • Reheater tube

Destruction inspection (Conduct a creep breaking test to evaluate the result by means of the Larson-Miller method.)

Boiler

Boiler header • Furnace evaporation header • Superheater header • Reheater header

Select the portion whose design temperature is 450 or more and the harshest (shortest design life) portion in terms of design out of the target portions shown on the left. (Take the time when cumulative operation time reaches 100,000 hours as a guideline.)

Structural examinations

Axle • High-pressure axle • Medium-pressure axle Casing • High-pressure internal casing • High-pressure external casing • Medium-pressure internal casing • Medium-pressure external casing

Steam turbine

Major valves • MSV • CV • RSV • ICV

Select the portion whose design temperature is 450 or more and the harshest (shortest design life) portion in terms of design out of target portions shown on the left. (Take the time when cumulative operation time reaches 100,000 hours as a guideline.)

Non-destructive inspection Hardness measurement Material degradation measurementMetallic structure test

144

Table 3.1.2.1-3: Examples of improvement in medium-capacity machines [Improvement in durability of plant] • Improvement of suspended superheater of boiler (Fig. 3.1.2.1-3) • Improvement of shape of disk base part of steam turbine rotor (Fig. 3.1.2.1-4) • Reduction in stress of steam turbine casing (improvement of shape) (Fig. 3.1.2.1-5) • Improvement of spray at exhaust chamber of steam turbine (Fig. 3.1.2.1-6) [Expansion of operation controllability and supervisory function] Improvement and enhancement of supervisory function for those that have high frequency of control at start up and shut down that have simultaneous operation that have difficulty in adjustment • Making the control of the boiler burner remote or automatic • Making the control of the drain valve and auxiliary machines of the boiler and turbine remote • Automatic start up of turbine from central operating panel (TSC) • Making the oil temperature control on the turbine bearing automated from a central operating panel • Making the injection of feed water and chemicals automatic from a central operating panel • Installation of furnace gas thermometer • Installation of turbine bearing a metal thermometer

For the purposes of securing the reliability and performing strict demand supply adjustment operation such as DSS (daily start up and shut down), etc. of the plant equipment designed originally based on the premise of base load operation given that aging progresses, we have basically planned : • An operation pattern that will contribute to demand-supply adjustment sufficiently and where start-up and shut-down loss is minimized • Securing of strength and allowance of a plant sufficient to cope with thermal stress, repeated stress arising from start up, shut down, load change, etc. and creep damage associated with secular use • Improvement of operability and enrichment of supervisory function so that the operator can cope with the situation within limited time and simultaneous operation • Early detection and handling when there is an abnormal condition in the plant • Establishment of optimum operation pattern through operation testing • Precise inspection and remaining life assessment for plants whose cumulative operation time has exceeded 100,000 hours (Table 3.1.2.1-1, 2) • Improvement of plant durability (Table 3.1.2.1-3, 4) • Improvement of operability and controllability, enhancement of monitoring function (Table 3.1.2.1-3, 4)

Disk

Processed pointCurvature radius

Big

Small

(a)Processing example

Curvature radius

Big

(b) Example of new shape

Fig. 3.1.2.1-3: Improvement of suspended superheater of boiler

Fig. 3.1.2.1-4: Improvement of shape of disk base part of steam turbine rotor

Fig. 3.1.2.1-5: Reduction in stress of steam turbine casing (improvement of the shape of casing)

Fig.3.1.2.1-6: Improvement of spray at exhaust chamber of steam turbine

Casing corner part Steam guide

Processing diagram of corner part

Casing

Diaphragm

Nozzle Packing casing cone

Final-stage blade

3.1.2.2 Renewal status of plant equipment in aged thermal power plants

The Thermal and Nuclear Power Engineering Society (Kanto affiliate) conducted a survey on the renewal status of plant equipment for thermal power plants (commercial thermal, joint thermal and private thermal) whose cumulative operating hours exceeded 100,000 hours in 1991 on a nationwide scale, and the results of such survey have been summarized as the “Report of a fact-finding survey on the renewal of thermal plants that have for a long time (January 1993).”

145

Table 3.1.2.2-1: Examples of improvement in large-capacity machines (Constant-pressure supercritical once-through system)

[Improvement of plant durability] • Improvement of superheater header part of boiler (Fig. 9) • Processing of membrane-end part of boiler (Fig. 10) • Improvement of structure of surrounding wall tie-bar of boiler (Fig. 11) • Improvement of passing-through part at boiler tube wall (Fig. 12) • Improvement of support system for main piping of boiler (Fig. 13) • Reinforcement of start system valve of boiler (Fig. 14) • Reinforcement of feed water control valve of boiler (Fig. 15) [Improvement of controllability and enhancement of supervisory function] Improvement of controllability to plan the improvement of controllability at start up/shut down and when the load changes • Digitization of APC control • Automation of boiler automatic burner • Expansion of automatic start-up control range of turbine • Improvement of controllability on the drain level of feed water heater • Bringing auxiliaries to group control (master) • Addition of life supervisory function for thick pressure-resistant part of boiler • Automation, enhancement of supervisory function and man-machine communication

[Before improvement] [After improvement]

. (a) Bringing header tube nozzle to flexible structure

(b) Corner R processing of header tube nozzle part Fig. 3.1.2.2-1 Improvement of superheater header

part of boiler

R processing of membrane bar stop-end part

Fig. 3.1.2.2-3 Improvement of structure of surrounding wall tie-bar of boiler

Fig. 3.1.2.2-2 Processing of membrane-end part of boiler

A-part Improvement of structure of tube leg at wall passing-through part

(a) Current structure

Fig. 3.1.2.2-4: Improvement of passing-through part at boiler tube wall

Fig. 3.1.2.2-5 Improvepip

• Corner R processing

Nozzle

Membrane bar Membrane barWater-cooling wall tube Tie-bar clip

Tie-bar Tie-bar

• U band • Stop-end refresh processing

(padding + R processing) Header

Header Outlet header of reheater

Tube leg

Old toeWall

Tube legNew toe Torque bracket

Shear lag

(b) Improved structure

Z-type valve Angle valve

ment of support system for main ing of boiler

Fig. 3.1.2.2-6: Reinforcement of start system valve of boiler

146

The following are the reasons for renewal according to the equipment of each plant, renewal rate and equipment whose renewal due to “deterioration and damage” exceeds 20% extracted from such report: (1) Boiler plant related (Fig. 3.1.2.2-8) 1 Furnace tube

Renewal rate: About 36% Renewal time: From less than 40,000 hours to 200,000 hours

The renewal peak falls within the range of 100,000 to 200,000 hours. 2 Superheater 1st

Renewal rate: About 25% Renewal time: From less than 40,000 hours to 200,000 hours or more

The renewal peak falls within the range of 120,000 to 140,000 hours. Reason for renewal: As many renewals have been performed after 120,000 hours, conceivable reasons for

renewal are creep damage, external high-temperature corrosion and ash erosion. 3 Superheater 2nd to 4th

Renewal rate: 2nd About 56% 3rd About 66% 4th About 70% Renewal time: 2nd: From less than 40,000 hours to 200,000 hours or more The renewal peak falls within the range of 100,000 to 120,000 hours. 3rd and 4th: From 60,000 hours to 160,000 hours The renewal peak falls within the range of 80,000 to 100,000 hours. Reasons for renewal: As there are many renewals for those whose main steam temperature is 550°C or

more, for those for WSS (weekly start up and shut down) operation and for heavy oil-fired ones, conceivable reasons for renewal are creep damage and high-temperature corrosion.

4 Superheater weld joint with dissimilar materials Renewal rate: About 47% Renewal time: From less than 40,000 hours to 180,000 hours. The renewal peak falls within the range of 80,000 hours to 160,000 hours. Reason for renewal: As there are many renewals for those whose main steam temperature is high and for

heavy oil-fired ones, conceivable reasons for renewal are creep damage, thermal stress fatigue and high-temperature corrosion.

Multistage pressure reducing

Single-seat globe valve Multistage pressure-reducing valve

Fig. 3.1.2.2-7: Reinforcement of boiler feedwater control valve

5 Reheater 1st & 2nd Renewal rate: 1st About 60% 2nd About 62% Renewal time: Renewals are distributed widely at 60,000 hours or more.

The renewal peak falls within the range of 120,000 hours to 160,000 hours for the 1st superheater and within the range of 100,000 hours to 120,000 hours for the 2nd superheater.

Reasons for renewal: From the viewpoint of the number of start ups, steam temperature, conceivable reasons for renewal are creep and thermal stress fatigue.

6 Reheater weld joint with dissimilar materials Renewal rate: About 60% Renewal time: 1st From 60,000 hours to 180,000 hours The renewal peak falls within the range of 100,000 hours to 120,000 hours.

2nd From 60,000 hours to 120,000 hours

147

The renewal peak falls within the range of 80,000 hours to 120,000 hours. Reasons for renewal: From the fact that there are many renewals of those whose steam pressure is high for

both the 1st and 2nd reheater, and in the case of the 1st reheater, there are many renewals of those for DSS operation, a conceivable reason for renewal is thermal stress fatigue.

7 Valves Renewal time: Form less than 40,000 hours to 180,000 hours

The renewal peak falls within the range of 100,000 hours to 120,000 hours. Reasons for renewal: From the fact that there are many renewals of those with many start ups, a

conceivable reason for renewal is seat leak. 8 Electrostatic precipitator (discharge electrode, collecting plate, hammering device, charging equipment (P/P))

Renewal rate: Discharge electrode About 57%

Dp n

PuRegulation

Nam

es o

f ren

ewed

equ

ipm

ents

Economizer tubeFurnace tube

Superheater – 1stSuperheater – 2ndSuperheater – 3rdSuperheater – 4th

Superheater – Mixed fitting1st reheater – 1st

1st reheater – 2nd1st reheater – Mixed fitting

2nd reheater – 1st2nd reheater – 2nd

2nd reheater – Mixed fitting1st superheater inlet header

1st superheater outlet header2nd superheater inlet header

2nd superheater outlet header3rd superheater inlet header

3rd superheater outlet header4th superheater inlet header

4th superheater outlet header1st reheater inlet header

1st reheater outlet header2nd reheater inlet header

2nd reheater outlet headerEconomizer inlet header

Economizer outlet headerMain steam pipe

Main steam pipe T・Y pieceReheat steam pipe

Reheat steam pipe T・Y pieceBoiler circulation pump

Drum safety valveFurnace outlet safety valve

Superheater outlet safety valveReheater outlet safety valve

Start-up system line safety valvePCV

High-pressure system valvesStart-up system valves

Fuel oil pumpCoal pulverizer

StokerMill exhauster

Fuel oil tank heaterFuel oil tank bottom plate

Forced draft fanInduced draft fan

Gas recirculation draft fanGas-mixing draft fan

Discharge electrode for electrostatic precipitatorElectrostatic precipitator collecting plate

Electrostatic precipitator hammering deviceElectrostatic precipitator charging equipment (P/P)

EP ash-handling ash flow pumpEP ash-handling blower

EP ash-handling ash feed pipeAir compressor for control

Auxiliary air compressor Air compressor for soot blower

NOx removal plant catalysisBottom ash-handling jet pump

Bottom ash-handling ash flow pipeDuct expansion

Desulfurization system absorberDesulfurization system oxidation tower

Desulfurization system G/G heaterDesulfurization system pump

Desulfurization system fan

Fig. 3.1.2.2-8: Reason for renewal a

Degradation damage

nd renewal rate Renewal

148

amage reventio

by boiler system equipment rate (%)

erformance pgrading

Collecting plate: About 46% Hammering device: About 39% Charging equipment (P/P) About 29%

Renewal time: From less than 40,000 hours to 180,000 hours or more The renewal peak falls within the range of 100,000 hours to 120,000 hours. 9 Duct extension

Dp

PuRegulation

High-pressure external casingMedium-pressure external casing

Low-pressure external casingHigh-pressure internal casing

Medium-pressure internal casingLow-pressure internal casing

High-pressure external casing high-temperature boltMedium-pressure external casing high-temperature bolt

High-pressure internal casing high-temperature boltMedium-pressure internal casing high-temperature bolt

High-pressure rotorMedium-pressure rotor

Low-pressure rotorHigh-pressure-stage rotating blade

Medium-pressure-stage rotating bladeLow-pressure-stage rotating blade

High-pressure-stage stationary bladeMedium-pressure-stage stationary blade

Low-pressure-stage stationary bladeMain steam stop valve valve box

Control valve valve boxReheat steam stop valve valve box

Intercept valve valve boxCombined reheat valve valve box

Main steam stop valve high-temperature boltControl valve high-temperature bolt

Reheat steam stop valve high-temperature boltIntercept valve high-temperature bolt

Combined reheat valve high-temperature boltHigh-pressure rotor thrust bearing

Medium-pressure rotor thrust bearingLow-pressure rotor thrust bearing

High-pressure rotor journal bearingMedium-pressure rotor journal bearing

Low-pressure rotor journal bearingMechanical governor-mechanism set

BFPT external casingBFPT internal casing

BFPT high-temperature boltBFPT rotor

BFPT rotating bladeBFPT stationary blade

Booster feed water pumpFeed water pump

Condenser tubeCondenser body expansion joint

Condenser electrochemical protectorCondenser cleaning device

Vacuum pumpEjector

Condenser pumpCirculating water pump

Sea water coolerHigh-pressure feed water heaterLow-pressure feed water heater

Feed water system valveMain steam system valve

Reheat steam system valve

Nam

es o

f ren

ewed

equ

ipm

ents

Fig. 3.1.2.2-9: Reason for renewal and re

Renewal rate: About 63% Renewal time: From less than 40,000 hours to 180,000

The renewal peak falls within the range There are many renewals of those with D

(2) Turbine system related (Fig. 3.1.2.2-9) 1 Internal casing High-pressure internal casing

Renewal rate: About 12% Renewal time: As the number of renewed uIn the range of 120,000 hours to 180,000 hours, ther

14

Degradation damage

Renewal rate

newal rate by t

hours or more of 100,000 houSS, WSS and

nits is few, a pee is small grow9

amage revention

(%)

urbine system equipment

rs to 120,000 hours. high-sulfur heavy oil.

ak does not appear clearly. ing trend.

erformance pgrading

2 High-temperature bolt (bolt that tightens horizontal flange of casing) Renewal rate: For high-pressure internal casing: About 58%

For high-pressure external casing: About 39% For medium-pressure internal casing: About 51% For medium-pressure external casing: About 22%

Renewal time: For high-pressure internal casing Many renewals were performed within the range of 80,000 hours to 140,000 hours.

High-pressure external casing Many renewals were performed within the range of 80,000 hours to 140,000 hours.

For medium-pressure internal casing Many renewals were performed within the range of 60,000 hours to 160,000 hours.

For medium-pressure external casing As the number of renewed units is few, a peak does not appear clearly. Reason for renewal: As many renewals were performed for those with many start ups and shut downs,

conceivable reasons for renewal are high-temperature creep and fatigue. 3 Rotor (high-pressure, medium-pressure)

Renewal rate: High-pressure axle About 14% Medium-pressure axle About 34% Renewal time: As the number of renewed units is few, a peak does not appear clearly.

There are many renewals associated with improvement of performance. 4 Rotating blade

• High-/Medium-pressure-stage rotating blade Renewal rate: High-pressure stage About 40%

Medium-pressure stage About 64% Renewal time: Many renewals were performed within the range of 80,000 hours to 100,000 hours.

There are many renewals associated with performance upgrading. • Low-pressure-stage rotating blade Renewal rate: About 35% Renewal time: Many renewals were performed within the range of 80,000 hours to 160,000 hours.

5 Main steam valve valve box Renewal rate: Main steam stop valve About 15%

Control valve About 15% Renewal time: Main steam stop valve Many renewals were performed within the range of 100,000 hours to 160,000 hours.

Control valve Many renewals were performed within the range of 120,000 hours to 180,000 hours

6 High-temperature bolt (bolt that tightens upper bonnet of steam valve) • Main steam stop valve Renewal rate: Main steam stop valve About 53% Control valve About 52% Combined reheat valve About 69% Reheat steam stop valve About 72% Intercept valve About 63% Renewal time: Main steam stop valve

Many renewals were performed within the range of 80,000 hours to 100,000 hours. Control valve

Many renewals were performed within the range of 80,000 hours to 100,000 hours. Combined reheat valve

Many renewals were performed within the range of 80,000 hours to 100,000 hours. Reheat steam stop valve

Many renewals were performed within the range of 80,000 hours to 120,000 hours. Intercept valve

Many renewals were performed within the range of 80,000 hours to 140,000 hours. Reasons for renewal: Many renewals were performed for those whose steam temperature is high, and those

with many DSS and WSS, so conceivable reasons for renewal are high-temperature creep and fatigue.

7 Turbine bearing Renewal rate: Low-pressure rotor thrust About 22%

High-pressure rotor journal About 37% Low-pressure rotor journal About 25%

Renewal time: Low-pressure rotor thrust As the number of renewed units is few, a peak does not appear clearly. High-pressure rotor journal, low-pressure rotor journal Although renewals were performed within a wide time period range, relatively many renewals were performed within the range of 80,000 hours to 140,000 hours.

8 Condenser Tube, body expansion joint Renewal rate: Tube About 66%

150

Body expansion joint About 36% Renewal time: Tube

Many renewals were performed from the initial stage of after operation start. Body expansion joint

Many renewals were performed within the range of 80,000 hours to 100,000 hours. 9 Feed water heater

Renewal rate: High-pressure feed water heater About 33% Low-pressure feed water heater About 24%

Renewal time: High-pressure feed water heater Many renewals were performed with the range of 100,000 hours to 120,000 hours.

Low-pressure feed water heater Many renewals were performed with the range of 80,000 hours to 120,000 hours.

(3) Electric plant related (Fig. 3.1.2.2-10) 1 Generator Rewinding of rotor

Renewal rate: About 27% Renewal time: Relatively many renewals were performed within the range of 100,000 hours to

160,000 hours. There are many renewals of those with DSS.

Reasons for renewal: A conceivable reason for renewal is insulation degradation of the winding. 2 Exciter Motor, AVR

Renewal rate: Motor About 23% AVR About 53% Renewal time: Motor:

Relatively many renewals were performed within the range of 100,000 hours to 160,000 hours. AVR: The number of renewed units increases suddenly from 80,000 hours and continues to 180,000 hours.

Reasons for renewal: Generally, many renewals were performed on large-capacity units and those for DSS. Conceivable reasons for renewal are insulation degradation of the winding or aging of the equipment.

3 High-voltage motor Rewinding of stator coil Renewal rate: Outdoors About 43% Indoors About 39% Renewal time: Relatively many renewals were performed within the range of 80,000 hours to

140,000 hours. There is a trend of increasing renewal of those for DSS.

Reasons for renewal: Conceivable reasons for renewal are insulation degradation of the winding. 4 Control center

Renewal rate: About 28% Renewal time: Many renewals were performed at 100,000 hours or more.

There is a trend of increasing renewal of those for DSS. Reasons for renewal: A conceivable reason for renewal is deterioration of major parts (NFB, thermal relay,

conductor, etc.) 5 Supervisory control panel Central electric supervisory panel, protective relay panel

Renewal rate: Central electric supervisory panel About 19% Protective relay About 25%

Renewal timing: With respect to the protective relay panel, there is a trend of increasing renewal from 120,000 hours.

Reasons for renewal: Conceivable reasons for renewal are aging of the relay, drop in reliability and type change for the purpose of improving operation accuracy (from the electromagnetic/mechanical to stationary type).

6 Power supply system Storage battery, rectifier, uninterruptible power supply system Renewal rate: Storage battery About 81%

Rectifier About 48% Uninterruptible power supply system About 22%

Renewal time: The number of renewed units increases suddenly from 80,000 hors or more. Reasons for renewal: For the storage battery, a conceivable reason for renewal is deterioration of the

electrode plate, separator, etc. For the rectifier, renewal was performed due to deterioration and in conjunction with replacement of the storage battery. For the uninterruptible power supply system, conceivable reasons for renewal are deterioration and renewal associated with capacity increase, and system change (making the control system redundant , making the control system free from instantaneous disconnection) for improvement of reliability

151

Damage prevention

Degradation damage

Performance upgradingRegulation

Replacement of generator stator

Rewinding of generator stator

Replacement of generator rotor

Rewinding of generator rotor

Generator hydrogen gas cooler

Generator stator cooling system

Generator hydrogen gas shaft seal oil system

Generator lead bushing

Exciter

Generator for exciter

Motor for exciter

Rectifier for exciter

Exciter AVR

Cooling system for exciter

Isolated-phase bus support bushing

Isolated-phase bus wall passing-through bushing

Outdoor high-voltage motor

Rewinding of outdoor high-voltage motor stator coil

Outdoor high-voltage motor rotor

Indoor high-voltage motor

Rewinding of indoor high-voltage motor stator coil

Indoor high-voltage motor rotor

Set of metal-clad panel

Metal-clad circuit breaker body

Set of power center panel

Power center circuit breaker body

Set of control center panel

Central electricity supervisory control panel

Protective relay panel

Power supply system storage battery

Power supply system rectifier

Uninterruptible power supply system

Main transformer lead bushing

Main transformer-cooling system

House transformer main lead bushing

House transformer-cooling system

Starting transformer main lead bushing

Starting transformer-cooling system

Special high-voltage switch circuit breaker

Special high-voltage switch disconnecting switch

Special high-voltage switch support bushing

Special high-voltage switch wall passing-through bushing

Special high-voltage OF cable

Special high-voltage CV cable

High-voltage power cable

Low-voltage power cable

Control cable

Nam

es o

f ren

ewed

equ

ipm

ents

Renewal rate (%)

Fig. 3.1.2.2-10 Reasons for renewal and renewal rate by electric plant equipment

7 Main transformer Cooling system Renewal rate: About 30% Renewal time: The number of renewed unit increases suddenly from 80,000 hors or more. Reasons for renewal: Conceivable reasons for renewal are corrosion of elements and oil leak.

8 Cable High-voltage cable Renewal rate: About 49% Renewal time: There are many renewals performed at 100,000 hours or more.

Cable with high renewal rate Breakdown by insulation class by class: Butyl rubber 70% Cross-linked polyethylene 27%

Breakdown by plant condition: Air/Culvert 48% Pipe line 35%

152

Damage prevention

Degradation damage

Performance upgradingRegulation

Unit interlock Auxiliary machine interlock

Combustion control unit Feed water control unit

Steam temperature control unit Burner control unit

Governor control unit Tb monitor vibration diagnosis system

TB monitor shaft vibration meter Tb monitor shaft eccentricity meter

Tb monitor shaft position meter Tb monitor revolution meter

Tb monitor difference expansion meter Tb monitor cam position meter

Unit computer Data logger computer

Environmental data-processing computer Fuel control computer

Water quality control analyzer Fuel analyzer

Exhaust gas NOx analyzer Exhaust gas SOx analyzer Exhaust gas O2 analyzer

Exhaust gas CO analyzer Exhaust gas dust analyzer Leak oil monitor analyzer

Flammable gas monitor analyzer NH3 monitor analyzer

Waste water COD analyzer Waste water PH analyzer

Feed water system actuator Fuel system actuator

Starting bypass system actuator Air system actuator

Exhaust gas system actuator Air dryer for control

Air pressure-reducing system for control Feed water flow transmitter

Main steam flow transmitter Spray flow transmitter

Fuel oil flow transmitter Fuel gas flow transmitter

Main steam pressure transmitter Fuel oil pressure transmitter

Fuel gas pressure transmitter Drum-level transmitter

Deaerator-level transmitter Feed water flow element

Main steam flow element Fuel oil flow element

Fuel gas flow element Conveyor scale

Nam

es o

f ren

ewed

equ

ipm

ents

Renewal rate (%)

Fig. 3.1.2.2-11 Reasons for renewal and renewal rate by instrumentation control plant (4) Instrumentation control plant related (Fig. 3.1.2.2-11) 1 Control unit

Renewal rate: Unit interlock About 18% Auxiliary machine interlock About 14% Combustion control unit About 68% Feed water control unit About 66% Steam temperature control unit About 67% Burner control unit About 44% Governor control unit About 32%

Renewal time: Renewal of any of equipment was performed within the range of 60,000 hours to 160,000 hours.

Reasons for renewal: For renewal due to degradation damage, conceivable reasons for renewal are failure

153

due to deterioration of the signal conversion unit, indication mechanism, etc. of each controller and increasing difficulty in procuring parts because of discontinuance of production of similar equipment. For the renewal due to performance upgrading, many renewals were performed due to nationalization of imported products or due to change from an air or mechanical system to an electric or digital type, and it is conceivable that many renewals had the objective of conversion to APC or full automation, etc.

2 Turbine monitor Renewal rate: Vibration diagnosis system About 57%

Shaft vibration meter About 67% Shaft eccentricity meter About 65% Shaft position meter About 57% Revolution meter About 49% Difference expansion meter About 60% Cam position meter About 56%

Renewal timing: Renewal of any equipment falls within the range of 40,000 hours to 180,000 hours. Reasons for renewal: There are many renewals due to degradation damage and damage prevention.

Conceivable reasons for renewal are failure attributable to the deterioration of each sensor, conversion amplifier, reorder, etc. or increasing difficulty in procuring parts because of discontinuance of production of similar equipment.

3 Computer Renewal rate: Unit computer About 59%

Data logger computer About 32% Environmental data-processing computer About 41% Fuel control computer About 22%

Renewal time: Unit computer Within the range of 60,000 hours to 180,000 hours

Reasons for renewal: Conceivable reasons that there are many renewals due to degradation damage are failure attributable to the deterioration of the calculation unit, each sensor, memory, typewriter, etc. or increasing difficulty in procuring parts because of discontinuance of production of similar equipment.

4 Analyzer Renewal rate: Water quality control analyzer About 71%

Fuel analyzer About 48% Exhaust gas NOx analyzer About 79% Exhaust gas SOx analyzer About 78% Exhaust gas O2 analyzer About 82% Exhaust gas CO analyzer About 29% Exhaust gas dust analyzer About 34% Leak oil monitor analyzer About 9% Flammable gas analyzer About 49% NH3 analyzer About 35% Waste water COD analyzer About 30% Waste water pH meter About 35%

Renewal time: The renewal of the water quality analyzer, fuel analyzer and exhaust gas NOx analyzer falls within the range of 20,000/30,000 hours to 180,000 hours. The renewal of exhaust gas SOx analyzer, exhaust gas O2 analyzer, exhaust gas CO analyzer, flammable gas analyzer, NH3 analyzer, waste water COD analyzer and waste water pH meter falls within the range of 40,000 hours to 80,000 hours.

Reasons for renewal: Conceivable reasons that there are many renewals due to degradation damage are failure attributable to deterioration of the calculation unit, each sensor, memory, typewriter, etc. or increasing difficulty in procuring parts because of discontinuance of production of similar equipment.

5 Actuator Renewal rate: Feed water system actuator About 37%

Fuel system actuator About 29% Starting bypass system actuator About 17% Air system actuator About 36% Exhaust gas system actuator About 37%

Renewal time: The renewal of feed water system, air system and exhaust gas system actuator falls within the range of 60,000 hours to 180,000 hours.

154

Reasons for renewal: Conceivable reasons that there are many renewals due to degradation damage are occurrence of many failures attributable to deterioration of the control mechanism, positioner, etc. or increasing difficulty in procuring the parts because of discontinuance of production of similar equipment.

6 Air source for control, air dryer, air pressure-reducing system Renewal rate: Air dryer About 42%

Air pressure-reducing system About 17% Renewal time: The renewal falls within the range of 40,000 hours to 140,000 hours. Reasons for renewal: Conceivable reasons for renewal are occurrence of many failures as a result of

deterioration owing to change of the control mechanism, tower, etc. or increasing difficulty in procuring parts because of discontinuance of production of similar equipment.

7 Transmitter Renewal rate: Feed water flow transmitter About 65%

Main steam flow transmitter About 61% Spray flow transmitter About 61% Fuel oil flow transmitter About 50% Fuel gas flow transmitter About 45% Main steam pressure transmitter About 63% Fuel oil pressure transmitter About 54% Fuel gas pressure transmitter About 44% Drum-level transmitter About 60% Deaerator-level transmitter About 57% Feed water flow element About 12% Main steam flow element About 8% Fuel oil flow element About 22% Fuel gas flow element About 7% Conveyor scale About 33%

Renewal time: Renewals of feed water flow, spray flow, fuel oil flow, fuel oil pressure, drum level, and deaerator-level transmitters fall within the range of 60,000 hours to 180,000 hours. Renewals of main steam flow and main steam pressure transmitters fall within the range of 40,000 hours to 200,000 hours.

Reasons for renewal: With respect to degradation damage and damage prevention, conceivable reasons for renewal are occurrence of many failures due to deterioration of each sensor, signal converter, etc. or increasing difficulty in procuring parts because of discontinuance of production of similar equipment. With respect to performance upgrading, conceivable reasons for renewal are performance upgrading of equipment and shift of control equipment to the electric type or digital type.

3.1.2.3 Inspection technology/remaining life assessment technology Although the strength design of a boiler’s pressure part to be used under high temperature and high pressure is performed based on the 100,000-hour creep strength of the material to be used, from the facts that units whose cumulative operation hours reach 100,000 hours are starting to appear, and problems with thermal power units with years of service such as breakage of the steam turbine rotor, cracks in the steam turbine casing, etc. were reported in U.S.A., etc., from about the beginning of the 50s of the Showa period, interest in the assessment of soundness, recovery from deterioration and renovation technology for the major structural portions of major equipment in thermal units with years of service has grown. For thermal power generation technology, technical advances such as upsizing, higher temperature and pressure, etc. moved forward rapidly after World War II, and the age deterioration phenomenon itself was worldwide unknown area. For this reason, at present, virtually every technology developed and having become operational, such as deterioration characteristics of the material with years of service, mechanism of age deterioration, inspection technology and inspection equipment for deterioration diagnosis, remaining life assessment technology, renovation technology for recovery from deterioration, deterioration progress supervisory technology, etc., is unknown and not yet developed, constituting technical issues that we must address. From the 50s of the Showa period toward the 60s, inspection and assessment technology and remaining life assessment technology coping with deterioration phenomena that became obvious with time were developed and became operational, and at the present time, the focus of its development has shifted to labor saving, automation, broader use of robots, etc. In addition, development and practical use of operation supervisory/diagnosis technology for the purposes of strengthening and enhancing the operation supervisory aspect is being pushed forward.

155

[1] Boiler equipment Age deterioration phenomena that became obvious in major structural parts of boiler plants (representative examples) (1) Long-duration high-temperature creep, high-temperature oxidation/steam oxidation • Final SH, RH pipe damage

In particular, downgraded portion of points subjected to material change, in-furnace points subjected to material change

(2) Repeated fatigue due to thermal stress • Damage of evaporation pipe

Cracks originating from weld zone of fittings adhered to pipe Damage originated from weld zones of burner, wind box, inspection hole mounting frame Cracks originating from weld zone of slit-type fin Crack due to corrosion fatigue from internal surface of pipe at nose, deflection arch and deflection parts.

• SH, RH pipes Crack originating from weld zone of fixing spacer fixture

• Evaporation pipe, SH and RH pipes (non-heated part) Crack originating from toe part of stub weld • Leak of boiler combustion exhaust gas

Seal structural part at intersection part between boiler nose part wall and furnace back wall-suspended pipe Corner part of side wall part of front and rear wall pipes at heat recovery part Header guard of RH, Eco, etc. of heat recovery part Header around furnace bottom and seal structure part of ceiling wall passing-through part at GR port guard Tube bending part such as TV, inspection hole, burner, manhole, etc.

(3) Repeated fatigue due to long-duration high-temperature creep, thermal stress • Superheater, reheater header

Crack at weld zone of nozzle • Main steam piping

Crack originating from internal part of weld zone Crack at weld zone of branch piping nozzle Inspection technology and assessment technology having became operational (representative example)

• Boiler tube diagnosis UT system (Target: Superheater, reheater)

• Major piping diagnosis robot (target: major steam pipe) • Stack casing inspection robot • Remaining life assessment by means of destruction test

(targets: evaporation pipe, economizer tube, superheater tube, reheat tube) • Remaining life diagnosis by means of stress analysis

(Targets: T & Y pieces of major piping, weld zone of tube-adhered fixture, fin-mounting area of tube, support lug part of tube, header stub)

• Remaining life assessment by means of non-destructive test (A parameter method, void area rate, crystal grain deformation) • (Targets: drum, header, header stub)

[2] Turbine equipment Age deterioration phenomena that became obvious in the major structure of turbine plant (representative example) • Breakage of high-pressure rotor • Surface crack at base R part of high-pressure rotor 1st-stage wheel • Bending of medium-pressure ROBIN rotor • Crack at low-pressure rotor wheel stud part • SCC of low-pressure rotor shrink-fit wheel part • Lifting of high-pressure part rotating blade • Erosion and crack on rotating blade of low-pressure part • Crack on final-stage rotating blade racing wire • Nozzle erosion • Crack on rotating blade tenon • Surface crack on corner part of high- & low-pressure housings • Crack on medium-pressure housing (origin: repaired weld zone) • Breakage of the high temperature bolt and damage to the bolt screw thread. • Crack on major valve casing (origin: repaired weld zone, internal defect)

Inspection technology having became operational (representative example) • Rotor center hole ultrasonic flaw detection technology • Rotor center hole magnetic particle flaw detection technology

156

• Rotor center hole hardness-measuring technology • Rotor center hole replica-sampling technology • Rotor & casing embrittlement diagnosis technology • Blade stud part inspection technology • Tenon ultrasonic flaw detection technology • Rotor wheel ultrasonic flaw detection technology • High-temperature bolt (stud bolt) ultrasonic flaw detection technology

Remaining life diagnosis technology having been commercialized • Rotor, casing, major valve body (crack occurrence assessment, crack propagation assessment) • Rotating blade (lifting)

Hydraulic jack

Boiler, front

Rea

r wal

l

Fron

t wal

l

Side

wal

l

Burner wind box

Side top header support structure

Jack down

Step rod (φ85) Support beam Hoisting bar (φ140) U bolt (φ100)

In furnace

Oil jack

Boiler steel frame

Collector beam

Side wall top header

Fig. 3.1.2.4-1 Concept in dismantling of furnace wall by means of jack down construction method

(3) Electrical equipment Age deterioration phenomena having become obvious (representative example) • Crack on end ring (18Mn-5Cr steel) of generator rotor • Wedge crack on generator rotor • Generation of copper powder of generator rotor coil • Drop in insulation of generator rotor coil • Drop in insulation of generator stator coil • Water leakage from generator stator coil

Inspection technology and life diagnosis technology having become operational (representative example) • Ultrasonic flaw detection technology for end ring of generator rotor • Measurement of looseness of wedge of the generator rotator • Generator stator-winding diagnosis • Generator rotator diagnosis • Analysis on dissolved gas in oil of major transformer • High-voltage motor insulation diagnosis • High-voltage cable insulation diagnosis (4) Measurement & control equipment

Diagnosis technology having become operational (representative example) • Control system diagnosis system • Standard maintenance tool

(5) Operation supervisory, diagnosis technology (representative example) • Operation support system (alarm guidance) • On-site problem detection system • Boiler combustion diagnosis system • High-pressure feed water heater tube leakage detection system • Patrol support by means of trend supervisory on operation data and alarm • Simplified vibration diagnosis of rotating auxiliary machines (pump, blower) • Valve control by means of handy terminal • Portable ultrasonic leak detector (high-pressure heater, drain valve) • Leakage detector by means of infrared camera (boiler casing) • Boiler wall thick part life diagnosis (main steam pipe T piece, circulation pump casing)

3.1.2.4 Large-scale renovation examples (renovation of boiler furnace) For the purpose of responding to electric power demand-supply adjustment, from about the mid-50s of the Showa period, modification of machines designed for the base load to the DSS model was made, and full-scale high-frequency start-up & shut-down operation has been performed. As the number of start ups and shut downs increases, many cracks on the boiler tube and leakages started to

157

158

appear in all areas of boilers, and, for this reason, inspection and repair result in longer time and higher cost. In particular, heavy damage is found in the metallic substance weld zone of furnace pipe walls, furnace headers and nozzle weld zones at the reheater header due to fatigue and creep, and fundamental measures are becoming necessary. In the future, as these portions are important components of boilers, it is impossible to take fundamental measures through partial renovation. In addition, from the viewpoint that cost and work will be enormous, complete blanket renovation of furnace evaporation pipes, headers, etc. is starting to be carried out.

Control unit r(replaced) Change of control system

• Generator output • Main steam temperature • Main steam pressure

2nd superheater pipe (replaced)

Reheat pipe (replaced) Steam separator

(added)

Furnaceevaporation pipe

(replaced)

Furnace-sidecasing

(replaced)

1st superheater

Evaporator

Economizer

Fuel Feed water

Air

Burner valve

FurnaceBoiler circulation pump (added)

Together with this renovation, partial renovation of an accessory plant was performed so that such boilers also have cutting-edge performance. Further, for the purposes of shortening the renovation construction work period and securing safety in construction work, the development and introduction of new construction such as jack down construction (Fig. 3.1.4.2-1) are sought. (1) Examples of structural improvement measures associated with renovation

(1) Modify the boiler from the skin casing structure to the membrane wall structure to plan a reduction in thermal stress. (2) Modify the boiler from a weld construction consisting of the furnace wall and tension plate to a slide structure to plan the reduction in thermal stress. (3) Cause the header nozzle part to have sufficient flexibility to plan the reduction in thermal stress. (4) Modify the furnace wall passing-through part to the double-sleeve structure to avoid a concentration of stress. (5) Make the root of the nozzle and weld zone at the fine end smooth to relieve the concentration of stress.

(2) Renovation work examples In the KANSAI Electric Company’s Himeji No.2 thermal power plant Unit No.2 (325 MW), a subcritical pressure boiler that began commercial operation in 1964, from 1992 to 1993 blanket replacement of the boiler was carried out. This unit was originally oil fired; however, in 1980, modification to convert it to LNG fired was made and since then, this boiler has served as base thermal power. Since its start of commercial operation, this plant has operated for about 170,000 hours (number of start ups and shut downs: 662), and in addition to normal age deterioration, due to the fact that this plant has been used for DSS operation from 1985, life consumption due to low-cycle fatigue advanced in all areas of the boiler, minor problems occurred frequently, and the time required for inspection and repair increased. Then, as a result of study of a repair program according to the increasing longevity program, as it is more advantageous to replace the furnace water wall part completely than to repeat minor repairs in terms of cost, then it was decided to carry out total replacement. Further, together with renovation, performance upgrading including improvement of thermal efficiency and acceleration of the time required for start up is planned through modification from a constant-pressure to a variable-pressure operation method. (Fig. 3.1.2.4-2). 3.1.2.5 Technology and construction method for shortening of the term of periodic inspection work In addition to the peak in the summer season, for the purpose of responding to firm growth of demand in the winter season, the timing of periodic inspections tends to be concentrated in spring and autumn. On the other hand, the term of periodic inspection work tends to become longer due to the increase in the amount of repairs associated with aging of plants, and in the future, as the aging of large-capacity machines will also proceed, further efforts to shorten the term of periodic inspection work are sought. As a method to plan the shortening of the term of periodic inspection work, in addition to the effective classification and planning of repair work associated with aging, improvement of construction method that includes the following are also pursued. • Improvement of work efficiency through mechanization and broader use of robots • Improvement of work efficiency through labor-saving tools • Cutback in amount of works through blanket replacement of large parts (service, repair, etc.)

In addition, measures will be also be implemented from the viewpoint of the plant (Table 3.1.2.5-1), including: • Earlier start of work through forced cooling stop of the turbine • Improvement of workability through scale-up of manholes

Fig. 3.1.2.4-2 Outline drawing of renovation work for Himeji No.2 thermal power Unit No.2 boiler.

159

Table 3.1.2.5-1 Improvement examples for shortening of the term of periodic inspection work Examples Outline

Adoption of forced cooling system for turbine Introduce the outside air into high- and medium-pressure casings through the injection of cooling air or by means of vacuum pump to shorten the cooling time.

Adopt a high-performance oil-flushing system. Use the flushing system with a fine-mesh filter. Have gas turbine rotor spares. Have the gas turbine rotor of the combined-cycle generator as a spare to replace it at periodic

inspection. Have steam turbine rotor spares. Have the steam turbine rotor for geothermal heat as a spare to replace it at periodic inspection. Adopt a gas turbine static blade-sealing alignment system. Although alignment at replacement was performed at the manufacturer’s factory, alignment has

become available through installation of the system at site. Adopt a casing-tightening hydraulic bolt. Change the high-pressure turbine casing-tightening bolt from the shrink-fit type to the hydraulic

tension type. Modify from MHG to EHG. — Have one set of EHG parts spares. Have one set of EHG parts as spares to replace them at inspection. Additionally install an overhead traveling crane. Install an overhead traveling crane additionally. Making the overhead traveling crane faster. Make traveling and hoisting speed faster to plan effective use of the crane. Develop scaffolding at the furnace bottom part. Carry in one set of folded stages from the furnace bottom to extend it on the furnace bottom. Adopt a mobile clinker hopper. Change the clinker hopper to the mobile type to facilitate carry-in of scaffolding. Adopt a turbine rotor dry horning unit. Change the work form from manual work to work with the horning unit to plan greater efficiency of

work. Adopt a hydraulic torque wrench. Adopt the hydraulic torque wrench for crossover pipe flange-tightening work at low-pressure casing. Install a lifting unit for dismantling of major valves. Adopt a simplified lifting unit for lifting work of the main check valve, etc. to plan greater

efficiency of work. Adopt a hydraulic bolt for coupling. Tighten the coupling by means of a hydraulic tension bolt. Improve in-furnace scaffolding. Change the scaffolding from steel pipe scaffolding to steel fit scaffolding. Install a floor for carry-in of boiler materials. Install an out-furnace stage for carry-in of in-furnace scaffolding and for material storage space. Install a shutter at the boiler sound isolation wall opening. Install an opening at the sound isolation wall of the boiler to facilitate carry-in of materials, etc. Conduct dismantling and inspection work on the electric valve with greater efficiency. Change the power supply connection of the electric valve to the connector system. Contrive dismantling and assembling jigs for the coil-end cover of the generator. Fix the bottom cover to the jig and then cause it to rotate to facilitate removal. Adopt an ultrasonic expansion-measuring instrument Measure expansion of bolts by means of the ultrasonic measuring instrument. Turbine blade clearance-measuring device Insert the sensor into the clearance between the turbine blades to perform automatic measurement to

process its data. Adopt a laser-type centering measuring device. Measure turbine alignment by means of a laser to calculate the corrected value automatically. Adopt a turbine casing lifting-level supervisory unit. Monitor parallelism of the housing to be lifted by installing an ultrasonic-type distance sensor at 4

corners to measure it. Install a crane for light parts. In addition to an overhead traveling crane, install a crane for lifting light parts. Rotor center hole horning unit Unit that performs horning of the turbine rotor center hole automatically. Adopt a hydraulic jack for dismantling of housing. Cause the measuring sensors installed at 4 corners of the housing to synchronize with the hydraulic

jack to lift it horizontally. Adopt a jig for groove alignment of the boiler-cooling wall pipe. Jig for groove alignment of boiler water-cooling wall pipe. Upsize the boiler manhole. Upsize the bore of the boiler manhole to plan greater efficiency. Develop an internal surface inspection system of the boiler header part. Insert it through the header inspection hole to make observation by means of TV and observation

with an optical microscope.

3.1.2.6 Support system for creating a work plan for increasing longevity In order to continue stable operation of an aged thermal power plant after 20 years or more from its construction and start of operation. While maintaining the economics and its function, identifying the function, performance and soundness of equipment and presentation (recovery from deterioration) are performed more efficiently than conventionally. For this reason, a support system to create a work plan for increasing longevity that takes plant reliability and economics into account has generally been introduced and made use of. Creation of a work plan for increasing longevity is performed along each step of flow in Fig. 3.1.2.6-1. 1 Selection of critical equipment (Fig. 3.1.2.6-2) Assuming that the service period will be 60 years, the operation time, 400,000 hours and the number of start ups and shut downs, 5,000, select equipment for which remaining life control is believed to be necessary out of all equipment comprising the unit. Assess the probability of the occurrence of failure, effect of failure on output when it occurs, term of recovery work and cost, safety (social influence level), etc. comprehensively by means of the FMEA technique. 2 Investigation of the background of accidents and failures Collect and organize the records of accidents and failures regarding the selected critical equipment, and investigate the failure mode, life consumption factor, etc. during a long-term service period. 3 Breakdown and defragmentation of equipment (selection of critical points)

Select critical equipment.

Investigate history of accidents and failures.

Break down and defragment equipment.

Select the critical portion.

Select remaining life calculation measures.

Calculate the unit life consumption rate.

Calculate remaining life.

Create a long-term maintenance program list.

Carry out profit calculation.

Create a work program list for increasing longevity.

Unit price table

Future operation conditionsMarginal processing value

Operation history

Fig. 3.1.2.6-1 Flow in creation of work program plan list for increasing longevity work

Equipment Measuring device

Electric device Boiler system

Turbine system

Equipment Life

consumption factor

Initial failure mode

Final failure mode

Effect-level assessment

Probability of failure

occurrence

Effect on output

Degree of difficulty of recovery from failure

Term of recovery

Recovery cost

Safety General assessment

Score Critical equipment

(100 points or more)

Hig

h-pr

essu

re tu

rbin

e

Rotor

External casing

Internal casing

Nozzle chamber

1st blade ring

Creep, low-cycle fatigue

Progression of manufacturing

defect Burst

Creep, low-cycle fatigue

Crack Leakage

Creep, low-cycle fatigue Crack Leakage

Creep, low-cycle fatigue Crack Leakage

Creep Deformation Axial inclinationAbnormal vibration

Fig. 3.1.2.6-2: Example of critical equipment selection

160

Break down the selected critical equipment to the group that conceivably has the same structure, function, and design condition, and then defragment them based on the detailed structure to select “Point to which maintenance control should be performed based on life assessment.” 4 Selection of critical portion (Fig. 24)

Critical point : Dummy grooveCritical portion : Heat group groove

bottom Initial failure mode : Crack Life consumption factor:

Low-cycle fatigue Low-cycle fatigue + High-cycle fatigue

Critical point : Center holeCritical portion : Control stage bottom Initial failure mode : Crack Life consumption factor:

Creep + Low-cycle fatigue

Critical point : Rotor body Critical portion : Central part and othersInitial failure mode : Deformation

Deterioration of characteristics

Life consumption factor : Creep, softening, embrittlement

Governor side

High-pressure stage

Control stage

Ultrahigh-pressure stage

GEN side

Critical point : High-pressure final-stage blade groove

Critical portion : 1st tooth blade groove shoulder corner

Initial failure mode : Crack Life consumption factor:

Stress corrosion crack, corrosion fatigue

Critical point: Ultrahigh-pressure 1st-stage blade grove(a) Critical portion : Blade groove shoulder

corner Initial failure mode : Crack Life consumption factor: Creep, High-cycle fatigue

creep + high-cycle fatigue (b) Critical portion : Contact surface at rotating

blade root Initial failure mode : Crack Life consumption factor : Fretting fatigue

Inlet sideOutlet side

Rotating blade

(a) (b)

Fig. 3.1.2.6-3 Example of critical portion in high-pressure turbine rotor

Expand the critical points to the portion level further to extract the portions where occurrence of failure is possible during a long-term service period to select the portions that are the target of remaining life control as a critical portion from them. 5 Selection of remaining life calculation method Remaining life calculation methods can be divided broadly into following 4 methods: • Theory analysis method (non-destructive diagnosis method) • Destruction test method • Statistical method • Trend control method Out of these 4 methods, select the most adequate method corresponding to the initial failure mode and life consumption factor. 6 Calculation of life consumption unit rate Using the respective methods, obtain the amount of life consumption per 1,000 hours of operation (φc) or amount of life consumption per one start up and shut down (φf).

Calculate unit life consumption rate.

Out

put

Tem

pera

ture

di

ffere

nce

Com

pute

r

FEM analysis Life assessment curve

Stre

ss

Repetitions

Life consumption unit life

Processing limit

Future operation conditions

Life consumption

unit life

Operation history

Calculate remaining life.

Limit processing value

Con

sum

ed li

fe

(Year)

Consumed life

Remaining life

Fig. 3.1.2.6-4: Calculate remaining life.

161

Table 3.1.2.6-1 Example of plan list for increasing longevity (Unit: million yen) Fiscal

System 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010

Boiler related 2226 83 62 15 35 10 268 3356 17 44 17Turbine related 712 85 1 35 37 83 95 1203 103 21 70Electric related 12 14 2 406 828 374 54 116Measurement related 12 20 70 300 320Total 3002 168 77 50 74 93 789 5457 494 419 523

Optimum renewal fiscal year

Uni

form

ann

ual c

ost o

f mai

nten

ance

cos

t

Uni

form

ann

ual c

ost o

f ren

ewal

Fiscal year in A.D.

Fig. 3.1.2.6-5 Example of profit calculation result

7 Calculation of remaining life (Fig. 3.1.2.6-4)

(1) Calculation of consumption life

φ = φc 1000

timeoperationCumulative + φf × Cumulative number of start ups and shut downs

(2) Establishment of marginal processing value The marginal processing value is a consumption life value that determines: ・start timing of inspection ・implementation timing of re-diagnosis ・implementation timing of repair and replacement And it is established in accordance with the equipment and portion.

(3) Establishment of future operation conditions Based on the power generation plan, the annual estimated operation hours (Ta) and estimated number of start ups and shut downs (Na) are assumed to calculate the future operating conditions by multiplying such number by life consumption unit.

(4) Calculation of remaining life Remaining life indicates the years until the unit reaches limit processing value and is calculated by means of:

Remaining life (years) = NaTa/1000

lifenconsumptio valueprocessingLimit fc ×φ+×φ

−

8 Creation of long-term maintenance program list A long-term maintenance program list represents the maintenance cost within the assumed service period on a year-by-year basis, and the maintenance cost is calculated from the cost required for inspection, diagnosis, repair and replacement on a portion-by-portion basis and the quantity of the corresponding portion.

(1) Entry of unit price Study the most adequate inspection, diagnosis, repair and replacement methods on a portion-by-portion basis to select its unit price from unit price list that this system has to enter it.

(2) Creation of long-term maintenance plan list There are 2 types of long-term maintenance plan list, on a portion-by-portion basis and equipment-by-equipment basis, and clearly shows the fiscal year when inspection, diagnosis, repair and replacement are required and the respective costs.

9 Implementation of profitability calculation (Fig. 3.1.2.6-5) Study the selection whose renewal of equipment or partial repair is required, and select the most economical renewal timing in the case of equipment renewal, as well as the optimum replacement range combining mutual equipment.

Creation of longer life work plan list (Table 3.1.2.6-1) With respect to all critical equipment, calculate the maintenance cost required within the future assumed service period, and an increasing longevity work plan list is obtained by summing such costs.

162

3.2.1 Causes of damage to boiler equipment (1) Trend of damage

An example of analysis of the ratio of aged deterioration damage for each component of a boiler and the damage ratio of its pressure-retaining parts by each cause are shown in Fig. 3.2.1-1. Such pressure-retaining components as the furnace wall, super-heater, re-heater, economizer, piping, etc. account for 67% of the entire boiler equipment. Thermal fatigue, corrosion fatigue, and creep damage account for 83% of the causes of total damage. As measures to improve the reliability of thermal power generation plants, it is important to prevent such fatigue, corrosion fatigue, and creep damage from occurring to such pressure-retaining components.

Ratio of occurrence of failure for each

component of boiler equipment

Cause of damage to pressure-retaining

components

Furnace wall 31%

SH/RH/ECO20%

Piping 16%

Valves 5%

Fan 10%

Non-pressure-retaining components 4%

Others 14%

Others7%

Thermal fatigue/

corrosion fatigue 68%

Creep 15%

Wear 5%

Corrosion 5%

Fig. 3.2.1-1 Ratio of the components becoming defective/ratio of cause of damage

Deterioration of materials

Overheating

Corrosion

Wear

Corrosion fatigue

• Drop in bearing force caused by use for a long period of time

• Change in material quality • Defective materials

Swelling out Deformation

Creep rupture

Rupture by spouting

• Fatigue occurring under the environment of the inner surface of the tank caused by filled water

Rupture by corrosion fatigue

• Thermal fatigue • Mechanical fatigue

Occurrence/growth of crack(s)

Rupture caused by fatigue

• Corrosion by low temperature

• Corrosion of the inner surface of tank caused by filled water

Occurrence/growth of crack(s)

• Clogging inside the piping caused by foreign materials

• Imbalanced flow of fluid within the piping

• Adherence/growth of scale within the piping

• Corrosion by high temperature

• Erosion by ash • Wear by high-velocity air

flow within the component

Excessive increase in load stress caused by the decrease in effective thickness

Rupture caused by static stress

Leakage

Fatigue

Fig. 3.2.1-2 Cause of damage to boiler equipment (2) Cause of damage

The types of damage generally experienced with the pressure-retaining components of the boiler equipment are shown in Fig. 3.2.1-2 for convenience. 3.2.2 Example of damage and measures to improve bearing force

Examples of typical damage experienced so far and measures taken to improve the bearing force are explained below. 163

(1) Creep and creep rupture Each pressure-retaining component of boiler equipment has been designed to have a creep rupture strength of

100,000 hours or longer. However, should the component be overheated beyond the designed temperature range for any reason, or should any decrease in thickness advance, creep damage may advance within a very short period time resulting in rupture. Typical causes of damage found in the examples are summarized as follows:

• Overheating due to an extreme decrease in flow rate within the piping caused by clogging with of foreign matters or by peeling off or accumulation of steam oxidation scale.

• Temperature rise at the piping wall caused by the growth of scale adhered to the inner surface of the pipes of the furnace evaporation piping, or the growth of porous-type scale with low heat transmission efficiency

For the accumulation of steam oxidation scale within the stainless steel piping of the super-heater, etc. and accumulated at the U-bend, it is considered effective to suppress the scale from growing if fine-grade steel is employed or the inner surface of the piping is shot blasted. Creep damage includes creep created over considerably long hours because the inner-pressure stress increases by the decreased thickness caused by high-temperature corrosion, etc. A lot of damage has been found caused by the scale adhering to the inner surface of the pipes of furnace evaporating piping. Standard water quality control of the supply water and tank water and implementation of proper acid cleaning is an important task. To determine the timing of acid cleaning, monitoring of the pipe wall temperature by a pulling-out check of the pipes at the time of regular inspection or by using a Cordal-thermocouple (embedded thermocouple) is available. (2) Thermal fatigue

Thermal fatigue occurs by the repeated effects of high thermal stress that is generated by the temperature differences among the member materials. The thermal fatigue generated by the start/stop operation of boiler equipment or by load fluctuation is a type of low-cycle fatigue in general. The surface of the broken part by thermal fatigue is uneven and rougher than that caused by mechanical fatigue where high-cycle fatigue is accompanied by vibration, etc. The surface of the cracked part is normally open to some extent. The causes generating thermal stress vary depending on the structure of each component of the boiler equipment. Examples of portions where thermal fatigue occurs and measures to reduce the stress are shown in Table 3.2.2-1 in a concrete fashion. (3) Corrosion fatigue

In the case of the inner water supply system of furnace, economizer, etc., such corrosion fatigue as cracking generated not only from the outside of the piping but also from the inside has been experienced. Corrosion fatigue is a phenomenon in which fatigue cracks are generated and grow because the strength against fatigue declines remarkably to a larger degree than the same in an air atmosphere, when the metal receives stress repeatedly in a corrosive environment. It is basically generated on the portion where thermal stress, etc. is large.

As a typical example of the relationship between thermal fatigue and corrosion fatigue, the tension plate and welded portion of the furnace wall are shown in Fig. 3.2.2-1.

At the portion where the tension plate has been welded directly to the furnace wall, thermal stress is generated by the temperature difference between the tension plate and the furnace wall in the direction of the piping axis and to right angles of the piping axis. The maximum stress is generated on the welded portion of the tension plate on the external face of piping. Thermal fatigue cracks are generated on the toe of the weld where stress concentrates.

On the other hand, stress is generated on the rear side of the weld on the inner surface of the piping. The stress on the inner surface of the piping is smaller in general than that on the outer surface.

164

Table 3.2.2-1 Portions where thermal stress is generated and measures to reduce the stress Portion Portion where fatigue damage

occurs Mechanism of generation of

stress Measures to reduce stress

① Furnace wall

Coo

l wat

er

If the boiler water temperature should change upon boiler start/stop operations, temperature difference occurs between the furnace wall and the sub-wall or between the sub-wall and the rear smoke duct wall, which generates stress on the fin edge of the furnace wall.

Arrange the fin edge in a large arch shape R-machining of fin edge

R-machining (9) 2 A h Arch

② Furnace wall seal box weld

Tool box

Concentration of stress

If the boiler water temperature should change upon boiler start/stop operations, temperature difference occurs between the furnace wall piping and the seal box, by which stress concentrates at the corner.

Change the shape of the seal box corner to an arch. Bend the seal box side in 2 steps.

R corner

2-step bending

③ Fixtures mounted on furnace

wall Tension plate

If the boiler water temperature should change upon boiler start/stop operations, temperature difference occurs between the furnace wall piping and the mounted fixture, which generates stress on the welded portion.

Change the structure of the furnace wall piping and mounted fixture to a sliding structure.

Tension plate

Slide ④ End bar and skin casing for

the hole of the ceiling End bar

Deformation

Welded portion

Pip

ing

on

the

ceili

ng

The entire portion tends to deform due to the temperature difference between the front- and rear-end bards at the ceiling hole, but is locked by the ceiling piping, resulting in the generation of stress on the welded portion. Due to the temperature difference between the end bar and the skin casing, cracking occurs on the skin casing.

Change the structure of the ceiling piping and end bar to a sliding structure. Change the skin casing to a 2-step bent type.

End bar

Piping on the ceiling

2-step-type skin casing

⑤ Skin casing below economizer

Due to the temperature difference between the wall piping surrounding the rear smoke duct and hopper, stress is generated on the skin casing, resulting in cracking.

Change the skin casing to 2-step projected bellow type.

165

Damage caused by corrosive fatigue Damage caused by heat fatigue

Generation of high stress

Generation of high stress

Low expansion Low expansion

Tension plate Tension plate High

expansionHigh expansion

Fig.10 Generation status of incompatiblility at the end portion of straight-finned

economizer

Fig.9 Typical example of heat and corrosive fatigue

Table 3.2.2-1 Portion where thermal stress is generated and measures to reduce the stress Portion Portion where fatigue damage

occurs Mechanism of generation of

stress Measures to reduce stress

⑥ Nozzle of super-heater and re-heating pipe head

Piping on the ceiling

Temperature difference occurs between the nozzles during start/stop operations, and bending stress is generated on the welded portion that has been locked between the nozzles and ceiling hole.

• Change the nozzles to the flexible type.

Flexible

⑦ Joint welded by dissimilar

metals

Coe

ffici

ent o

f lin

ear t

herm

al

expa

nsio

n Inconel welding electrode

Present style Improved

style

SUS rod

SUS steel Cr-Mo steel

Due to the difference in the carbon content, carbon migrates to the metal to be welded from low-alloy steel, yielding a decarbonized layer as a result, and the strength on the low-alloy side declines. By the difference in thermal expansion between the austenitic stainless steel to be welded and the low-alloy base steel, thermal stress is generated on the portion welded. Because of its high temperature, creep damage also occurs.

By using Inconel-family welding electrodes and by reducing the linear expansion difference, reduce the stress. Prevent the strength from declining by preventing the carbon from migrating.

Welding at factory using Inconel welding electrodes

⑧ Saddle spacer welding portion

Welding-typesaddle Fixed

Within a structure supported by a spacer fixed by welding to the hanging pipe of the horizontal-type super-heater/re-heater, thermal stress is generated on the spacer-welded portion due to the temperature difference between the upper and the lower pipes.

Employ a flexible spacer.

Flexible saddle spacer

Slide

Fixed

⑨ Welded portion of small-diameter nozzle of pipe header

Piping reaction force

Deformation (29) Pi i ti

If the air vent pipe and drain pipe of pipe header are the type of such structure as being locked in the housing hole, thermal stress is generated at the welded portion of the nozzle of pipe header.

Change the small-diameter pipe to a flexible type. The form of nozzle of pipe heater should be butt welding type.

Hole to be fixed

Flexible bending piping

166

Portion Portion where fatigue damage occurs

Mechanism of generation of stress Measures to reduce stress

⑩ Welded portion to fix the anchor plate Portion where

cracking occurs Filler plate

Tie bar

Anchor plateStand-off

Due to the temperature difference between the anchor plate and furnace wall piping occurring by start/stop operations of a boiler, stress concentrates at the welded portion of the anchor plate.

Separate the anchor plate to fit it by full arc welding and make the size smaller.

Filler plate Driber

Anchor plate Stand-off

Stopper

⑪ Membrane-edge connecting waterwall and cage walls

Membrane

Waterwall pipe

Portion where cracking occurs

Due to the temperature difference between the waterwall and cage piping occurring by start/stop operations, stress concentrates at the connection and membrane edge.

Refresh the connection membrane edge and provide R-machining to the welding stop end.

Membrane

Waterwall pipe

⑫ Welded liner of

super-heater/desuper-heater Stopper

Fore

part

of

the

tank

Desuper-heater main body

Spray nozzle

Liner

Support WeldiSupport ring Base pipe

Protection cylinder

Fitting of liner (Welding type) Portion where

cracking occurs

By the ON/OFF injection from the spray nozzle of the super-heater/desuper-heater, the liner is bumped and stress concentrates at the liner-welded portion.

Change the structure of the spray nozzle and improve the method of fixing the liner by changing to the pin type.

Stopper

Desuper-heater main body Spray nozzle Support Protection

cylinder

Pin Pin

Improved structure (pin type)

⑬ Measures against damage to main piping support lag

Main piping

Thermal insulation material

Support lag

Portion where cracking occurs

Temperature difference occurs on the supports inside and outside the thermal insulator, and excess stress concentrates at the support-welded potion.

Change the support lag to shear lag. Shear lag

Thermal insulation materialBand

Hanging bolt

Main piping

⑭ Ceiling hole

Ceiling piping

Crwon

Due to the temperature difference between the crown and the piping, stress concentrates causing cracking to occur.

Use a sleeve through the ceiling hole and avoid direct welding of the crown to the piping.

Ceiling piping

CrownTo add a sleeve

⑮ Connection of loop pipes

Tie rod Sliding spacer

Hanging loop pipe

When there is a temperature difference at operation start, cracking occurs at the portion where a linkage fixture has been installed due to stress concentration.

Use a sliding spacer at the portion where high temperature is transmitted and to avoid any locking. Change the tie lag in the rear heat transmission portion to an oval-shaped lag to soften the concentration of stress.

(Single lag) (Oval lag)

⑯ Inner casing of ceiling enclosure

The corner casing cannot absorb the expansion force from 3 sides, and cracking causes gas leakage.

Use a corrugated-type expansion at the corner.

Corrugated-type expansion

Pipe header at furnace side wall

Pipe header at furnace front wall

167

However, in a corrosive environment, strength against fatigue declines, which causes cracking at the inner face within a pipe by corrosion fatigue.

As a characteristic of a cracked surface caused by corrosion fatigue, many cracks are accompanied by pits caused by the corrosion along the cracks.

As basic countermeasures, such actions to soften the thermal stress are considered important. In such a case, it is required to change the tension plate support to a sliding type and improve the structure so that the thermal stress may be softened.

Examples of other corrosion fatigue are introduced below: ① Straight fin end of economizer piping (Fig. 3.2.2-2)

The occurrence of cracking was experienced at the straight fin end of the economizer piping, caused by thermal stress accompanied by intermittent water supply in order to keep the drum at a constant level at the operation start of the boiler.

Cracking has started from the inner surface of the piping. Corrosive fatigue is the cause. ② Ligament of the pipe header at the inlet of the economizer (Fig. 3.2.2-3)

The occurrence of cracking was experienced at the ligament of the pipe header at the inlet of the economizer due to the same cause as above. This was also caused by corrosion fatigue. (4) Mechanical fatigue

In the case of mechanical fatigue, the cracking is a type of transgranular cracking in general. The ruptured face has a fine fatigue face, and no extension by rupture was detected.

Pipe header at the inlet of economizer

Nozzle at outside furnace Pipe header at outside furnace

Fig. 3.2.2-3 Example of corrosion fatigue of the inner ligament of the pipe header nozzle at the inlet of the economizer (5) High-temperature corrosion

The surface stainless steel pipe affected by high-temperature corrosion has been damaged by corrosion in a pockmarked fashion. The corroded portion is composed of an oxide layer – a polysulfide layer – a carbonized layer – base metal from the outer piping surface. From the viewpoint of microstructure, the corroded and carbonized structure of grain boundary is found. A drop in expansion as well as a drop in strength can be detected.

168

Table 3.2.2-2 Classification of measures to improve bearing force

Cause Measures Subject portion Phenomenon

(6) Low-temperature corrosion

Aged strength drop by creep at welded portion

Assessment of remaining life by replica, ultrasonic testing, TDFD, ELFOSS, UT inspection

Pipe header of super-heater/re-heater, main-/high-temperature longitudinal re-heating steam piping, around welded portion, elbow/Y-piece-welded portion

Wear

Restriction on elongation by heat

Add flexibility Pipe header stub, finish of sealing

Thermal shock

Change of shape, improvement of material, improvement of the shapes of seat and piping

Desuper-heater spray, small-diameter piping with main piping (drain pressure tank)

Dissimilar metal welding (SUS/Cr-Mo)

Inconel solvent Joint of different piping material, fixture of different material

Corrosion fatigue

Change of structure and shape, water quality control

Fixture welded to furnace wall piping, ligament at the inlet of the economizer

High-temperature fatigue, oxidation

Improvement of bearing force of material, addition of extra welding

Super-heater, re-heater Furnace wall

Oxidation of steam (SUS piping)

Fine-particle SUS materialInner face shot blast

Super-heater, re-heater

R-machining, chamfering, change of shape

Shape the stress concentrates

Corrosion

Fatigue (including creep fatigue)

Creep

Fin-end portion, pipe header lid at the corner of the burner wall box, expansion for the smoke duct

Expansion of casing Piping-supporting fixture, back-stay prevention fixture Sliding

Furnace wall, super-heater, re-heater

Protector, pipe thermal spraying

Coal ash, soot blow

The AH element, seal plate, etc. are main damaged caused by low-temperature corrosion. It has been reported that the expansion at the AH outlet, damper, etc. were affected by sulfuric acid corrosion when HS oil had been used.

In addition, such an example was reported where corrosion was generated on the outer surface of the furnace or the furnace wall of the rear smoke duct caused by condensed sulfuric acid in the steam-condensed water while the boiler was kept at standstill. (7) Measures to improve bearing force

As explained above, the components composing the boiler equipment receive various types of damage depending on the environment of use, most of which are combinations of several damaging factors. With respect to such damage, various measures to improve the bearing force, which are classified and detailed in Table 3.2.2-2, have been taken. 3.2.3 Technology to assess the remaining life

The methods to assess the remaining life of boiler equipment can be divided into the following 3 types: • Stress analysis method such as the finite element method, etc.

• Destructive test method

• Non-destructive test method

Except the stress analysis method, it is not possible to assess the remaining life if you use only any one of above methods. Assessment of remaining life is carried out by combining the methods. (1) Stress analysis method

This is a method of obtaining the life consumption by calculation based on the equipment subjected to 169

assessment, the geometric shape of the part, the operation history such as temperature, stress, etc., the strength against creep rupture, and the properties of the materials. The finite element method using a computer makes it possible to analyze the stress of a complex structure.

With respect to the properties of the material to be used for the analysis, it is required to include the safety ratio in the laboratory data to some extent considering possible variations of the properties. Therefore, the assessment result leans towards the safe side.

With respect to such operation history as the temperature, stress, etc. to be used for the analysis, calculation is performed by dividing the operation history into several typical patterns. In order to cope with the recent complex operation history, the remaining life is sometimes assessed by installing a life-monitoring device at the pipe header at the outlet of the super-heater, water separator, boiler circulation pump, etc. (2) Destructive test method

This is a method of estimating the remaining life through various types of destructive tests by taking out test specimens from the components actually put under operation. This test method is usually employed for components (typically, the boiler tube) from which test specimens can be easily taken out. The advantage of this method is that the remaining life of a given material can be assessed directly, including its history at the time of manufacture, even if the temperature or stress history of the material in the past is not made clear. The disadvantage is that sampling is required, the portion where the test specimen has been taken out needs to be repaired, and time and expense are required for creep rupture testing, fatigue testing, etc.

As a measure making it possible to perform destructive testing by using much smaller test specimens, destructive testing through a miniature test is available. As shown in Fig. 3.2.3-1, its effectiveness has been verified.

Comparison of strength against creep rupture between a conventional test specimen of 1 Cr 0.5 Mo Steel and a miniature test specimen

Conventional test specimen

Miniature test specimen

Stre

ss (M

Pa)

Miniature test specimen

Time of rupture

Conventional test specimen

Fig. 3.2.3-1 Miniature creep rupture test

(3) Non-destructive test method The advantage of the non-destructive test method is that any critical component with respect to the stress can be

assessed in a short time without sampling. This test is used together with the assessment of remaining life through stress analysis.

This method varies depending on the material quality or state of damage to the subject component. In Table 3.2.3-1, the non-destructive assessment method for creep damage and fatigue damage is shown.

170

Table 3.2.3-1 Non-destructive method of assessing the remaining life of components affected by creep/fatigue damage Low-alloy steel Steel Subject

damage Method of assessing remaining life

Method as described in Attachment 3 of the

Electricity Utilities Industry Law Base metal Welded

portion Base metal Welded portion

Deposition intergranular distance method ⎯ ⎯ ⎯

Hardness-measuring method ⎯ ⎯ ⎯ Structure comparison method ⎯ ⎯ ⎯ AC electric resistance method ⎯ ⎯ Void (cavity) area ratio method ⎯ ⎯ Void density method ⎯ ⎯ A-parameter method ⎯ ⎯ ⎯ ⎯ Crystal grain deformation method ⎯ ⎯ Carbide structure-measuring method ⎯ ⎯ ⎯ Ultrasonic method ⎯ ⎯ ⎯ Structure-quantifying method ⎯ ⎯ ⎯ ⎯

Creep damage

CMA density spectrum method ⎯ ⎯ ⎯ ⎯ Fatigue Microscopic-crack method ⎯ ⎯ ⎯

① Creep damage (a) Deposition intergranular distance method

This method is used for the assessment of creep damage of low-alloy steel base metal. Low-alloy steel is a material whose strength against creep has been raised by depositions and shows ductile creep damage. When used for many hours in a high-temperature atmosphere, the intergranular distance of this disposition becomes larger and, at the same time, resistance against deformation declines, causing the creep to accelerate. This phenomenon is represented by the creep distortion–time curve in general. The change depends on the temperature and stress of the subject component. By measuring the intergranular distance between particles of disposition, the creep distortion at the time of assessment can be obtained. Therefore, the behavior of creep distortion thereafter can be predicted, and the creep remaining life can be assessed. The intergranular distance of disposition is obtained by image processing of the replica taken out from the subject component using an electrolytic discharge-type scanning electron microscope (Fig. 3.2, 3-2).

Replica

Point

DispositionScanning line

Mean free-path

Scanning-type electron microscope

Cre

ep ra

te c

onst

ant

Average intergranular distance (µm)

Fig. 3.2.3-2 Disposition intergranular distance method

171

(b) Hardness-measuring method The crystal grains of 9 Cr base metal steel are very fine, and this base metal has a hard structure of initial

hardness. Different from low-alloy steel, no metallic structural change can be detected even when creep damage grows. However, its hardness tends to drop gradually.

Therefore, by measuring the hardness and referring to the master curve that indicates the relationship between the hardness and the amount of damage, the life consumption ratio can be assessed (Fig. 3.2.3-2).

Vickers hardness

The amount of creep damage

Fig. 3.2.3-2 Hardness-measuring method (c) Structure comparison method

This method is very effective for the assessment of components affected by low-alloy steel welding heat that indicates fragile creep damage. Comprehensive assessment of life is carried out by comparing the standard structure corresponding to the life consumption ratio by taking out the replica/extracted replica from the component subjected to assessment and by using 3 parameters of deposition distribution pertaining to mechanical damage such as creep voids or microscopic-cracks generated as the creep damage grows, optical microscopic structure pertaining to the change in the distribution of the metallic structure, or carbide using various types of microscopes and the change in the shape or size of the deposed carbide.

As shown in Fig. 3.2.3-3, the factors for assessment of respective damage are divided into 3 steps or 4 steps. By combining them, the life consumption ratio is estimated comprehensively within a range of 8 categories. For example, when mechanical damage is IID, the microscopic structure is IIIM, and deposition distribution is IIP, the comprehensive damage category of the life consumption ratio by creep rupture is estimated to be E, namely 50 – 60%.

By combining various factors for assessment of life, assessment with high precision becomes possible in the entire range covering the first half and second half of life.

172

Replica

Disposition


Life consumption ratio by creep breakage (%)

Extracted replica

Component surface (etched surface)

Optical microscope Analysis electron microscope

Creep cavity Micro-crack Metal structure

Damage factors Comprehensive damage category Mechanical

damage

Microscopic structure

Deposition distribution

Fig. 3.2.3-3 Structure comparison method

(d) AC electric resistance method

This method is effective for the assessment of creep damage of components affected by welding heat (hereinafter referred to as the HAZ portion) of low-alloy steel and 9 Cr steel.

Volta

ge d

rop

ratio

def

ined

by

initi

al v

alue

Material not used yet

Material damaged by creep

Life consumption ratio by creep breakage (%) Fig. 3.2.3-4 AC electric resistance method

The creep damage of the component affected by welding heat from these steels is a type of fragile damage and generates creep voids at the grain boundary. As the generation of voids increases, the electric resistance tends to become stronger (Fig. 3.2.3-4). The amount of damage is assessed by using the electric resistance ratio of unused material and the electric resistance ratio of the component being assessed, and by referring to the master curve indicating the relationship with the amount of damage. Assessment accuracy has been improved by making it

173

easier to grasp the level of damage proximate to the surface by using an alternative current. In addition, it is required in this method to spot weld a platinum wire to the subject component. If an electrode has once been installed, building of a scaffold, thermal insulation, removal/restoration of the exterior plate, and polishing of the subject component for inspection are not required thereafter. Therefore, the costs for inspection can be reduced. In addition, it is possible to make measurement at any time during operation. This method can also be used for monitoring the main piping, etc. (e) Void (cavity) area ratio method

As shown in Fig. 3.2.3-5, voids are generated at the grain boundary when the HAZ portion of low-alloy steel or 9 Cr steel is affected by creep damage. The number of voids increases as the damage grows. The voids become a crack after growing/combining (namely, the area of voids increases), and finally result in the rupture of the component material. In this method, the ratio between the total area of voids generated within the observation visual field and the total observation visual area is defined as a void (cavity) area ratio. Using this ratio together with the master curve prepared by its correlation with the degree of creep damage, the life is assessed in this method (Fig. 3.2.3-5).

Replica


99% reliable section of creep damage ratio

Cav

ity a

rea

ratio

S0

Creep damage ratio φc

Welding metal (570°C) Regression curve 99% reliable section

Fig. 3.2.3-5 Void (cavity) area ratio method

Incidentally, the behavior to generate voids is different in such low-alloy steels as 2.25 Cr-1 Mo steel, etc. and 9

Cr steel. It is required to use the master curve suitable for the respective type of steel. (f) Void density method

The ratio between the number of voids in the observation visual field and the observation area is defined as cavity density. Referring to the master curve indicating the relationship between the cavity density and the amount of damage, assessment of the life of the component subjected to assessment is carried out in this method. (g) A-parameter method

This is a method to be used for the assessment of creep damage at the HAZ portion of low-alloy steel. This method was developed by English researchers. Creep voids generated as creep damage grows are generated at the grain boundary. Draw an optional scanning line in the metal structure of the subject component. The ratio of the number of grain boundaries where voids have been generated against the number of grain boundaries that intersect this scanning line is defined as the A-parameter. The life of the component subjected to assessment is assessed in this method by referring to the master curve indicating its relationship with the amount of damage (Fig. 3.2.3-6).

174

Relationship between A-parameter and life consumption ratio by creep rupture

Amount of creep damage (%)

Fig. 3.2.3-6 A-parameter method

(h) Crystal grain deformation method

The base material of low-alloy steel used for boiler equipment has been made considerably soft considering easiness of machining and welding. Therefore, level of generation of voids when receiving creep damage is lower than that of the HAZ portion, but instead plastic deformation generates easily. Under such circumstances, the crystal grain is expanded gradually to be long and narrow in the applied stress direction and becomes uniform. The level of this uniform style is quantified by the standard deviation of the frequency distribution of the maximum-diameter angle (an angle created by the direction of the maximum diameter of the crystal grain and the direction of applied stress). This is a method of assessing the life using this standard deviation and the master curve prepared by the correlation with the degree of creep damage (Fig. 3.2.3-7).

(b) Material damaged by creep

99% reliable section 99% reliable section of creep damage ratio +/- 0.09

Regression curve

(a) New material

Freq

uenc

y Fr

eque

ncy

Direction of stress

Direction of stress

Maximum diameter

Crystal grain

Def

orm

atio

n co

unt S

m (d

egre

e)

[Relationship between deformation coefficient and creep damage ratio]

• Applied to the assessment of creep damage of Cr-Mo steel base metal

• Method of assessing the remaining life focusing on the fact that the crystal grain deforms as the creep damage grows

Deformation

coefficient Sm (standard deviation)

Formal distribution

Piping material (500-650°C)Heat transmission pipe material (570-600°C)

Creep damage ratio φc

Fig. 3.2.3-7 Assessment of creep damage to Cr-Mo steel base metal through the crystal grain deformation method

(i) Carbide structure-measuring method The base metal of low-alloy steel used for a boiler and the HAZ portion makes such structural changes as

deposition of carbide, condensation/large sizing, etc. as the creep grows. The structure of carbide also changes. At the initial stage of life, there is a lot of Cr-enriched carbide represented by M₂₃C₆. However, as the damage grows, it changes to Mo-enriched carbide such as M₆C. This method focuses on such structural changes of carbide. In this method, life is assessed using the master curve prepared in the correlation between the weight ratio of Mo/Cr and the degree of creep damage.

The Mo/Cr weight ratio is obtained by taking out very a small amount of specimen from the component subjected to assessment, extracting carbide by dissolving it in a suitable device, and measuring the weight of Cr and Mo by high-frequency plasma emission-analyzing apparatus. Figure 3.2.3-8 shows an example of the master curve of this method, which shows stress dependency. (j) Ultrasonic method

Upon the incidence of ultrasonic waves into the component, rear scatter noise is generated. Because the noise characteristics correspond to the number of generated voids and/or microscopic-cracks of the damaged component, it is quantified to specify this as a parameter to assess creep damage (noise value). Taking the noise wave after the

175

incidence of ultrasonic waves from the component surface to the 1st bottom echo, and by carrying out power spectrum analysis, the area within a certain frequency range is calculated to define it as the noise value. The assessment flow in the ultrasonic method is shown in Fig. 3.2.3-9.

σ: Application of reaction

Mo/

Cr w

eigh

t rat

io

Creep life ratio t/tr (Carbide structure-measuring method)

HAZ-reproduced component with SRsteel

Fig. 3.2.3-8 Carbide structure-measuring method

(b) Material damaged by creep

Data measurement of component for assessment

Pulse receiver Oscilloscope PC

Search unit

Component for assessment

Frequency analyzer

Noise analysis

1st bottom echo

(a) New material (unused)

Am

plitu

de (d

B)

Am

plitu

de (d

B)

Noise value

Frequency (MHz)

Frequency (MHz)

No fine cracks are detected.

Fine cracks aredetected.

Life assessment

• Comparison of noise between component for assessment and unused material

Certified curve

Life ratio

Noi

se v

alue

ratio

Noise value

Fig. 3.2.3-9 Life assessment flow in ultrasonic method

(k) Other methods to assess creep remaining life

In the remarks column of Attachment 3 as explained above, it is stated that the “Application of any other method than the above is permitted on the condition that it is recognized (by a committee with participation of people of experience or academic standing) to have accuracy equivalent to the above methods. Various methods other than the above have been developed. The names of such methods are mentioned below. a) Structure-quantifying method

The following 2 means are included: • M₆C deposition ratio

• Spheroidizing ratio of carbide

b) CMA (Computer-aided X-ray Microscopic-analyzer) Density spectrum method ② Fatigue damage (a) Microscopic-crack method (Replica method, MT copying method, etc.)

The methods for life assessment against creep damage as explained above are the methods of assessing the life against fatigue damage to the carbon steel, base metal, or HAZ portion.

If these components receive fatigue damage, microscopic-cracking of a level that can be observed by replicas only at the initial stage of fatigue life occurs, grows, and finally grows to a crack that can be detected by a non-destructive test such as PT, MT, etc. Therefore, the life can be assessed by detecting such microscopic-cracks by a replica. Figure 3.2.3-10 shows life assessment curves.

176

Magnetic powdercopying film

Low-alloy steel welding stop end

Magnetic powder

Oxidation scale

Magnetic field

Max

imum

leng

th o

f cra

ck m

easu

red

by

MT

copy

ing

met

hod

(mm

) M

axim

um le

ngth

of c

rack

mea

sure

d by

M

T co

pyin

g m

etho

d (m

m)

Life consumption ratio by the generation of macro-cracks (%)

Life consumption ratio by the generation of macro-cracks (%)

Carbon steel welding stop end

Average value curve

Average value curve

99% reliability curve

99% reliability curve

Crack detection boundary

Cra

ck

dete

ctio

n bo

unda

ry

Fig. 3.2.3-10 Microscopic-crack method (MT copying method)

3.2.4 Development and automation of inspection technology

With respect to the regular inspection of boiler equipment, the use of high efficiency, high-precision assessment devices has been required under such circumstances where the inspection process needs to be simplified, inspectors are getting old, 3K jobs (dirty, dangerous, and tough jobs) need to be eliminated, and damage needs to be quantitatively assessed corresponding to a requirement to rationalize aged boiler maintenance in line with the liberalization of electric power. In Table 3.2.4-1, various types of inspection methods and related automation are shown. (1) Type IV cracks of high-temperature thick wall pipe of large diameter

Cracks occurring to the boiler pipe header and to the welded portion of the thick wall main piping of large diameter are classified according to the locations of occurrence and are shown in Fig. 3.2.4-1. The cracks frequently experienced as creep damage are Type III and Type IV damage and are respectively characterized as damage to the rough-grain areas and fine-grain areas of the HAZ portion.

177

Table 3.2.4-1 Various types of inspection methods and related automation

Subject Damage to material Method of inspection and detection Automation

(1) Thermal fatigue of piping external metals

PT, MT UT from outside furnace

Furnace waterwall piping

(2) Inner piping corrosion fatigue

High-frequency array UT Spiral UT

Automatic inspection device using a multi-sensor within the furnace

(1) Creep Replica method, hardness method

Void recognition device by image processing

(2) Fatigue PT, MT Replica method (microscopic method)

(3) High-temperature corrosion, wear, and thickness decrease

Inner piping UT Automatic-measuring robot

(3) Steam oxidation scale High-precision UT method

Coil of super-heater, re-heater, and economizer

(4) Wear of horizontal heat-transferring piping

High-velocity laser method Inner piping UT UT thickness gage for narrow portion

Automatic inspection unit

Pipe header and main piping

(1) Type IV crack, inner crack

TOFD method Electronic focus sector scanning Ultrasonic noise method

Image-processing device

Type I: Crack in welded metal Type II: Crack classified as Type I, which has expanded from

the welded portion to the portion affected by heat (HAZ)

Type III: Damage to the rough-grain area of the portion affected by heat (HAZ)

Type IV: Damage from the fine-grain area of the portion affected by heat (HAZ) to the range of the partially transformed area

Welding material

Base material Base material

Transmitter Surface wave

Diffracted wave

Receiver

Crack

Direct reflection wave (same as conventional one)

Diffracted wave

Wave reflected on the bottom

Wave diffracted on the crack bottom

Wave diffracted on the crack top

Fig. 3.2.4-1

Classification of damage to a welded portion Fig. 3.2.4-2 Principles of TOFD method

Type III damage (damage in a rough-grain area) appears on the external surface of a pipe, whereas Type IV

damage (damage in a fine-grain area) occurs within a thick wall pipe and expands toward the surface. Impure substances contained in the steel play an important role in Type IV cracks. (2) Inspection method for Type IV cracks

Typical inspection methods for Type IV cracks occurring within a pipe having a thick wall are explained below. The inspection method is used alone or jointly with other methods. ① TOFD method

As an inspection technology able to assess Type IV cracks occurring from the inside of a thick wall precisely and quantitatively, the TOFD (Time of Flight Diffraction) method has been developed and put to practical use, which is an ultrasonic wave flaw detection method using 2 search units for transmission and receipt. A comparison with the conventional angle beam method is shown in Fig. 3.2.4-2.

The conventional method was in principle designed so as to catch reflecting echoes from a defect. Therefore, there were some cases where inspection was not possible depending on the direction of the crack. It was also

178

difficult to capture the defect size in a quantitative manner. On the other hand, the TOFD method catches the wave diffracted from the tip end of a crack. Therefore, it is

not affected by the direction of the defect. In addition, it can assess the length (depth) of a crack based on the transmission time of the diffracted wave. As a result, inspection in a precise and quantitative fashion has become possible. ② Electronic focus sector scan ultrasonic testing

The principles of measurement by ELFOSS UT are shown in Fig. 3.2.4-3. This device can perform wide-angle scanning by focusing an ultrasonic wave beam through the delay circuit to improve resolution and defect inspection accuracy. Two search units are used for the TOFD method, whereas this device has such a characteristic that inspection of the narrow portion is made possible because wide-angle flaw detection is performed by only 1 search unit.

Trigger pulse for activation

Angle of deflection

Delay circuit

Vibrator

Focus Electronic focusing by

delay circuit Sector scan by delay

circuit Electronic focus sector

scan If the activation timing of

the vibrator is changed with the

same interval in the right and left directions, an

ultrasonic wave beam focuses. In addition, the focal

depth can be freely set by the duration of

the timing.

If the activation timing of the vibrator is changed at

the same interval, the ultrasonic wave beam is deflected. In addition, the deflection angle can be

freely set by the duration of the timing.

If the timing and duration of the activation of the

vibrator is changed from time to time, the direction

can be changed continuously by focusing an ultrasonic wave beam.

Fig. 3.2.4-3 Principles of ELFOSS UT

③ Ultrasonic noise method As explained in the section on the method of assessing remaining life pertaining to creep damage, such features

as noise intensity rises in the case of a material with voids or microscopic-cracks being utilized in the ultrasonic noise method. Early assessment has become possible for Type IV damage occurring within the welded joints of high-temperature thick wall pipes of large diameter. In addition, by scanning the search unit in the right-angle direction against the weld line and by installing a time gate in the direction of plate thickness, map images of the damaged portion can be obtained through divided measurements as shown in Fig. 3.2.4-4.

179

Plate thickness direction (mm)

Scan a search unit and apply gate by splitting the corresponding time width.

Clarification of the points of damage for image processing

Scanning

Portion affected by weld heat

Deposited metal portion

Base material

Base material Base material

Direction to move search unit (mm)

Example of flaw detection result

The noise value is displayed on a color map.

Deposited metal portion



Tim

e sp

lit g

ate

Fig. 3.2.4.-4 Image processing of flaw detection results through the ultrasonic noise method

3.2.5 Chemical cleaning (1) Purpose and timing of chemical cleaning ① Purpose of chemical cleaning

The purpose of carrying out chemical cleaning of boiler equipment is to remove any and all foreign materials and scale adhering to inner face of the evaporation piping during construction or operation of the boiler, thereby preventing any problems from occurring to the boiler, to recover its efficiency and maintain it under good conditions.

The purpose of chemical cleaning performed during the construction of boiler equipment is to remove any and all mill scale adhered during the manufacture of the boiler piping and fat and oil adhered during installation, to remove any foreign materials entered such as sand, etc., and to prevent any problems from occurring during operation thereafter.

Although impure substances brought into the boiler equipment when installing a condensate demineralizer or improving water treatment are reduced, these substances still remain as scale adhering to the inner piping due to the following causes: (a) Intrusion of corrosive substances through the water supply system and their adherence to the water supply

system

(b) Condensation and deposition of dissolved salts

(c) Corrosion of the materials of the boiler piping

Such impure substances cause overheating of the piping materials, generation of scaling, formation of local cells, or corrosion due to condensed salts and lead to future swelling out or explosion of the piping.

As shown in Table 3.2.5-1, the thermal conductivity of scale largely varies depending on its chemical ingredients. Because the size of scale is smaller than that of piping materials, adhered scale blocks thermal conduction causing overheating or heat loss of piping materials.

Table 3.2.5-1 Thermal conductivity of metal and scale Type Thermal conductivity (W/m・K)

Mild steel 45 ~ 70 Scale containing silicate as its major ingredient 0.2 ~ 0.5 Scale containing iron oxide as its major ingredient 0.9 ~ 2.3 Fat and oil 0.1 Water 0.6

The water vapor oxidized scale generated in the steam system peels off during operation and accumulates in the

U-shape pipe of the super-heater piping, resulting in its explosion. Its fragments may fly over to the turbine and damage the blade.

180

② Timing of chemical cleaning For the timing of chemical cleaning after the start of operation of boiler equipment, the boiler manufacturer

specifies the standards of cleaning depending on the amount and thickness of adhered scale. On the other hand, the operators at electric power companies also specify their own respective standards. The standards commonly used for cleaning are shown in Table 3.2.5-2. The value mentioned there is only a general guideline. Therefore, it is desirable if an independent cleaning timing is established. In addition, this value should be determined based on the portion where the maximum amount of scale adheres to individual boiler equipment. Full care should be paid to any change in the portion where the maximum amount of scale adhered due to a change in the boiler operation method or fuel change.

Table 3.2.5-2 Amount and thickness of adhered scale for which chemical cleaning is required Normal pressure

Type

8Mpa class 12Mpa class 18Mpa class Beyond boundary pressure

⎯ 90 ~ 135 75 ~ 105 Coal-fired boiler

⎯ 400 ~ 450 250 ~ 350 ⎯

90 ~ 120 75 ~ 105 60 ~ 90 Coal/oil mixture-fired

boiler 300 ~ 400 250 ~ 350 200 ~ 300 ⎯

75 ~ 105 60 ~ 90 45 ~ 75 24 ~ 36 Oil-fired boiler

250 ~ 350 200 ~ 300 150 ~ 250 80 ~ 120 Gas-fired boiler Same as above Same as above Same as above Same as above

Note 1) The upper row in each column indicates the amount of adhered scale (mg/cm²), the and lower row indicates the scale thickness (µm).

Note 2) The amount of adhered scale is the value at the flame side (180°) of the inner evaporation piping. Note 3) The amount of a once-through boiler of 18 Mpa class or smaller shall be 2/3 of the value shown in above table. Note 4) Even if the actual values are less than above, it is recommended to carry out chemical cleaning when the boiler has been

operated for 50,000 hours or longer. (2) Nature of scale

The scale adhered to new boiler equipment is mostly mill scale (magnetite: Fe₃O₄) generated during the course of pipe manufacturing. On the other hand, the nature of scale largely varies depending on the quality of the supply water or refill water, treatment of the supply water or boiler water or materials of low-/the high-pressure supply water heat transmission piping of a heater between the steam condenser and the boiler. Even with the same boiler, the amount and ingredients of scale vary depending on the sampling position or whether it is the flame side/furnace material side. Table 3.2.-3 shows examples of analyzed scale ingredients in the evaporation piping and steam system. Its characteristics are outlined below: (a) With respect to boilers A, B, and C using a heater of copper alloy steel in the water supply system, the scale

contains copper.

(b) With respect to boilers B and C, such refractory scale as white ZnAl₂O₄ (zinc aluminate) or NiFe₂O₄ (nickel ferrite) called spinel scale may be generated if zinc (Zn), Aluminum (Al), and/or nickel (Ni) is contained.

(c) The scale of boilers D, E, and F using a heater for the steel piping in the water supply system is mostly iron oxide (Fe₃O₄).

(d) Boilers D, E, and F are of same high-pressure, once-through type, but the boiler water treatment for boilers D and E is AVT (all volatile treatment) to remove Fe₃O₄ (magnetite). As shown in Photo 3.2.5-1, the scale has a corrugated surface.

181

Table 3.2.5-3 Examples of chemically analyzed scale ingredients Average adhering amount

Chemical content

Boiler Pipe specimen (mg/cm2)

Fe3S

O4

Cu

ZnO

Al 2O

3

NiO

MgO

CaO

P 2O

5

Cr 2

O3

MoO

MnO

Ref

ract

ory

by a

cid

A Right-side wall pipe 65.3 58.3 1.9 1.1 <0.1 <0.1 10.3 13.3 10.6 - - - 2.1 B Front wall pipe 25.4 33.0 34.5 15.1 0.7 14.5 <0.1 <0.1 0.3 - - - 0.2 C Front wall pipe 20.4 73.0 2.8 10.0 0.9 <0.1 0.8 1.7 4.9 - - - 1.9 D Front wall pipe 24.1 97.5 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 - 1.8 - 0.7 <0.1E Front wall pipe 23.4 97.9 <0.1 <0.1 <0.1 <0.1 - - - - - - 0.7 F Front wall pipe 9.6 97.9 <0.1 <0.1 <0.1 <0.1 <0.1 0.2 - - - 0.5 <0.1G Secondary

super-heater 38.4 65.9 <0.1 <0.1 - 11.3 - - <0.1 17.8 1.5 1.7 0.3

H Re-heater 125.0 95.4 <0.1 <0.02 <0.1 <0.2 - <0.4 - 1.8 0.9 0.4 1.3 I Main steam pipe 125.3 88.1 <0.5 <0.5 <0.5 <0.5 <0.5 <0.5 <0.5 3.3 0.8 <0.5 -

Note 1) A: Boiler for own power generation (VU-60) 6.8 MPa 60 t/h B: Forced circulation boiler (Mitsubishi) 19.2 MPa 860 t/h C: Natural circulation boiler (Hitachi) 17.2 MPa 1,135 t/h D: Once-through boiler 26.3 MPa 1,640 t/h E: Pressure-variable once-through boiler (AVT) 25.0 MPa 2,300 t/h F: Pressure-variable once-through boiler (CWT) 25.0 MPa 2,300 t/h G: Once-through boiler (super-heater piping: SUS316HTB) 26.9 MPa 1,500 t/h H: Pressure-variable once-through boiler (re-heater piping:STBA24) 25.0 MPa 2,300 t/h I: Once-through boiler (main steam piping:STBA24) 25.0 MPa 1,900 t/h

Note 2) A –F: Adhered amount on flame side G – I: Adhered amount around entire circumference

Inner layer

Outer layer

Base material

Photo 3.2.5-1: Corrugated scale of AVT treatment

boiler Photo 3.2.5-2: Steam-oxidized scale

(x 100 magnification)

With respect to boiler E, fine-grain Fe₂O₃ from CWT (combined water treatment) adheres to the magnetite, and the scale has smooth surface. (e) Boilers G, H, and I generate vapor-type scale. Cr-Mo steel (low-alloy steel) has been used for these

boilers. Two-layer scale, called steam-oxidized scale; one in the neighborhood of piping materials with a high content of chromium and the other at steam side with a high content of iron oxide are generated as shown in Photo 3.2.5-2.

182

Table 3.2.5-4 Operation of a boiler and required cleaning process

Flus

hing

Cle

anin

g w

ith

amm

onia

Cle

anin

g by

de

grea

sing

Was

hing

with

w

ater

Cle

anin

g w

ith

acid

Was

hing

with

w

ater

Prev

entio

n of

ru

st b

y ne

utra

lizat

ion

Fina

l was

hing

w

ith w

ater

During construction ⎯ Copper content: high ⎯ Copper content: low ⎯ ⎯ ⎯ After

operation Copper content: none ⎯ ⎯ ⎯

Remarks : Implement. Implement if necessary. (3) Cleaning method ① Cleaning process

The cleaning method varies depending on the operation of the boiler equipment or the ingredients of the scale. Typical cleaning processes are shown in Table 3.2.5-4. (a) Boiler during construction

The main purpose of cleaning is to remove mill scale, oil and fat, and/or foreign materials. In these years, the degreasing process is mostly omitted by adding degreasing agent during acid cleaning. (b) Boiler after operation

i) Starting from acid cleaning of the scale mainly containing ferrous, hardening ingredients and/or a small amount of copper, copper-dissolving/-enclosing agent is added during acid cleaning if copper is contained.

ii) Before acid cleaning, ammonia cleaning is performed as pretreatment in order to dissolve the copper content.

In lieu of the above i) and ii) cleaning, chelating cleaning is sometimes carried out. Its cleaning process is:

Ferrous removal cooling copper removal/rust prevention washing with water

(4) Planning and implementation of cleaning Planning of cleaning includes understanding the overall structure of the subject boiler, studying the cleaning

specifications through investigation of scale, selection of a method of treating wastewater, and planning the implementation method. The planning procedures are shown in Fig. 3.2.5-1.

183

Investigation of the subject of cleaning

Specifications and materials of the unit State of evaporation amount, pressure,

operation hours, etc. State of water treatment control such as

quality of supply water, quality of boiler water, chemicals used, etc.

Practical experiences in cleaning Type of fuel

Scale ingredients Amount of adhered scale Scale generation rate Deterioration level of the material

Scale dissolution test Material deterioration test Investigation of customer’s

environmental conditions such as wastewater standards, etc.

Availability of wastewater treatment equipment at customer side

Experimental wastewater treatment

Study of wastewater treatment system

Dissolution test

Planning of cleaning specifications and requirements

Planning of cleaning process Planning of treatment of wastewater and

exhaust gas Approval of power source, water supply

source, heat source, etc. Preparation of cleaning flow and work

procedures Checking of the safety and sanitary level

related to construction and training

Visual inspection Amount of corrosion to be checked

by a test piece Amount of scale removed by

cleaning

Report of cleaning implemented Advice regarding maintenance and

water treatment method Checking of operation after cleaning

Implementation

Inspection

Summary

Investigation of scale

Fig. 3.2.5-1 Flow of chemical cleaning planning for boilers

① Guideline for implementation

Items to be considered when planning a guideline for implementation are given below:

(a) Outline of the unit subjected to cleaning

(b) Scope of cleaning and amount of cleanser

(c) Handling of components not subjected to cleaning

(d) Relationship between actual construction and temporary construction, size, quantity

(e) Types of cleaning chemicals, density and cleaning conditions

(f) Cleaning process and criteria for determining the completion of cleaning

(g) Method of checking and inspecting the effect of cleaning

(h) Method of receiving waste cleaning fluid and procedures to treat it

(i) Utilities (pure water, steam, power, air, etc.)

(j) Flow of each process, temporary storing place, and piping route

(k) Outline and detailed process

An example of the cleaning system for typical-type boiler equipment is shown in Fig. 3.2.5-2.

184

Actual construction line Temporary construction

Pressure gage

Flow meter

Sampling

Thermometer

Mixing header

Mixing heater

N2 gas W

ater

su

pply

line

Steam drum

Sid

e w

all

Sid

e w

all

Fron

t/rea

r w

alls

Inspection nipple

Temporary level gage

Circulation pump

Blower

Draw pump

Hydrazine pump

Ejector

Tank

Chemicals injection pump

Pure water

Steam

Fig. 3.2.5-2 Flow of cleaning system of the natural circulation-type boiler

Main steam piping

Pure water

Cage

Steam separator

Evaporator

Ceiling wall

Cold water

Economizer

To blow line

Actual construction line

Temporary construction line

Level gage

Pressure gage

Flow meter

Thermometer

Sampling

Stea

m

sepa

ratio

nta

nk

SH Water-filling pump

Main closing valve of turbine

N2H4 tank

N2H4 pump

Water-sealing pump for the components not subjected to cleaning

Test piece seat

Mixing heater

Circulation pump

High-pressure supply water super-heater

Main supply water piping

Blow line Steam

EjectorChemicals

tank

Chemicals injection pump

From tank-lorry

Fig. 3.2.5-3 Flow of cleaning a once-through boiler

185

3.2.6 Circulation Pump (1) Preventive maintenance of circulation pump

Circulation pumps for boilers have been employed for boiler equipment having a capacity of 150 MW or more since around 1955. The circulation pump is divided into the injection type and the glandless type (canned motor type, submerged motor type). Currently, about 400 units of these 2 types of pumps are operated for domestic thermal power generation. Many non-conformance events occurred at the initial stage of introduction. As a result of structural improvement and completion of the details for inspection items thereafter, such non-conformance events have been drastically reduced and the reliability has been largely improved. However, 30 years have already passed since the installation of some circulation pumps as shown in Fig. 3.2.6-1. Some of them are being replaced gradually, but more than half of them have been used for 15 years or longer. The preventive maintenance of such units has become a critical issue. (The descriptions from the next section are examples of circulation pumps made by Fuji Electric.)

25 ~ 29years

Num

ber o

f del

iver

ed u

nits

(uni

t)

Total number of units

30 years orlonger

20 ~ 24years

15 ~ 19years

10 ~ 14years

5 ~ 9years

0 ~ 4years

Fig. 3.2.6-1 Years of operation after delivery

Pump case (Renewal cycle: 35 – 40 years)

Generation of cracks

Impeller (Renewal cycle: 25 – 30 years) Generation of

cracks Abnormal vibration

Heat exchanger (Renewal cycle: 35 – 40 years)

Accumulation of scale

Fatigue/corrosion of welded portion

Cavity temperature rise

Water leakage Rotor (Renewal cycle: 25 – 30 years) Popping out of

rotor bar Corrosion/wear of

steel core

Deflection of ammeter

Abnormal sound → Abnormal vibration

Renewal cycle of thrust bearing Thrust plate: 8 – 12 years

Pad: 16 – 20 years Abnormal wear Lift/peeling off of bearing material

Abnormal sound/abnormal

vibration

Renewal cycle of journal bearing Sleeve plate: 8 – 12 years

Pad: 16 – 20 years Abnormal wear Lift/peeling off of bearing material

Abnormal sound/abnormal

vibration

Motor case (Renewal cycle: 35 – 40 years)

Expansion of in-low clearance Deformation of gasket Uneven tightening Warming shortage Overlapping of thermal insulation materials

Abnormal sound → Abnormal vibration Steam leakage, water leakage Cavity abnormal temperature rise

Stator (Renewal cycle: 25 – 30 years)Wear of press ring

Loosening, dislocation,

corrosion, or wear of steel core

Abnormal sound → Abnormal vibrationShortened life of

coil

Renewal cycle of coil winding PVC: 8 years

XLPE: 12 – 16 years Wear of coil wire • Slipping down of coil • Loosened cleat wire • Deterioration of

insulation materials Insulation drop →

Ground fault/unstable life

Fig. 3.2.6-2 Deterioration of main parts and renewal cycle

186

① Non-conformance events As explained above, non-conformance events have been reduced to date, and the reliability of the circulation

pump of boiler has been largely improved. However, there still remain many plants for which no structural improvement has been implemented so far. It is required therefore to recheck the non-conformance events in the past and reflect their results in the completion of details of inspection items and on the plan for repair. Non-conformance events of major parts are outlined in Fig. 3.2.6-2, which shows the deterioration phenomena and renewal cycle of major parts (renewal cycle with addition of the effect of the bearing force improvement plan to the past experiences). ② Guideline for implementation of preventive maintenance and inspection

Inspection items are divided into “general inspection items” and “special inspection items.” General inspection mainly involves visual inspection, whereas non-destructive testing is the main item of special inspection, which should be started from the 10th year after the start of operation to obtain remaining life assessment data. Concretely, the target regular inspection cycle should be set at 4 years, and a long-term plan of “details for checking/inspection items” and “details for repair items ”should be developed Items to be implemented should be confirmed at the start of the respective regular inspection. ③ Concept of measures for improvement of bearing force and examples of implementation

Measures for improvement of bearing force of the circulation pump of boiler equipment are promoted under the 2 concepts below, aiming to respond to any change in the operation method of power generation plants (conversion to WSS/DSS), extension of the inspection cycle, and prolongation of operation life:

(a) Improvement of structure, materials, and work method

(b) Enrichment of inspection items (early detection of non-conformance and early countermeasures)

Typical examples of implementation are shown below: i) Forged pump case

The conventional pump case was a cast product of the volute type. As a measure to improve the bearing force of the pump case, a spherical-shape forged pump case has been employed for about 15 years. Compared with the volute-type cast pump case, the spherical-shape forged pump case is simple in its configuration and the reliability of its materials is high. It is suitable for a plant with frequent start/stop operations in a high-temperature, high-pressure atmosphere (plants using DSS, etc.)

ii) The motor stator coil has been changed to cross-linked polyethylene wire.

Coils manufactured before 1980 were made of PVC wire, which involved the issue that the rewinding cycle was short because hardening/fragility of the insulation coat was accelerated due to reduction of the plasticizer.

iii) Employment of a single-basket-shaped stator of the closed slit type

The double-basket-shaped stator of the open slit type was used as a standard stator in the past. DSS operation (repeated transient vibration torque and/or thermal stress at the start of operation) was not considered in its structure. As a measure for DDS operation, a single rotor of the closed slit type has been employed.

iv) Implementation of special precise inspection

Visual inspection is more than enough for the initial stage of plant operation (within 10 years). However, after 10 years when the renewal cycle timing of parts approaches, special precise inspection mainly composed of non-destructive testing is carried out in addition to visual inspection. Through early detection of and early action against any non-conformance by determining the timing of renewal, the life of parts can be prolonged.

187

3.2.7 Fan (1) Measures to improve reliability and guideline for maintenance and inspection

The present time is called a maintenance age. The number of social systems and production systems subjected to maintenance are accumulating at a continuously increasing speed. According to a certain trial calculation, the ratio of costs for maintenance was 10% of social capital investment during the 1970s, whereas it increased to 30% during the 1990s and to 50% by 2020. Under these circumstances, maintenance costs keep increasing; how to cope with this in a quantitative manner, how to improve cost efficiency keeping improvement of reliability, and how to select the type of acceptable maintenance have become serious issues.

Because the fans installed at power generation plants are kept in operation for a long period of time from the start of operation until the time they are disposed of, the accumulated number of units has been increasing. It is required to make clear what inspection items are to be applied to these fans and to implement them under a controlled cycle and implement feedback and feedforward without any reserve. Because efficiency and rationalization of maintenance costs is directly linked to the management, it is required to develop a general image of maintenance, determine what is presently missing, and implement these items in a well-planned manner. ① Axial fan

As boiler capacity becomes larger, the rotating-type variable axial fan suitable for large-capacity boilers with reduced power consumption under partial load has been widely used as a ventilating fan for power generation equipment other than FDF, IDF, PAF, BUF, and high-temperature GRF. Control for improvement of reliability is further required, because the structure of the rotating-type variable mechanism is complex and the number of parts is larger than the same of the centrifugal fan.

As a result of measures taken for the improvement of reliability [1] with consideration paid to the problems with axial FDF experienced over a period of 15 years since 1970, the employment of axial fans started, and the problem occurrence ratio has been suppressed to its minimum. However, in view of the facts that the installation of axial fans increased from 1985 onward when many thermal power generation plants were constructed, and that its usage has expanded, it is desirable to carry out precise inspection of fans used for many years in order to further secure their reliability. ② Centrifugal fan

Although the reliability of centrifugal fan has been improved, many fans have already been in use for 20 years or longer. It is required to plan and implement measures to improve their reliability further taking into account any aged deterioration or any change in operation from what was expected at the start of operation.

Because the operation of thermal power generation plants corresponds to the peak power generation capacity, the number of start/stop operations has increased, which, as a result, requires the improvement of the bearing force of impellers, bearings, and couplings.

(a) Stress change occurs at the impeller caused by the change in RPM due to start/stop of operations. Especially with GRF, low-cycle fatigue occurs due to repeated thermal expansion caused by temperature fluctuation. If you start the operation of GRF at room temperature, the temperature of the intake gas rapidly changes and the vibration becomes several times larger for some time than the vibration experienced under stable, steady operation. This is an effect of the difference in thermal expansion caused by the temperature difference among the components of the impeller. When the temperature of the impeller becomes stable after continuing operation in a stable gas temperature atmosphere, the amplification of vibration gradually lowers and the operation becomes stable. In particular, when a riveted joint is used, this phenomenon frequently appears. Therefore, if a riveted structure has been used for the impeller, it is recommended to change it to a welded structure and remodel the connection of the impeller to the shaft/hub to a reaming bolt connection structure from the rivet-fixed type. Because the effect of thermal distortion concentrates on the riveted structure, non-destructive testing needs to be carried out for the components concerned when the fan is not in use or regular inspection is carried out. In the case of the structure of the axis–boss shrink fit, any vibration that may be caused by the decrease in the shrink-fit margin or loosening due to the transitional difference in temperature distribution is of concern. It may be required to increase the shrink-fit margin or change to an integrated rotor of the axis–boss.

If the level of adherence of the mating portion of the axis–boss shrink-fit structure changes as the time passes, that the vibration may become stronger or the torque transmission ability may drop are concerns. Ultrasonic waves can be used to test the level of such adherence. Figure 3.2.7-1 shows the inspection principles when a clearance is available for testing.

(b) Any fatigue damage that occurs to the face of the tooth at the gear coupling due to start/stop operations is also a concern. Complete inspection is required. It is recommended to change to a tooth face with improved bearing force or to a flexible coupling having no contact with the face of the tooth.

188

(c) Stress occurring at the impeller is strong. When carrying out non-destructive testing at regular inspection, such a case is found where the portions and number of occurrences of damage increase as time passes. In case there is concern that complete reliability may not be secured through regular inspection or repair only, it is required to change to an impeller of a type whose generated stress has been reduced by increasing its wall thickness or improving its welding quality.

③ Precise inspection of large-sized fan Large-sized fans are disassembled and maintained at each regular inspection. Items subjected to precise

inspection of the respective parts of the centrifugal fan that can be implemented for such aged deterioration phenomena as corrosion, wear, cracking, etc. are shown in Table 1. Because problems with large-sized fans can lead to operation stop of the unit or to load limit, it is recommended to carry out full assessment at respective regular inspection, etc.

The fan is equipped with attachment devices other than the main unit such as the lubricating device, silencer, measuring apparatus, etc. It is required to secure the reliability of these devices as well as securing the reliability of the main unit. For inspection of the main unit, disassembling, which requires many processes, is necessary. Because fewer processes are required for disassembling inspection of attachment devices, it is recommended to carry out regular maintenance once a year.

Assess the output of the echo from the hub bottom (Bn) and from the shaft bottom (W).

Impeller

Impeller hub

Shaft

Mating portion

Hub

Sensor

Shaft

Tran

smitt

ed

wav

e

Fig. 3.2.7-1 Assessment of the degree of adherence of hub/shaft

(2) Cause of life consumption

As the causes of consumption of life of the fan, corrosion, wear, fatigue, etc. can be mentioned. Because power generation plants are located near the sea, corrosion caused by salt needs to be taken into

consideration. For the intake of atmospheric air by FDF and PAF, it is required to assess the strength of the silencer against corrosion. Caution is required to be paid to the pit generation of aluminum alloy used for the rotating blade of the axial fan caused by salt corrosion and clogging created between slide clearances. With respect to IDF and BUF, because drain with strong corrosive features is generated when the moisture contained in the gas condenses while the gas temperature drops when the fan is not in use, it is required to make assessment in this respect.

With respect to wear, there is a record of a survey conducted in USA. As a result of a large-scale survey to clarify the cause of problems conducted by EPRI ⁽²⁾ ⁽³⁾ in order to improve the reliability of coal-fired thermal power generation plants, it was found that IDF was one of the most serious causes for drops in operation efficiency.

The main cause of problems with IDF was wear caused by the fly ash contained in the exhaust gas. Researches of the following items are presently under way in order to improve the wear resistance of IDF:

(a) Characteristics and level of wear of the fan at power generation plants and related costs required for countermeasures against it

(b) Improvement of computer models to estimate the wear damage to the fan

(c) Assessment of the effect of relative wear by various types of fly ash 189

(d) Assessment of the cost for the armor system of the blade-shaped centrifugal fan that can be replaced on site

The researches are mainly focused on the centrifugal blade-type fan, which cannot in most cases be applied directly to the axial fan, which is the mainstream in Japan.

Axial-type IDF, many of which have been introduced in Japan from around 1985, have already been used for 10 years or longer. It is considered that such study will become necessary as the same EPRI conducted for wear.

Fatigue is divided into low-cycle fatigue caused by start/stop operations and high-cycle fatigue occurring during normal operation. It is required to fully assess the rotating blade of the axial fan because damage to it is highly expected. A study is required to be conducted for low-cycle fatigue if the frequency of start/stop operations has increased in the course of the change in the operation method to more than when the unit was initially installed. 3.2.8 Corrosion of boiler equipment occurring in its water zone and countermeasures against it

Introduction

The purposes of controlling thermal power generation plants by establishing a reference value for each item of water supply, boiler water, and steam is to prevent any problems from occurring to the equipment composing the thermal power generation plant caused by corrosion and/or scale due to the quality of water used and to continue the operation of the plant in a safe and smooth manner. As the pressure and temperature of the main steam rise higher, the thermal efficiency of the plant becomes higher. However, the plant is likely to be affected by corrosion or scale, and the level of such effect becomes higher. Therefore, water quality control is an important task that affects the thermal efficiency and operation efficiency of the unit.

While the water treatment engineering of boiler equipment has remarkably advanced in these years, accidents often occur from thermal power generation plants caused by the water used by aged equipment or DDS operation. Those staff responsible for water quality and the staff in charge of operation and maintenance of the plant are required to understand the importance of water quality control and endeavor to improve it.

Problems arising from water are roughly divided into issues of corrosion, fragility, (cracking) and scale. As shown in Fig. 3.2.8, most of the problems relating to water occur when multiple causes are combined. Upon occurrence of any problem, its cause must be analyzed and assessed in detail to establish adequate countermeasures.

An outline of various types of problem and their causes, handling, and preventive measures is given below.

Attack by ammonia Erosion of turbine Clogging Rise in differential

pressure Thermal conduction

was blocked Alkali corrosion Fragile crack caused by

hydrogen Crack caused by stress

(①

oewth

Defective design and construction Inadequate materials Defective design of orifice Clogging with foreign

materials Uneven thermal load

Defective operation maintenance

Defective storage Defective water treatment Defective control of

combustion

Fig. 3.2.8 Problems and related causes

1) Problem caused by adhered scale and countermeasures against it Problem caused by overheating In the period in which raw water was used for refilling, hard contents contained in the raw w

n the evaporation unit as white scale of calcium carbonate, which caused overheatinvaporation piping due to its thermal resistance. Currently, due to advanced technology in the ater, dissolved contents from the materials in the condensed water supply system change toe evaporation unit. The main ingredients of the scale are magnetite (Fe₃O₄), copper, etc. By carrying out chem

190

Drop in efficiency Opening by swelling-out breakage

Oxidization of steam Insufficient flow rate Adherence of scale Corrosion of entire unitCorrosion of partial unitCarry over Leakage of seawater

ater were deposited g problems of the manufacture of pure scale and adhere to

ical cleaning of the

boiler equipment at an adequate timing, it is very seldom that the evaporation piping is damaged by overheating due to the thermal resistance of the scale itself.

Cross section of scale

Appearance of the portion of leakage

Photo 3.2.8-1 Example 1 of problem caused by overheating due to adherence of scale The causes of problems by overheating due to scale adhering to the evaporation piping occurring in these years

are considered to be as follows: (a) Due to improper water control, very soft magnetite scale is generated and grows to form a steam layer in

the clearances among the scale layers.

(b) A steam layer is formed in the portion in which the scale has been peeled off from the steel face and lifted due to the temperature fluctuation caused by start/stop operations of the boiler equipment under a condition where a relatively large amount of scale has adhered.

(c) If the amount of Cu, ZnO, CaO, etc. has become very large within a given scale layer when the composition of scale largely fluctuates due to the change in quality of the supply water, the scale is peeled off from that portion, film boiling occurs there, and a steam layer is formed as a result.

(d) When any scale remains in the chemical cleaning process of boiler equipment and any clearance is created between the piping materials and the scale, that portion becomes a hot spot and a steam layer is formed there. Almost all of these problems occur after the operation of the plant has started.

Photo 3.2.8-1 shows an example in which the scale has swelled out and broken open in an oval shape within the furnace of the evaporation piping (STB42) located on the upper side of the burner. This is a case where heat conduction is blocked when soft-type scale (200 – 250 μm) has adhered to the inner face of the piping, peeled off within the layers, and lifted and opened due to the excessive rise in the metal temperature of the piping. In the area surrounding the opening, many cracks are generated in the pipe shaft direction. As countermeasures against this, the generation of soft-type scale is suppressed by the removal of scale through chemical cleaning, reduction of melted oxygen in the condensed water and in the drain system of the low-pressure supply water heater, deoxidization at the time of starting operation, etc.

Cross section of scaleAppearance of the portion

of leakage

[Metal]

(Scale thickness 0.33 – 0.49 mm)

Photo 3.2.8-2 Example 2 of problem caused by overheating due to adherence of scale

Photo 3.2.8-2 is an example of a case the scale was overheated, swelled out, and opened within a very short

period of time (creep breakage in a short time); as a result, the unit was operated under such a condition that the amount of scale adhered to the inner piping exceeded the amount for which chemical cleaning was required (thickness 450 μm, amount of adherence 85 mg/cm²), the scale layers adhered to the inner piping were peeled

191

off and lifted, and heat conduction was blocked by the steam layers generated between the scale layers. As countermeasures against this, it is required to capture the level of scale growth by regular pipe sampling inspection and determine the adequate timing of chemical cleaning. ② Corrugated scale

At a plant where volatile matter treatment is undertaken as a method of treating supply water, there are many experiences where the scale adhered to the inner evaporation piping of a furnace shows a corrugated pattern. Especially with respect to the supercritical sliding-pressure once-through boiler, the average rate of flow in the piping becomes higher. Therefore, scale with this corrugated appearance increases the break-through resistance of the furnace, which may cause problems in operation. The cause of the generation of such corrugated scale has not yet been clarified. The scale is considered to be generated under such a condition that chemical factors and fluid dynamics factors have been combined. Namely, dissolution and deposition of the component materials in a high-temperature, high-pressure atmosphere as chemical factors and cyclic structural change of turbulent boundary layers as fluid dynamic factors are considered combined, whereby such corrugated scale was generated.

Photo 3.2.8-3 shows an example of the corrugated scale generated within a supercritical sliding-pressure once-through boiler. In this case, the amount of adhered scale is not so great that chemical cleaning is required, but problems in operation have occurred because the break-through resistance became stronger due to the shape of such scale. As countermeasures, the scale is removed by chemical cleaning in order to reduce the break-through resistance. Thereafter, it was clarified that the generation of such corrugated scale could be suppressed by changing the supply water treatment to oxygen treatment, according to certain European literature ⁽⁹⁾⁽¹º⁾and the test results of oxygen treatment verification carried out in Japan ⁽¹²⁾. This oxygen treatment method has the advantage of a reduction in running costs, including the prevention of such corrugated scale from being generated. Therefore, this oxygen treatment method is currently being rapidly introduced to once-through boilers in Japan.

Adhered scale (inside of furnace)

Direction of flow

Photo 3.2.8-3 Adherence of corrugated scale

③ Scale adhering to the components

There is such a case where an increase in break-through resistance and fault movement is caused by the considerable amount of magnetite scale partially adhering to such components as the orifice for flow rate adjustment at the inlet of the evaporation piping of the forced circulation boiler, the spray water control valves of the super-heater and re-heater, the drain control valve of the supply water heater, the flow meter for the supply

192

water (flow nozzle), and the high-pressure supply water system (strainer of the water supply pump, rectifying cylinder of the high-pressure supply water heater, heater piping), etc. This scale adheres to portions where there is no thermal load, which however is present in the evaporation piping. It is considered that the adherence of scale is a phenomenon that occurs when chemical factors, fluid dynamic factors, and static electric factors (charged grains) are combined.

Iron

conc

entra

tion

(µg/

l)

Temperature (°C)

Fig. 3.2.8-1 Solubility curve of magnetite

Because the main ingredient in the chemical factors is magnetite and scale is generated at portions with such high temperatures as 180 or more, and as one can reason by analogy from the solubility curve ⁽¹²⁾ in Fig. 3.2.8-1, the portion where scale has adhered becomes oversaturated by the degree of solution of magnetite under such temperatures and becomes an area where fine grains of magnetite are created. As fluid dynamic factors, the scale has adhered to the portion whose boundary layers proximate to the metal surface are thinner than other portions in the high rate of flow in the area in which the flow path has become narrower. This indicates that the scale adheres to such portions with high probability of the created magnetite fine grains colliding with the metal surface. As static electrical factors, when such oxide as magnetite is submerged into water, the surface of the oxide is charged and comes to have electrical potential (zeta potential) by certain type of static electrical phenomenon. The intensity level of this electrical potential is related to the characteristics of the grain surface. If the grain size becomes smaller, the characteristics of the surface become stronger. Namely, the activity of the surface becomes especially strong immediately after the fine grains of magnetite are created. Because scale is generated to such portions where the various factors above are combined, the scale does not always adhere to the same portions of similar plants.

Photo 3.2.8-4 shows an example of a unit that has become uncontrollable due to adhered and solidified magnetite scale in the high-velocity portion of the stem throttle of control valve for the spray water of the super-heater.

Photo 3.2.8-5 shows an example of scale containing copper as its main ingredient selectively adhered and solidified at the orifice inlet of the water drum where the flow rate has been reduced. In this case, the copper content dissolved from the supply water heater equipped with copper alloy piping due to a failure in the supply water treatment was brought into the boiler and selectively adhered to the orifice.

193

Photo 3.2.8-4 Example of scale adhered to control valve

Scale Orifice

Photo 3.2.8-5 Example of scale adhered to orifice

Such a failure in supply water treatment can be avoided by improving the treatment system. The adherence of

magnetite scale as mentioned above occurs even in such area where supply water treatment has been carried out properly. Even by changing the conditions of the portion to which the scale has adhered (for example, change in the pH, hydrazine density, etc.), only the adhering portion changes its location to some extent, and it does not lead to any satisfactory solution. As a measure to resolve this issue of scale adherence, oxygen treatment, which has been employed as a countermeasure against corrugated scale, is effective.

Photo 3.2.8-6 shows an example of improvement for the removal of magnetite scale adhered to the rectifying cylinder of a high-pressure supply water heater at a power generation plant where oxygen treatment has been adopted. Such problems as efficiency drop, vibration, etc. caused by magnetite scale adhered to the impeller of the water supply pump have also been resolved by oxygen treatment.

194

Volatile matter treatment Oxygen treatment

Photo 3.2.8-6 Example of scale adhering to rectifying cylinder of high-pressure supply water heater

(1) Corrosion at the furnace water-wall tube of coal-fired boilers ① Corrosion

It has been known from long ago that strong corrosion occurs at the furnace wall of coal-fired boilers by flame impingement (flames hit the waterwall piping directly in the neighborhood of the burner zone)⁾¹⁴⁾.

In such a case, it is considered that the area exposed to flames is locally placed under low oxygen partial pressure, because a lot of unburned carbon, FeS₂, etc. are contained in the adhered ash. As shown in the chemical formula below, FeS₂ contained in the adhered ash reacts with the Fe contained in the waterwall piping to yield FeS. Because FeS contains more grid defects than such oxides as Fe₃O₄, the protective capability of the coat becomes poorer, causing strong corrosion.

Recently, many cases are found such as the 2-step-type combustion process being employed for many boilers for power generation as a measure to satisfy low NOx yield. In such a process, the area in the neighborhood of burner zone becomes an atmosphere of low oxygen partial pressure containing H₂S. Figure 3.2.8-2 shows the impact of air ratio on the balanced structure of gas when Datong (Chinese) coal containing 0.63% S is burned at 1300. When the air ratio is 0.8 or less, it is obvious that a lot of reduced contents such as H₂, CO, H₂S, etc. is contained in the combustion gas. In particular, when such coal containing a lot of S content is used as fuel, the H₂S density becomes higher, creating a severe corrosive environment. Corrosion of the furnace waterwall piping caused by high-temperature sulfide becomes a critical issue. Coal combustion gas is composed of CO₂, CO, H₂O, H₂S, COS, N₂, etc. As a result, the environment has become a family of so-called C-H-O-S. The critical factors of corrosion are oxygen partial pressure and sulfur partial pressure in the atmosphere. In an atmosphere where the oxygen partial pressure is high, oxidation plays a leading role in the corrosion of materials, whereas in an atmosphere where the sulfur partial pressure is high, sulfuration plays a leading role. In an atmosphere where oxidation is the leading player, the protective characteristic of the oxidized coat becomes excellent, resulting in a negligible level of corrosion. On the other hand, in an atmosphere where sulfuration is the leading player, the protective characteristic of the sulfide coat becomes remarkably poor, resulting in strong corrosion.

With respect to the corrosion occurring in an atmosphere of low oxygen partial pressure and high sulfur partial pressure, it is considered that the reaction mentioned below is the leading player.

195

Datong coal (S content: 0.63%) Combustion gas temperature: 1300°C

Air ratio

Fig. 3.2.8-2 Impact of air ratio on the balanced structure of combustion gas at 1300

H2S + Fe → Fes + H2 ...........................................................................................................(7)

2CO + SO2 + Fe → FeS + 2CO2 ..........................................................................................(8)

This corrosion gradually grows to complete corrosion in general. At a portion where repeated thermal stress is strong, the corrosion may grow in a groove shape in the direction of the circumference (which is called “elephant-hide alligator-skin cracking”).

Other than the above corrosion, it has been reported that corrosion involving such vitriols as X₂SO₄, X₂S₂O₇ (X: Na or K), etc. contained in the adhered ash or pyrosulfate can occur when the SO₃ density in the combustion gas is high ⁽¹⁵⁾. However, cases of corrosion of the waterwall piping by these alkali compounds are not reported very frequently.

In UK where coal containing lot of Cl is used, acceleration of corrosion of waterwall piping caused by HCl contained in the combustion gas has been reported. Because the coal currently used in Japan contains a very small amount of Cl, no corrosion caused by HCl contained in the combustion gas has been reported to date. From the standpoint that poor-quality coal may be used in future as fuels to be used diversify, it will be required to capture well the influence of HCl on corrosion.

196

② Examples of corrosion and countermeasures A cross section of corrosion of the furnace waterwall piping of a coal-fired boiler that occurred in which the

2-step combustion process has been employed in order to reduce NOx is shown in Photo 3.2.8-7. The flame side has been evenly corroded, and the amount of corrosion was 0.15 – 2.00 mm/year.

Flame side

Photo 3.2.8-7 Cross section of corroded potion of furnace waterwall piping

Photo 3.2.8-8 includes EPMA photographs of corrosive scale. The scale in the outer layer is composed of FeS, whereas the inner layer is composed of a mixture in which Fe₃O₄ is the main content. It is typical corrosion in an atmosphere of low oxygen containing a considerable amount of H₂S. In the neighborhood of the waterwall piping surface where strong corrosion occurred, it is indicated that the content of H₂S in the combustion gas was 300 ppm, H₂ was 1.5%, and CO was 6.1%, and the air ratio at the moment of combustion was 1 or less.

An example of groove-shape corrosion of the waterwall piping is shown in Photo 3.2.8-9. The appearance of the corrosion is similar to that occurring at heavy oil-fired boilers. The causes of such groove-shape corrosion are considered to be follows. Namely, the oxidized coat on the piping surface has cracked by repeated thermal stress arising from any combination of adhered substances to the inner piping (Fe₃O₄), condensation of air bubbles, or local falling off of scale from the surface of the furnace piping. It is considered that corrosive gas entered through the cracks and that the corrosion was accelerated at this gas-entered portion ⁽¹²⁾.

Considerable actions to prevent corrosion of waterwall piping are as follows: (a) Measures to be taken in the design

(b) Selection of materials

(c) Employment of surface treatment

The most effective action is the use of coal with a lower S content. Such actions as employment of low-NOx burners, use of fined coal to promote complete combustion, increase in the oxygen partial pressure on the piping surface by filling boundary air (to create an air curtain along the waterwall piping) over the waterwall piping surface, etc. are also considered effective ⁽¹⁸⁾ ⁽¹⁹⁾.

Photo 3.2.8-10 shows the EPMA observation result of the scale on the piping surface before and after filling of boundary air. By filling of air, the scale mainly containing sulfide has changed to scale mainly containing oxide.

As measures against groove-shape corrosion, suppression of the generation of substances adhering to the inner portions through thoroughgoing water treatment or prevention of air bubbles from condensation by employing rifle pipes is considered effective ⁽²º⁾.

197

Na: X-ray image

S: X-ray image

Cl: X-ray image

K: X-ray image O: X-ray image

C: X-ray image

Fe: X-ray image

Photo 3.2.8-8 EPMA observation result of corrosive scale on waterwall piping

Photo 3.2.8-9 Appearance of groove-shape corrosion of waterwall piping in the neighborhood of the burner

With respect to the materials, use of the double piping system composed of an outer pipe made of materials

excellent in corrosion resistance such as SUS 347 H, SUS 310 S, etc. and an inner pipe made of carbon steel is considered ⁽²¹⁾. These materials have been already put to practical use where the materials are exposed to severe combustion gas containing H₂S and HCl ⁽²²⁾.

For surface treatment, chromizing treatment by raising the Cr density by having Cr diffuse and penetrate is effective for prevention of corrosion also. In addition, thermal spray coating of corrosive materials by plasma thermal spray is effective for prevention of corrosion. Thermal spray process using 50 Cr – 50 Ni as its material has been put to practical use.

In the case of thermal spray, however, entry of gas into the layer of the metal/thermal spray cannot be avoided. This process has not yet been put to practical use as a permanent countermeasure.

198

O: X-ray image

S: X-ray image

O: X-ray image

S: X-ray image Fe: X-ray image

Fe: X-ray image

Before filling boundary air

After filling boundary air

Photo 3.2.8-10 EPMA observation result of corrosive scale adhering to waterwall piping before and after filling boundary air

199

200

3.3 Water Chemistry for the Boiler 3.3.1 Transition and Summary of Water Treatment Technology

The current water treatment technology we use in Japan derives from the U.S., introduced together with the so-called ‘new type of thermal power’ system. 3.3.1.1 Transition of 170K-Class Water Treatment

A 170K-class unit was imported and installed at Osaka Power Plant. The boiler used was a forced circulation type made by Combustion Engineering. Water treatment using the 170K-class unit showed a series of problems, and the method used for water treatment changed several times. (1) Initial Criteria for Water treatment

The initial criteria under which water quality was controlled in Osaka Power Plant are shown in Table 1. Caustic treatment was used, in which sodium hydroxide and sodium phosphate were injected into the boiler water. (2) Hide-Outs and Turbine Scales

Power generating efficiency decreased due to the hide-out of phosphate ions in the boiler water (and subsequent increase in pH), and also due to deposition of sodium phosphate scales to the turbine blades (See Table 3.3.1-1). Consequently, the downwash of the scale at the time of turbine start up contaminated the condensate water . To eliminate the hide-out, disodium salt was used and the phosphate ion concentration was maintained at 0.2 - 2.0 ppm, which was the upper limit that Osaka Power Plant was able to manage.

However, hide-outs still existed and it was relatively difficult to control pH at an appropriate level. 1) Low Phosphate Treatment

In early 1961, a test for treating water with low phosphate treatment started. Following the results, monobasic sodium salt was used, but since it failed to reduce the pH to the threshold limit value of GA1, i.e. 8.5 to 9.5, the value remained to be 9.5 to 10.0.

In September 1961, Mr. Grabowski of C.E. made a presentation at the New Nagoya Power Plant and Thermal Power Division of Kansai Electric Power Company, Inc. and showed that reducing boiler corrosion does not necessarily require increasing pH, but the key is to protect the magnetite protection coating. He also pointed out that Coordinated Phosphate Treatment requires the pH value to be maintained at the 9.5 to 10.0 level only (as experienced by a boiler manufactured by C.E.) and suggested keeping the concentration of phosphate ions on the concentration curves of trisodium salt and pH.

Table 3.3.1-1: Example of Analysis of Depositions on the Turbine Blades at Osaka Power Plant (Unit: %) Ignition Loss SiO2 Fe2O3 Na2O CuO PO4

High Pressure Moving Blade 1st to 4th stages * Medium Pressure Static Blade 1st to 4th stages * Medium Pressure Moving Blade 3rd to 5th stages Medium Pressure Moving Blade 6th to 8th stages Medium Pressure Static Blade 8th to 9th stages Medium Pressure Static Blade 10th stage Medium Pressure Static Blade 11th stage Medium Pressure Moving Blade 13th stage Low Pressure Static Blade 1st to 3rd stages Low Pressure Moving Blade 2nd stage

4.1

10.0

20.3

24.2

17.0

10.4

13.3

22.1

2.1

2.2

1.5

16.0

23.7

23.9

12.3

2.9

3.1

9.0

1.9

1.9

36.4

14.5

15.1

19.8

48.7

61.3

53.5

34.8

84.7

54.5

38.2

49.0

58.4

49.7

41.8

40.1

42.3

63.9

5.8

6.7

3.1 -

6.4

1.7

2.6

3.7

1.7

1.4

5.0

7.2

45.6

23.8

0.2

0.5

0.2

0.8

0.2

0.1

0.2

0.1

(Blade composition: 12 stages for High Pressure and 13 stages for Medium Pressure and 6 stages for Low Pressure) * As shown in the original document

2) All volatile Treatment Mr. Grabowski also showed a method involving the use of volatile chemicals to treat boiler water. He noted (1)

the use of volatile chemicals requires thorough monitoring of any condenser leakage and assurance of pre-boiler system operation, (2) if a leakage occurs, phosphate salts must immediately be injected, (3) the phosphate salts work effectively against the leakage of sea water at a concentration of 5ppm or above, below which they are ineffective and (4) after reaching the cationic conductivity of 0.5µS/cm, the phosphate salt must be kept at 10 to 15 ppm.

He also instructed that the cationic conductivity of boiler water be maintained at 2 to 3µS/cm during normal operation and to feedwater at a pH of 8.8 to 9.2.

R

atio

of s

ilica

con

tain

ed in

stea

m to

that

in w

ater

(%)

Water Quality Requirements

Average pH Silica (ppm.)

201

Fig. 3.3.1-1: Impact of Pressure, pH and Concentration of Silica in Water to the Ratio of Silica under the Steam

Generation Volume of 5lb/h and in Static Condition

Pressure (psig)

1 Distribution ratio by Jacklin & Bronar

Fig 3.3.1-2: Curve of Acceptable Silica Concentration in Boiler Water

Acc

epta

ble

Con

cent

ratio

n of

Sili

ca in

Boi

ler

Wat

er (S

iO2,

ppm

.)

5 Coulter, et al: pH 7.8 - 9.0

(Silica Concentration in Steam: 0.02 ppm.)

Drum Pressure (kg/cm2G)

Acceptable silica concentration in boiler water in order to retain silica concentration in steam to 0.02 ppm or below

Silic

a (p

pm.)

202

Fig. 3.3.1-3: Acceptable Silica Concentration in Boiler Water (by C.E.)

(3) Shift of Criteria for Condensate water and Make-up water The reference pH value shown by Gilbert for condensate water and make-up water at Osaka Power Plant was

8.6 to 8.8. The value depends on the volume of ammonia generated by the decomposition of hydrazine. However, an increase of hydrazine injection caused a surge of pH to nearly 9.0. Gilbert explained that ammonia would attack the copper alloy condenser tube if the pH value of ACD was high. In this case, the value should be kept to 8.8. However, the company also explained that if the pH of ACD were below 9.8, the pH of the condensed and make-up water might be around 8.6 to 9.0.

So as criteria for pH and hydrazine concentrations, the ceiling was set to 0.05 ppm for hydrazine and 8.9 for pH respectively, so that they can be maintained at these levels, even when there is a load variation. As it is difficult to limit the hydrazine concentration to 0.01 ppm or below in a stable manner, and as there is concern regarding the accuracy of the analysis, the lower limit of hydrazine concentration was set as 0.01 ppm. (4) Silica and Silica Purge

In order to avoid bad influence to a turbine by silica scales, it is necessary to limit the volume of silica contained in steam. As for the limit, the following three reports were issued in the U.S.:

1) Experience shows silica in steam should be contained to 0.03 ppm or below to avoid any scales being deposited to the turbine.

2) No deposits was found in the low- and medium-pressure turbine blades of a turbine with 150MW, 170k and 550°C when the silica concentration in the high pressure turbine exhaust was retained to approx. 0.01 to 0.02 ppm.

3) Experience shows no silica is deposited on turbine blades when the silica concentration is kept at 0.02 ppm or below.

Fig. 3.3.1-4: Boiler Pressure and Maximum Permissible Silica Concentration Limit in Boiler Water

Pressure (psig)

All volatile treatment: pH: approx. 9.0

Indicates silica concentration should be kept below this line during normal operation to avoid any deposits.

Silic

a C

once

ntra

tion

(SiO

2 ppm

)

Indicates silica concentration can reach this line when a turbine is restarted after regular repair or when it is rapidly operated. However, the concentration should be closer to the real line and close attention should be paid to the concentration of silica in moisture.

Drum Pressure (kg/cm2°C)

203

Table 3.3.1-2: Maximum Permissible Total Soluble Solid Material in Steam (Unit: ppb)

Material Permissible

Concentration for Continuous Operation

Permissible Concentration for

Conditioned Operation

Permissible Concentration for

Intermittent Operation NaCl Na2SO4Na3PO4NaOH SiO2

40040060308

200020001506020

4000 4500 300 150 45

Total 898 4230 8995 Table 3.3.1-3: Example of Measurement for Drum Carry-Over in the U.S.

Name of Power Plant Chestexfield Ashtabula Load 170 MW Drum Pressure 2600 psig 2500 psig Na Concentration of Boiler Water

Concentration in Steam Carry-Over Ratio

7.5 ppm 0.0021 ppm 0.028%

10.6 ppm 0.0029 ppm 0.027%

NaCl Concentration of Boiler Water Maximum Carry-Over Ratio

9.4 ppm 0.057%

15.3 ppm 0.047%

Referring to the above reports, the concentration of silica was set to 0.02 ppm or below. The tolerance of silica concentration in boiler water depends on the ratio of silica distribution in saturated steam. It also depends on pressure and pH, as shown in Fig. 3.3.1-1. Based on the distribution ratio, Fig. 3.3.1-2 is drawn and C.E used Fig. 3.3.1-3.

In Fig. 3.3.1-2, the silica concentration is 0.18 to 0.19 ppm under the pressure of 186 to 188k and a pH of 7.8 to 9. From these data, the silica concentration was set as 0.2 ppm.

As the silica concentration in boiler water tends to rise when the boiler starts operation, due to the silica scale deposited on the turbine low-pressure blades being washed away by wet steam, a silica purge must be implemented to raise pressure by blowing the boiler, while ensuring the silica concentration is limited to within the designated value. This is the main cause of delays and increased load when starting the drum type boiler. Therefore, looser values were set, as shown in Fig. 3.3.1-4, for the concentration of silica when starting a boiler.

The silica-washing device installed in a drum manufactured by Babcock-Hitachi K.K. showed a remarkable ability to reduce the silica concentration in steam, doubling the permissible concentration of silica in the boiler. (5) Total Soluble Solid Materials

As for the total soluble solid materials, the following reports were issued in the U.S.: 1) Fig.3.3.1-2 shows the permissible concentration for continuous operation (the maximum concentration that

does not cause significant silica deposit after operating a turbine for 8,000 hours), the permissible concentration for conditioned operation (the maximum concentration after repeated stopping and restarting or under such operation conditions as variable pressure operation) and the permissible concentration for intermittent operation (the maximum concentration that does not cause any silica deposits for a relatively short turbine operating period). The permissible concentration for continuous operation is approx. 1 ppm.

2) No significant silica deposits were observed in the concentration range of 0.1 to 0.2 ppm. 3) In order to operate a turbine without any washing for an extended period, the silica concentration must be

controlled to 0.05 ppm or below. Experience at Himeji No. 2 Power Plant showed that deposits were rarely seen when a turbine was operated

with cationic conductivity of 0.3µS or below. This corresponds to a silica concentration of 0.05 ppm. Based on the above, the cationic conductivity and the silica concentration were determined as 0.3K-µS/cm or

below and 0.05 ppm or below, respectively. In order to determine the limit value for the total solid materials in the boiler water, the carry-over ratio of the

drum should be considered. With this in mind, the following data is issued: 1) The design value is 0.25%. 2) The value measured in the U.S. is 0.05% or so, as shown in Table 3.3.1-3. 3) The value measured in Himeji No. 3 Power Plant was approx. 0.15%.

Based on the above, the value was determined as 0.2%, taking safety into consideration, and the total solid materials in boiler water was set as 25 ppm.

As the measurement of water quality under the all volatile treatment is 5K-µS/cm, or 10K/µS/cm at worst, the total solid materials in boiler water was determined as 10 ppm.

First stage low temperature reheated steamSecond stage low temperature reheated steam

Ignition Fourth steam Normal valuesCombined

feeding

Time (h) Fig. 3.3.1-5: Trend of Hydrogen Concentration

after chemical cleaning

No. 2 bearing vibration increased by three-hundredth. Loads decreased by 3MW.

Tim

e

High temperature reheated steam (ppb)

Fig. 3.3.1-6: Hydrogen Concentration when Two First Stage Blades of Curtiss Turbine Flied Apart

Permittivity of Cation-Exchange Resin

Tim

e

Condensate tEntrance of ECOExit of WW

Second stage low temperature

reheated steam

KC-floc used

Fig. 3.3.1-7: Electrical Conductivity and Hydrogen when Water Starts to Pass Through a Filter

OH

- in

Boi

ler W

ater

(ppm

) (as

CaC

O3)

Throttle Pressure (kg/cm2)

: Crater-shaped corrosion observed at least once ×: No crater-shaped corrosion observed This is a chart indicating the relationship between the alkali level of hydroxy-ions in boiler water and pressure. The safe and unsafe domains for a boiler showing crater-shaped corrosion are indicated as a dotted line.

Fig. 3.3.1-8: Relationship of Alkali Level and Crater-shaped Corrosion

204

205

(6) Hydrogen As for hydrogen, it only indicates the corrosion condition of a tube and no measures can be taken based on it . The generation of hydrogen can be determined as stable, because it remains commensurate with the surface

area, regardless of the volume of steam generated. Thus, it seems normal that the hydrogen concentration doubles when the flow decreases by half. As seen in the example where the hydrogen concentration is 2 to 4 ppb under stable operation, the chemical reaction of iron and water continues to a certain extent, even under stable operation. This means the magnetite coating undergoes a cycle of damage and recovery to a certain extent.

Increased hydrogen generation tells that the following events are happening: 1) The magnetite coating incurs significant damage. : E.g.: After chemical cleaning, the magnetite coating is

removed and thus hydrogen increases, returning to the normal level as the coating is formed (Fig. 3.3.1-5). 2) A new steel surface has appeared. : E.g.: Iron powder is generated by the flying apart of turbine blade; a

new metal surface appears on it, on which a chemical reaction progresses rapidly (Fig. 3.3.1-6). 3) The metal temperature has surged abnormally. : There is a report that the hydrogen concentration increased

by about 10 ppb when a reheating pipe caused creep damage for a relatively short period. 4) Organic materials (sugars) inputted have been decomposed (Fig. 3.3.1-7): There is a report that fine resin

leaked out from a condensate demineralization tower when water was introduced into it immediately after replenishing the resin.

(7) Malfunctioning of Boilers in the U.S. Table 3.3.1-4 indicates the result of investigations by the American Society of Mechanical Engineers (ASME)

on 116 boilers in the U.S. from 1950 to 1959. As stated in the table, 40% of boiler showed some pipe damage, while 28% of boilers caused crater-shaped corrosion, which is considered alkali corrosion. Fig. 3.3.1-8 shows the relationship between crater-shaped corrosion and hydroxy-ions, expressly showing how the concentration of the latter may decline as pressure goes up.

206

Table 3.3.1-4: Outline of 116 Boilers in Use No. of Boiler % No. of Boiler %

Manufacturer A B C D E

Pressure (kg/cm2) 63 or below 64-91 92-126 127-155 156 or above

Capacity 90 or above 90-225 225-337 337-450 450 or above

Overheat Temperature (°C) 427 or below 427-496 496-552 552 or above

Reheating Temperature (°C) 496 or below 496-552 552 or below No reheating

Fuel Fine charcoal powder Gas Oil Others (chain grate-fed charcoal and coal)

Year of Operation Before 1950 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959

Economizer Used Not used

42 49 14 8 3

13 21 53 26 3

6 33 26 28 23

0 15 80 21

0 68 3 46

72 36 4 4

7 4 5 7 18 28 18 11 7 8 3

101 15

36 42 12 7 3

11 18 46 22 3 5

28 23 24 20 0

13 69 18 0

59 2

39

62 31 3.5 3.5

6 3 4 6

16 24 16 9 6 7 3

87 13

Deaeration unit Used Not used

Final treatment of makeup water

Deionizer is used. Steam evaporator is used.

Boiler water treatment Sodium sulfite Hydrazine Caustic soda Phosphate Potassium salt Organics

Condensate water treatment Morpholine Cyclohexylamine Ammonia

Problems No corrosion losses observed to pipes Crater-type corrosions Burnout due to overheat Bubbles observed Overheat at the top of pipes Orifice Others

Pitting corrosion of header Pitting corrosion of suspending metals Corrosion of separation tube

Attachment to header Corrosion of feed heater Turbine attachments Carry-over of silica

Acid washing With acid washing No acid washing With initial acid washing No initial acid washing

Total number of acid washings

Once Twice Three times Four times Five times

89 27

50 66

80 41 81 111 3

13

76 12 12

70

32

5 2 1

13

2

3

2

5 13 9

(Water-soluble 9) 3

34 82 45 37

44 (Initial acid washing 22)





77 23

43 57

69 35 70 96 3 11

65 10 10

60

28 (20 to 23% are of

fluidity hindrance.)

4 2 1

11

2

2

2

4 11 8

3

29 71

For users, chemical cleaning to prevent any damage, for manufactures, designs to avoid hot spots or fluidity hindrances, for consultants, the removal of dissolved oxygen, carbon dioxide and sulfur dioxide from the pre-boiler, curtailment of dissolving iron and copper and research into controlling the pH level are requested.

Tem

pera

ture

of B

oile

r Wat

er a

nd P

ipe

Mat

eria

ls (°

F)

Con

cent

ratio

n of

NaO

H (p

pm)

Concentration

Mother Water of Boiler Water

Temperature

Concentrated boundary Film

Fig. 3.3.1-9: Heat Transmission Film and Concentrated

Film at the Heat Transmission Surface

Alk

ali C

orro

sion

B

rittle

ness

aga

inst

H

ydro

gen

Stre

ss C

orro

sion

A

sh

Cor

rosi

on

Cor

rosi

on

Fatig

ue

Ash

C

orro

sion

O

verh

eat

Num

ber

of C

ases

Exam

ple

Year

Fig. 3.3.1-10: Accidents Occurring to Power Generation Boiler Pipes

(8) Alkali Corrosion Sodium hydroxide was used to control pH in the boiler water. As Fig. 3.3.1-9 indicates, a boundary film was

formed around the boiler pipes of the heat transfer surface where boiling occurred. The sodium hydroxide in the boiler water increased in concentration because it was left on the heat transfer surface as the water boiled up. As sodium hydroxide has high solubility, it was not deposited on the surface, but instead, a film of highly

207

208

concentrated sodium hydroxide formed. According to an example of calculation , a 100k-class boiler containing 100 ppm of sodium hydroxide shows the temperature of inner surface of a pipe is increased by 5°F when it is heated at comparatively low heat flow rate of 25,000 BTU/ft2・h, boosting the sodium hydroxide concentration by 10%.

Another report shows that the temperature of the inner surface of the pipe is increased by 30°C at some hot spots . The occurrence of such hot spots is considered attributable to film boiling that is likely to occur due to the enlargement of heat flux in a large-sized boiler, steam blanket, lack of flow rate, inappropriate burner positioning, contact of flame due to insufficient combustion control, biased combustion and gas flow, inclusion of slabs in the welded parts, blow holes and lack of fusion.

Sodium hydroxide can result in corrosion of steel at a concentration of 5%. When the concentration reaches 5% or above, it dissolved the protective oxide layer, causing the inner metal surface to become exposed and corroding it due to the reaction of water and steel. The hydrogen generated by the reaction penetrates into and damages the steel.

In Japan, alkali corrosion cases were also reported. As Fig. 3.3.1-10 shows, statistically speaking, this has been responsible for the highest proportion of boiler accidents having occurred to date.

The alkali corrosion is otherwise known as a caustic attack, or in the U.S., as crater-shaped corrosion, due to its shape. These differ from conventional caustic embitterment.

After such accidents, the injection of sodium hydroxide was stopped. Thereafter, a new finding was reported: namely that hideout of phosphate ions causes not trisodium phosphate but 2.65-sodium biphosphate at 689F and 2.85-sodium biphosphate at 572F respectively. This means 0.35 to 0.15 of trisodium phosphate in the system is in the form of sodium hydroxide. It thus emerged that phosphate containing sodium less than 2.6 sodium biphosphate should be used.

However, even if such phosphate is used, it was found that pH in boiler water increased past this level due to trisodium phosphate. The cause was identified as a leakage of sodium from a deionized water system. In order to avoid leakage, a double-bed operation was used to place the deionized water system just after regeneration to the latter stage. Due to the fact that the movement of the system used at the latter stage to the front stage resulted in a more significant leakage, a mixed type system was installed at the latter stage to use it dedicated to a polisher. Thanks to such measurements, no further alkali corrosion has been reported since 1963. (9) Shift to All volatile treatment

The No. 2 boiler of the Karita Power Plant (a 170k forced circulation boiler) started its operation in June 1959 showed alkali corrosion to evaporator tube at the 3,700th hour. This was attributable to sodium hydroxide and the use of the chemical was stopped. The investigation showed that powder scale was attached to its turbine blades, especially the final stage of the medium-pressure turbine, mainly consisting of sodium bicarbonate. Gilbert suggested the use of sodium acid phosphate to maintain the pH level of boiler water and make-up water at around 8.5 to 9.5. The phosphate ion was dramatically reduced two days after feeding phosphate ions into the drum. As this resulted in heightened conductivity of saturated steam and carryover to the turbine, the use of phosphate ions was terminated in February 1960.

As the volume of hydrazine was maintained, the pH level of boiler water was lower by 0.4 to 0.5 than that of make-up water, which was well below 8.5 and maintained at that level.

Regarding the risk of leakage within the condenser, trisodium phosphate is fed and a drum blow operation is started. After repairing the leakage, the concentration of phosphate ions decreased to 0.0 ppm.

In the U.S, all volatile treatment was used due to the heavy carryover having occurred, and an ambiguous trial just to maintain turbine operation started. For them, there was no choice other than the use of the treatment method.

Table 3.3.1-5: pH Control Methods of the Pressure Boiler of 130kg/cm2 or above by C.E.

How to Adjust pH No. of Boilers (1) Caustic Based pH10.5 to 11.0

(Caustic alkali and phosphate ions are used.) 37

(2) Low Caustic Control pH10.0 to 10.5 (Same as above) 8

(3) Cordinated Phosphate – pH Control pH10.0 to 10.5 (Coordinated phosphate treatment is used and caustic alkali is not used.) 29

(4) All volatile treatment: pH 8.5-9.0 (Hydrazine and ammonia treatment is used and no solid chemicals are used.) 21

(10) Chemicals Fed into the Drum

To avoid alkali corrosions, no free sodium hydroxide should exist and the pH should be minimized. However, decreasing the pH level is not desirable in preventing oxygen corrosion and sodium hydroxide helps the removal of magnesium as sludge during leakage into a condenser. To maintain pH at a certain desired level, the phosphate ion concentration should be increased, but doing so can stain turbine blades. Therefore, where the increase of pH is not abandoned, sodium hydroxide should be used to prevent the turbine blades from deposit. In order to avoid both turbine scale and alkali corrosion, only volatile chemicals should be used, although doing so is not safe for preventing the seawater leakage of the condenser. There is no single medicine to cure all such difficulties. Thus, it was decided to implement a comprehensive examination of conditions surrounding the boilers to decide which priority should be chosen and favored. Table 3.3.1-5 shows the experience of the U.S. in 1959. (11) Problems in All volatile treatment

At the Kansai Electric Power Company, the use of volatile chemicals was applied to all units of 250KW or above. However, there were reports of white laminated scales deposited in a boiler, as shown in Fig. 3.3.1-11, with total thickness of 0.4mm and a total of around 11 to 12 white laminations. Each ingredient, especially the 11 to 12 layers of copper and zinc, corresponded to the number of white laminations. There were some 86 instances of seawater leakage within the condenser, of which 10 were serious. This also corresponds to the number of white layers. The total calcium and magnesium content was 1 to 4% in the form of CaO, which is several times larger than other boilers. Rice associated the generation of hydrogen to decreased pH, due to leakage of the condenser, causing accelerated corrosion. He also determined that the corrosion became uncontrollable when volatile chemicals were used, and that no hydrogen corrosion would occur when coordinated phosphate treatment was used. O’Neal also commented that coordinated phosphate treatment was used in lieu of volatile chemicals to avoid any hydrogen embitterment.

Base Material of ScaleScale Thickness

All volatile treatment

Approx. 15%Approx. 80%In

gred

ient

st

reng

th

Approx. 10%

Scale Thickness (when scanned in an oblique directions)

209

Fig. 3.3.1-11: Microscopic Diagram of Scales on the Burner Side

(Source: Kurosawa et al.)

Total

Num

ber o

f Boi

lers

in O

pera

tion

125K Class

170K Class140K Class

210

Fig. 3.3.1-12: Trend of High Pressure Boilers

Year

Perc

enta

ge in

the

Entir

e Tr

eatm

ents

%

Sodium phosphateSodium

hydrate


Potassium salt treatment

Fig. 3.3.1-13: Trend of Boiler Water Treatment

Year

Dick revealed that all volatile treatment is not a panacea against water damage accidents, because there were

reports of significant water wall tube trouble occurring in some boilers subjected to all volatile treatment. Decker disagreed with the use of all volatile treatment on drum type boilers, stating that such treatment was

rarely beneficial to them. Every few years, an accident involving increased differential pressure to a forced circulation boiler with all

volatile treatment happens due to the attachment of scales to the orifice installed at the inlet of water pipes. This is likely to occur just after chemical cleaning and the likelihood of such incident is based on how the initial crystallization occurs. The scale present consists of mainly magnetite. As anti-scale measures, (i) removal and cleaning, (ii) modification of the orifice shape, (iii) change of the orifice material to soft steel, and low phosphate treatment were carried out.

Subsequently in 1970, of four heavy oil boilers of virtually the same design, two treated by volatile chemicals were involved in an accident, while another, although not involved in an accident, was affected by a zinc scale attachments on a considerable scale. On the other hand, boilers subjected to low phosphate treatment showed no such zinc scale deposits. It was considered that the use of low phosphate treatment not only stops zinc deposits but also decomposes part of the zinc oxide deposits into zinc phosphate.

Thus, conventional method of all volatile treatment was switched to phosphate treatment at boilers used mainly by Electric Power Development Co. Ltd. Table 3.3.1-6 shows a comparison performance comparison featuring the scale generation behaviors of 23 boilers before and after changing the mode of water treatment. As the table shows, 4 units showed that phosphate treatment worked well in reducing the generation of scales. No unit showed an increase of scale when phosphate treatment was used in lieu of all volatile treatment. Even taking into consideration the change in operating conditions, phosphate treatment showed a reduction of scale generation. It was therefore considered that phosphate treatment was applicable for restricting scale deposits in the generating tubes of a boiler.

211

Table 3.3.1-6: Survey on Boilers whose Water Treatment Method was Switched from All volatile treatment to Low Phosphate Treatment

Materials for the Condenser Materials for the Feed Heater Unit No.

Capacity (MW)

Furnace Type

Circulation Method

Start of Operation

(Year/ Month)

When Phosphate Treatment was

Started (Year/Month)

Condensate water Part Air Cooling Part Low Pressure Part High Pressure Part

36 375 Single Natural 47.11 53.3 Aluminum brass Nickel-plated aluminum brass

Aluminum brass Monel metal

65 156 Divided Natural 39.8 48.1 Aluminum brass Cupronickel Copper arsenate Monel metal 66 156 Divided Natural 41.2 45.3 Aluminum brass Cupronickel Copper arsenate Monel metal 67 350 Divided Natural 44.1 51.5 Aluminum brass Monel metal Aluminum brass Carbon steel 70 156 Divided Natural 39.6 53.2 (50.6P→A) Aluminum brass Cupronickel Copper arsenate Monel metal 78 156 Divided Natural 48.4 53.5 Aluminum brass Titanium Aluminum brass Alloy steel

Carbon steel 86 250 Divided Natural 42.7 49.4 Aluminum brass Cupronickel Aluminum brass Alloy steel 87 250 Divided Forced 43.7 52.6 Aluminum brass Nickel-plated

aluminum brass Aluminum brass Alloy steel

88 250 Divided Forced 44.1 52.2 Aluminum brass Nickel-plated aluminum brass

Aluminum brass Alloy steel

89 265 Divided Natural 42.5 47.5 Aluminum brass Titanium Copper arsenate Carbon steel 90 265 Divided Natural 44.9 50.10 Aluminum brass Titanium Copper arsenate Carbon steel 106 350 Divided Natural 47.2 54.1 BKCB Titanium Aluminum brass Carbon steel 108 265 Divided Forced 35.10 54.11 Aluminum brass Nickel-plated

aluminum brass Copper arsenate Monel metal

109 265 Divided Forced 37.9 54.10 Aluminum brass Titanium Copper arsenate Monel metal 112 350 Divided Forced 41.7 53.6 Aluminum brass

Titanium Titanium Copper arsenate Alloy steel

113 350 Divided Natural 42.1 55.12 Aluminum brass Titanium Copper arsenate Alloy steel 118 350 Divided Forced 43.1 53.7 Aluminum brass

Cupronickel Aluminum brass Aluminum brass Alloy steel

122 250 Single Forced 45.6 54.1 Aluminum brass Cupronickel Aluminum brass Alloy steel 147 350 Divided Natural 44.9 53.6 Aluminum brass Titanium Aluminum brass Alloy steel

Carbon steel 148 350 Divided Natural 45.1 55.12 Aluminum brass Titanium Aluminum brass Carbon steel 168 250 Divided Natural 46.1 51.3 Aluminum brass Cupronickel Aluminum brass Carbon steel 169 350 Single Forced 48.1 53.12 Aluminum brass Cupronickel Aluminum brass Carbon steel 181 400 Single Forced 52.9 56.3 Aluminum brass Titanium Aluminum brass Carbon steel

Table 3.3.1-7: Corrosion Damage Reported in the U.S. on Drum Boilers (125k Class or above)

Periods New damage reported

Number of units in operation

Ratio of annual damage occurred

1955-1960 1961-1965 1966-1970

48 39 27

219 385 481

3.6% 2.0 1.1

* Corrosion newly discovered (12) Water Treatment and Boiler Accidents in the U.S.

Fig. 3.3.1-12 shows statistics concerning high pressure boiler damage reported in the U.S. up to 1970. Fig. 3.3.1-13 also shows a shift of boiler water treatment methods. It shows that the use of sodium hydroxide keeps declining, while that of sodium phosphate is on the rise. All volatile treatment peaked in 1963 (25%) and declined thereafter, falling as low as 5% in 1970.

Table 3.3.1-7 shows the trend of boiler corrosion, in terms of year on year decline. However, even recently, 1% of boilers used today are prone to corrosion damage. As Table 3.3.1-8 shows, 170k-class boilers tend to corrode, even when volatile chemicals are used to treat the boiler water. This ratio is relatively high compared with other classes, indicating why this type of treatment was declined.

However, a few cases of damage allegedly occurred due to the all volatile treatment not showing any evidence of boiler water contamination. The users believed that the all volatile treatment maintained the deionized water at the specified level. Experience supports the theory that it was the boiler design making it prone to corrosion) although Klein considers such an idea to be illogical.

Boiler corrosion cases in the U.S. peaked in the early 1960s, when various reports were submitted by both users and consultants. They reported that corrosion was caused by (i) basically inappropriate boiler design, (ii) contact of flames, (iii) insufficient circulation of boiler water and (iv) excessive evaporation speed. A few users suggested that strict control should be imposed to the heat load exposed to furnaces. 3.3.1.2 Appearance of Subcritical Pressure Once-through Boilers and the Necessity of Condensate

Water Treatments As the type of boiler evolves from natural circulation to forced circulation to once-through types, stricter water

treatment technologies have been called for. A once-through boiler requires departure from the conventional type of water treatment technology, because all foreign matter introduced to the boiler, together with the feed water, tend to become deposits on boiler tube and turbine blade.

A once-through boiler requires ultra pure water, which contains a concentration of solid substances as low as

212

just 5 to 500 ppb. In order to obtain such low concentrations, it is necessary to avoid mixing solid substances into the feed water.

Generally speaking, the following are considered sources of solid substances into feed water : Solid substances derived from construction phase A mixture of cooling water into the system, due to leakage of the condenser Corrosion products derived from the feed water system Solid substances in supplementary feed Solution of resin from the deionized water system

Among them, leakage of the condenser occupies the largest part. To eliminate this, condensate purification equipment is installed. This unit is relatively effective in removing dissolving metal.

When an equilibrium between the protective coating on the metal surface and the water contacting it is lost, due, for example, to load fluctuation, and system start-up and stoppage, the volume of corrosion products contained in feed water rapidly increases.

In order to maintain the stability of the metal surface during a system stoppage, the system should be carefully protected while not in use, with the use of hydrazine water of high pH and nitrogen sealed in tubes.

At the same time, a clean-up operation should be implemented before the system restarts operation. As for the solution of resin from the deionized water unit, resin slightly dissolves into water during the initial phase of the unit start-up, until it reaches a stable condition, meaning fine-powdered resin mingles into the supplementary feed. The effects thereof were not the subjects of research in 1959. In those days, it was considered that installation of a fine pore filter before and after a condensate purification equipment was effective in removing fine powder resin.

As chemicals capable of solidifying must not be used, hydrazine, ammonia and amines should be used. However, amines are not recommended because they decompose at high pressure.

Table 3.3.1-8: Damages to Boilers by Pressure and by the Water Treatment Method Pressure 125k Class 140k Class 170k Class Total

All volatile treatment Sodium Phosphate Treatment Sodium Hydroxide Treatment Potassium Salt Treatment

5 12 41 5

7 3

18 5

11 8 4 -

23 23 63 10

Total 63 33 23 119

The No. 2 Boiler of the Himeji No. 2 Power Plant was imported from the U.S. and commenced comercial operation in 1964. Table 3.3.1-9 shows the water quality standards, set based on actual operation performance. This boiler uses hydrazine and ammonia injected into the outlet of its condensate purification equipment and ammonia into the outlet of its deaeration unit. (1) Condensate purification equipment

Condensate purification equipment normally consists of a mixed bed condensate demineralization tower and filters placed in front of it. Some act as a polisher of supplementary feed.

The following three major objectives are associated with the use of condensate purification equipment: 1) To prevent damage to the entire system due to the leakage of a condenser 2) To purify the supplementary feed 3) To remove corrosion products from the feed water system

The secondary objectives include: 1) to purify various drain water before it enters into the system, and 2) to purify the system during the initial start-up and shutdown operations.

Contamination of ion-exchange resin by metal oxides and pressure loss of the condensate demineralization tower impose significant impacts to the system when a high-speed ion exchange takes place. Due to this, a filter is installed just before the tower to remove them and prevent the ion-exchange resin from deterioration.

As for appropriate materials to use for the filter, cellulose, diatomite, leaf-type and other fine pore filters are recommended.

As for the condensate water condensate demineralization tower, although the impact of dissolved resin to the tower has not yet been clearly identified, some say that the use of a filter, which puts after the demineralization tower, can eliminate leakage from the tower. Himeji No. 2 Power Plant employs a combination of filters made by United Filters, Inc. and Permutite Company. The label said the design conductivity was 0.2µS/cm, and that design silica, iron and copper concentrations were 7 ppb, 5 ppb and 5 ppb respectively, although these are not guaranteed values.

The imported item of Himeji No. 2 Power Plant uses a horizontal leaf type pre-coat filter and Solka floc

213

BW-100 and -40 were used at a ratio of 1 to 1 as filtering agents. The agents heavily leaked out, were deposited on the resin surface of the condensate demineralization tower caused pressure loss of the tower. The reasons for the leakage were attributed to leaf-end gaps, the distance of leaf hubs, the non-parallel arrangement of the same, variations in the water flow, an excessive design flow rate and the screen structure. The pressure loss exceeded the design value, after pre-coating, it was attributable to the excessive flow rate and an overly small shaft and shaft hole diameter. To eliminate the pressure loss, the system underwent renovation, but the loss still exceeded the design value. So, other filter unit was added.

It is believed that the black carbon precipitated to the boiler tube when the filtering agent leaked in large amount.

The condensate demineralization tower showed resin leakage, which was attributable to the gap of the disc strainers, the distances between a disc strainer and the strainer plate and between the strainer plate and a bottom plate.

In addition, sending resin to the regeneration tank led to massive amounts of residual resin accumulating at the bottom of the tower, resulting in an insufficient regeneration process and imbalances between the cation and anion resins. This was due to a structural defect at the bottom of the tower.

As the pressure loss of the condensate demineralization tower became abnormally high, resin with less than 60 mesh was filtrated using a filter (Permatite Q and S-1). The total annual fraction ratio came to approx. 45%, with the ratio of damaged anion resin particularly high. As for the cause of the fractured resin, this was found to be attributable to the relatively high design flow rate of 119m/h (51). Due to such experiences, a flow rate of 80m/h was recommended.

However, despite such measures, the water purity showed no improvement. Even after investigation by a Japanese condensate demineralizer manufacturer, no causes were identified. So, the staff was so desperate for help that they used a sieve to remove small resins, whereupon the water purity showed improvement.

Table 3.3.1-9: Criteria for a Once-through Boiler

Pressure Subcritical Pressure

Supercritical Pressure

Condensate water Cl- ppm O2 ppb

0 40

0 40

Demineralized Condensate water

Electrical Conductivity µS 0.2 0.2

Feed Water pH O2 ppb N2H4 ppb Cationic conductivity µS/cm Fe ppb Cu ppb SiO2 ppb

9.0-9.5 5

5-30 0.2 10 5 20

9.2-9.7 5

5-30 0.2 10 3 10

Values not shown in the form of a range (a ~ b) are the maximum allowable values.

(2) Cleanup (Inlet of an economizer in front of a boiler unit of Himeji No. 2 Power Plant)

Lowest Range

Highest RangeStart of switchover to a high pressure heater (second time)

Tim

e

B-Line Feed Water

Feed water rate increased(200t/h → 260t/h)

Fig. 3.3.1-14: Example of Flow-Out of Suspended Particles after Increase

in the High Pressure Heater Flow Rate and Switchover of Lines

Picture 3.3.1-1: Suspended Particles in Condensate water

(during the clean-up process in supercritical pressure boiler) The cleanup process basically aims to remove foreign matter from the lines swiftly, stabilize the metal surface

and regulate water quality to ensure the normal operation of the unit. In order to realize this, each system must be separated and it must be cleaned up from condenser to the boiler, achieving and maintaining satisfactory water quality level.

There are the following two ways of purifying water: Blowing down Condensate purification equipment

The aim of the blowing is to uplift water quality, when boiler water quality has deteriorated to such an extent that it would damage a condenser.

After the blowing process, the water passes through a filter, bypassing a condensate demineralization tower, to reduce the iron concentration at the outlet of the filter to 30 ppb or below. Subsequently, the water can be fed to the condensate demineralization tower.

Suspended particles must be removed as far as possible via flushing before discharging them from the lines. The flow rate is a decisive factor during flushing and cleanup. For this purpose, the flow rate should be

increased by using a single line system, or by adding shocks to the flow rate by switching lines reciprocally. Fig. 3.3.1-14 shows examples of a large volume of suspended particles flown out of the system via the addition of shocks after switching to the high pressure heater.

The temperature of the boiler should be monitored to raise it in a phased manner by considering the relationship between the temperature and the maximum allowable iron concentration. The flow rate must be retained as high as possible. (3) Analysis of Iron Concentration using a 0.45µm Millipore Filter

Analysis of iron concentration must be done as swiftly as possible, as it is an indicator to determine the appropriateness of each cleanup process. The accuracy is relatively unimportant. A technique involving filtrating a certain volume of sample water to visually inspect the residues and compare its color with a standard specimen was introduced by Gilbert.

The technique was developed by B&W. Firstly, a filter of 0.1µm was used, but it soon emerged experimentally that more than 90% of particles that cannot be removed via filtration of condensed and feed water can be

214

215

eliminated by a 0.45µm filter. In Japan, as shown in Table 3.3.1-10, it was found that, when except a deaerator tank, iron particles of 0.45µm or above in size occupied more than 60% at the position requiring final assessment and at the time measurement required just after starting up the plant.

These results were based on the plant being in continuous operation mode, with shutdown rarely occurring. The successful results were attributable to the fact that large-sized particles detached from the boiler surface, etc., were flown out temporarily into water in the lines when the plant started operation.

However, stopping the system frequently during the DSS operation gradually reduces the volume of such particles of larger size. On the other hand, it was reported that the size of the needle-shaped corrosion products generated in a condenser during plant stoppage and restart consisted of FeOOH of 0.02 to 0.1µm and that other products, such as magnetite, also detached from the boilers of which the size and shape were apparent in the form of thin films of approx. 0.02µm and square-shaped products of approx. 0.1µm respectively. As the ratio of fine particles of 0.45µm in size or below newly generated from the system letup to restart tended to increase, it was necessary to examine whether a filter of 0.45µm should be continuously used or not.

Recently suggestions include, given the variability in the optical properties of FeOOH, Fe2O3 and Fe3O4, that the color strength of these three products should be measured and quantified.

Table 3.3.1-10: Particle Diameter Distribution of Suspended Iron Oxides in System

Water when the System is Started No. 4 Unit of Himeji No. 2 No. 2 Unit of Takasago No. 4 Unit of Kainan *3

Iron Collected by a Millipore Filter*2



Specimen Sampling Time*1

0.45 0.22 0.025

Sampling Time*1

0.45 0.22 0.025

Sampling Time*1

0.45 0.22 Outlet of the Condensate pump

1

2

180 86 43 72

179 85 60

100

1899060

100

1 2 3 4

28892

18991

26193

22894

29394

19694

27398

23893

30999

207100277

98239

99

1 2 3

83 88 41 96 63 94

35 93 42 97 64 95

Deaerator tank 1 2

17 30 30

100

23 40 30

100

234030

100

1 2 3 4

134317591858

826

1653165518581858

2790269030972994

1 2 3

3 37 6

73 4

54

4 50 6

73 4

63

Inlet of an Economizer

1

2

86 57 31

100

116 77 31

100

1268431

100

1 2 3 4

3362123517452861

3770164717493270

4789319127774087

1 2 3

11 73 11 83 6

88

12 80 12 87 6

96

Outlet of Furnace

1

2

86 57 31

100

116 77 31

100

1268431

100

1 2 3 4

2558194824

10023

100

2865276824

10023

100

43100

40100

24100

23100

1 2 3

44 94 58 95 3

72

45 96 58 96 3

72

* Sampling Time [No. 4 Unit of Himeji No. 2] [No. 2 Unit of Takasago] [No. 4 Unit of Kainan] 1973.6 1: After boiler inspection 2: Cleanup at 117°C

1973.5 1: Acceptance of cleanup before

the ignition of the boiler

2: Furnace fluid: 120°C 3: Furnace fluid: 190°C 4: Furnace fluid: 300°C

1973.8 1: Immediately before the ignition of the boiler2: Inlet of Primary SH: 200°C 3: Inlet of Primary SH: 350°C

*2 The upper line of each column shows the concentration of iron collected by 0.45µm, 0.22µm and 0.025µm filters, while the bottom line shows the percentage of iron oxides of more than 0.45µm, .22µm and 0.025µm in size relatively to the total iron.

*3 The iron concentration values are rounded off to the nearest whole number.

3.3.1.3 Emergence of a Supercritical Pressure Unit The first supercritical pressure unit commenced commercial operation at Anegasaki Power Plant in Chiba

Prefecture in December 1967, followed by the No. 4 unit in the Himeji No. 2 Power Plant in March 1968. In the U.S, the first supercritical pressure unit to commence commercial operation was the 315k Class Philio No. 6 unit (125MW) in 1957. Based on the experience of this unit, a 246k Class, 538°C unit was developed after 1964, and in 1966, such units occupied almost half of all power generation capacities developed for steam-power generation.

As in the U.S. a supercritical pressure unit started operation without identifying the movement of impurities in

water under supercritical condition, many troubles of copper scales to a high pressure turbine was reported due to the copper dissolved in steam. The troubles became a synonym of a trouble peculiar to supercritical pressure unit, that had not been experienced in subcritical pressure units. For example, Avon No. 8 unit experienced a copper scale deposit of 1.5 to 2.3mm in thickness and 5.5 lbs. in weight during its three-year operation, and the load of 250 MW was decreased to 216 MW.

Subsequently came the introduction of a supercritical pressure unit aiming to meet the increased power demand during periods of high economic growth. As the new unit handles supercritical pressure, completely different from subcritical pressure, the latest water treatment system was employed to handle the latest water treatment technology at that time.

There remained some challenges to be overcome in water treatment after the introduction of a supercritical pressure unit. The following are the experiences of the No. 4 unit in the Himeji No. 2 Power Plant. (1) Copper and Condensate purification equipment

As explained above, the concentration of copper must be minimized, as it deposits on turbine blades. Referring to experiences in the U.S, 2 ppb was determined as a target for the copper concentration.

To eliminate copper, there are two methods; namely removing copper alloys from plant and using condensate purification equipment.

For the former, steel pipes were used for the feedwater heater in lieu of copper alloy pipes. The No. 4 unit of Himeji No. 2 Power Plant only used copper alloy for its low-pressure heater No. 1 and 2.

There was no alternative to the use of aluminum brass and copper dissolved from a condenser can be removed by condensate purification equipment. However the condensate purification equipment manufactured by Graver and installed as the No.4 unit of Himeji No. 2 Power Plant met the guaranteed value of 0.3µS for electrical conductivity, while the iron, copper and total dissolved solid material concentrations of 10 ppb, 3 ppb and 35 ppb respectively were only the target values. The condensate demineralizer used was an external regeneration system.

With this in mind, the criteria for the copper concentration of the No. 4 unit of Himeji No. 2 Power Plant was set to 3 ppb, although the actual concentration could be contained at almost 2 ppb. All the units installed in the Power Plant thereafter used steel pipes for all feed heaters and the criteria was changed to 2 ppb, which was successfully met thereafter.

Temperature

Flui

d Te

mpe

ratu

re (°

F)

Iron

Dep

ositi

on V

olum

e (g

/ft2 )

Deposition

Length of Generating Tubes (ft)

(Source: B&W reports) Fig. 3.3.1-15: Iron Deposition to Generation Tubes (2) Cold Cleanup at a Temperature of 177°C

The following two were preconditioned for boiler cleanup activities: 1) To minimize foreign substances slipped into boilers to minimize deposit generation and hence reduce chemical

cleaning 2) To shorten the cleanup time

The flow ratio of WW maintained by a BCP is said to accelerate the cleanup because the contaminant reverts to the form of suspended particles and flows out from the boiler .

A cold cleanup refers to the initial cleanup process of a boiler previously used, which can literally also be cold. For the No. 4 unit of the Himeji No. 2 Power Plant, C.E suggested that a cleanup be done while keeping the boiler 216

217

ignited. As Fig. 3.3.1-15, a result of experiments at B&W, shows, the iron starts depositing in a generating tube at a temperature of 450°F (232°C). Therefore at this temperature or below, iron deposited at the inlet of an economizer need not be considered. In actual practice, a boiler is cleaned by keeping its temperature at 350°F (177°C) at the outlet of the WW to suspend foreign matter deposited on the wall of pipes, which is then removed using a condensate purification equipment. The temperature at the outlet of the WW may be risen up to 400°F. It is reported that Breed and Philo experienced circumstances whereby most iron oxide contained in feed water went through the boiler at an outlet temperature of 260°C to 288°C, and all of it was deposited on the pipe wall at a temperature of 316°C or above.

The following are the criteria for giving final approval to a boiler that cleanup be completed at a temperature of 177°C:

Inlet of an economizer: Iron 50 ppb Copper 20 ppb Silica 30 ppb Oxygen 10 ppb Outlet of a WW: Iron 500 ppb

If the iron concentration is 500 ppb or below at the outlet of WW when the deaeration feed water contains 50 ppb of iron, the contamination on the WW surface is minute. This is because it is said that experiences indicate that iron will not separate out at the WW when the temperature at the outlet of the WW is 218°C at the highest.

Experiences also indicate that the permissible level of iron concentration at the outlet of WW may be up to 500 ppb, rather than 50 ppb, without sacrificing cleanup effects or boiler performances.

Lax water quality is allowed after cleanup so that units can be installed in juxtaposition to obtain a reasonable flow rate. Moreover, since the cleanup does not take long, slightly deteriorated water quality will not cause any scales to be deposited.

Based on experience, when the iron concentration comes to 50 ppb, both copper and silica concentrations satisfy the limit values, and the acceptance of cleanup is determined by measuring the iron concentration at the inlet of an economizer only.

If the iron concentration at the inlet of the economizer reaches 50 ppb, both copper and silica concentrations at the same location should satisfy the limit values.

After the cleanup at 177°C, the temperature may be uplifted. During the temperature rise, the iron concentration at the inlet of the economizer should be kept at 50 ppb or below. Beyond 177°C, even if the spillover of the economizer is closed, the iron concentration can be maintained at this concentration or below. In the case that the concentration exceeds this value, the spillover should be increased, whereupon, the iron concentration can be kept at this level till combined input. (3) Steel Pipe Heater and pH Rise

As a result of a test , when the pH of the feed water at the inlet of the economizer was uplifted to 9.5, the iron separated out into the feed water system was significantly reduced. As for copper, no significant change was observed after the pH uplifted to 9.5. So the value of 9.5 was determined for pH.

The reason that the pH was limited to 9.5 was because more ammonia should be used if the value exceeds this level and because of curtailing chemical costs due to an increase in the number of regenerations and due to a deterioration in the water intake capacity of a desalination tank. 3.3.1.4 Advancement of Condensate purification equipments (1) Ammonia-Type Resin

Though the cation and anion resins contained within a condensate demineralization tower do not lose the function of removing ions such as sodium, iron and copper for the former and chloride ions and silica for the latter, they are prone to break down due to the ammonium ion exchange caused by a pH regulator.

Based on experience, an idea was proposed to use NH4 type ions as exchanger bases for cation resin to optimally utilize the resin functions. In the U.S, a series of simulation tests was conducted at the end of 1966, followed by the practical implementation of the method.

At the No. 4 unit of the Himeji No. 2 Power Plant, a series of tests using an actual unit was conducted from 1969 to 1970, following an experiment using a small-sized unit.

Consequently, it emerged that this method can withstand even a leakage occurring within the condenser. However, the findings were attributable to the fact that during the test, the unit was handled with extreme care, the ratio of regeneration was almost 100% and that due to this, the water quality at the inlet was excellent. With this in mind, the design of actual units required thorough consideration of various points. This consideration was made after the test and the method was implemented. (2) Electromagnetic Filter

A pre-coat filter was in use for 20 years. The shortcomings were the fact that it took 2 hours to regenerate and

that dissolving a pre-coat agent requires handling by operators, because this did not take place automatically. In addition, the pre-coat filter requires a pre-coat agent, resulting in a high running cost, and effluent sludge needs to be treated.

An electromagnetic filter was first used in a Kiel Power Plant (320MW). At the Power Plant, condensate water and a low pressure drain were treated at a temperature of 130°C, while the capacity of the electromagnetic filter was disclosed in 1966, as shown in Fig. 3.3.1-16. As the system water temperature rose at startup, the magnetite volume also increased. Alongside the same, the ratio of removing foreign matter was on the rise; 90% of total iron and 97% of magnetite were removed successfully. However, as the concentration of total iron fell to 10 ppb, the ratio was reduced to 80%.

After an electromagnetic filter had been developed in Japan, it rapidly spread to all newly installed units. The filter generates a high gradient magnet field by subjecting a solenoid coil to a direct high density current. It shows high removal performances against ferromagnetic and paramagnetic iron oxides when a filler is charged into the tower, making it an electric magnet, in combination with the mechanical filtration of the filler. The water feed filtration velocity (LV Value) is approx. 10 times higher for an electromagnetic filter than that for a conventional filter, and the whole unit can be miniaturized. As it facilitates regeneration relatively easily, this helps save significant amounts of both energy and labor. Moreover, no stand-by unit is required because flushing the unit takes as little as 15 minutes, including the preparation time. In addition, no meticulous operation control is necessary, the volume of effluents from of the unit is modest, no filtration auxiliary agents or other chemicals are required and the volume of sludge can therefore be minimized.

However, the shortcomings, according to a report, include its inability to remove high levels of paramagnetic α-Fe2O3 and α-FeOOH of fine powders and amorphous bodies, while its capacity to remove irons is slightly worse .

Normal Operation(0.3m/s)

Rat

io o

f Rem

oval

(%)

MagnetiteTotal iron

Heat up of a plant

Iron (pbb)

218

Fig. 3.3.1-16: Performance of an Electromagnetic Filter

(Source: Condensate water of the Kiel Power Plant)

(3) Hollow Fiber Membrane Filter

The first thermal Power Plant to use a hollow fiber membrane filter was Goi Thermal Power Plant No.2 in April 1988 . In operating DSS, as the conventional pre-coat filters fell short in terms of responding to the requirement of high speed condensate water purification and enhancement of boiler water quality, one of the three existing pre-coat filter was replaced by a hollow fiber membrane filter so that all condensate water generated at system start-up could be treated. The hollow fiber membrane filter is of polyethylene resin, with filtration holes of less than 0.1 micron in diameter on its surface and removes foreign matter on the external surface of the fiber. The foreign matter it captures is discharged from the membrane surface by backwashing filtrated condensate water from the inside of the filter by vibrating the membrane fiber using air pressure.

As the operation needed only involves drawing water and backwashing, they can be remotely controlled from the central control room.

Although the report states that the concentration of iron at the inlet reached 200 ppb, that at the outlet was kept below the detectable limit, as always and the cleanup time was successfully curtailed by half to a third of the original.

The bottleneck of this system, however, is the substantial initial cost required. 3.3.1.5 Introduction of an Oxygen Treatment Method

As indicated by the arguments thus far, water treatment technologies after the WWII have been solely reliant on the U.S. However, on a global basis, methods of water treatment employed may also originate from outside the

U.S., e.g. from Europe, where methods unique to this continent are used.

In Germany, an oxygen treatment method was developed in the latter half of 1960s and registered in VGB in 1972. Due to the lack of any ammonia attack on the condensate water pipes, the iron concentration in feed water can be retained, at least at a level equivalent to that of volatile chemicals and other benefits, and this oxygen treatment method has penetrated all over Europe. Indeed, the former Soviet Union employed the method in the mid-1970s for practical use. In Japan, the method has been applied to all boiling water reactors (BWR), in systems where chemical treatment cannot be applied to the primary cooling system, and a good operational record has been accumulated to date.

It was considered that, in order to apply the oxygen treatment method to thermal power plants in Japan, it was necessary to more clearly identify its impacts on reducing scales, the effects of curtailing boiler differential pressure, the influences on turbine materials, water treatment conditions when stopping and starting the system and other aspects. For this purpose an ‘oxygen treatment method assessment committee’ was established, featuring the membership of 10 electric power companies and the Central Research Institute of Electric Power Industry (CRIEPI). CRIEPI aimed to identify the above issues and commenced basic research into the practical use of oxygen treatment method to a once-through boiler in April 1988. The research period was 2 years.

Joint basic research carried out by the 10 electric power companies and CRIEPI was primarily focused on the following three examination items, and the committee was used as a venue for discussing and assessing in a comprehensive manner.

(i) The impacts of oxygen on the anti-corrosive performance of boiler pipes against high temperature water (ii) Impacts of oxygen on SCC and the corrosion fatigue of steam turbine materials (iii) Assessment as to how the oxygen treatment method is used in plants outside Japan and the provision of

temporary guidance for the practical application of this method to an actual system Various types of these tests, as well as a case assessment of how the method is used in overseas plants, were

summarized as shown below. The oxygen treatment method was found to have at least equivalent scaling and anti-corrosion performances to

the all volatile treatment method. Case assessment showed that the oxygen treatment method had the effect of curtailing a surge of boiler differential pressure and decreasing the generation of scales as well as no new system reports being reported. With such affirmative results, it was confirmed that the oxygen treatment methods could represent an ideal feed water treatment method to a once-through boiler. 3.3.1.6 Introduction of the Oxygen Treatment Method

The Chubu Electric Power Company Limited conducted an experimental research involving the application of the oxygen treatment method to the No. 1 unit of its Chita No. 2 Thermal Power Plant jointly with Hitachi, Ltd. (boilers) and Toshiba Corporation (turbines) in 1990. This was the first of its kind in Japan.

Some favorable results were obtained, including the curtailment of the differential pressure surge of boilers, decreased BFP powers and prolongation of the chemical cleaning intervals of boilers. As no adverse effects of corrosion and erosion were observed, the system was assessed as being applicable for practical use.

With such favorable assessments, the system will be introduced mainly to 18 once-through boilers of supercritical pressure class or above.

The following chapters explain the result of the research, final assessments and introduction plans.

Application of the Oxygen Treatment Method

Decreased iron concentration in the feed water

Change of scale characteristics

Decreased iron volume fed to the pumps

Curtailment of wave-shaped scale production

Curtailment of overheating to the generating pipes Decreased scale generation

velocityCurtailment of the differential

pressure surge of boilers

Curtailment of pressure surge at the outlet of a feed water

pump

Decreasing power

consumption of the feed water

pumps

Prolonged chemical cleaning intervals

Enhanced cost performance

Enhanced reliability

219

Fig. 3.3.1-17: Expected Effects of the Oxygen Treatment Method

Table 3.3.1-11: Research Plan Item

220

3.3.1.6.1 Characteristics of the Oxygen Treatment Method

The oxygen treatment method aims to prevent corrosion by generating a protective layer of trivalent iron oxide (Fe2O3 or hematite) by infusing a minute volume of oxygen (20 to 200µg/l) under an ultra pure water environment (0.2µS/cm or below) with pH between 6.5 to 9.0. Compared to an AVT protection layer of magnetite, hematite has relatively lower solubility, finer particles and generates sleeker protection layers, meaning the performance as shown in Fig. 3.3.1-17 can be expected.

There are two ways to use the oxygen treatment method; one is the neutral water treatment method and the other, the combined water treatment method (CWT), where the pH environment is 8.0 to 9.0. During this research, CWT was employed because it was once used before in Germany and because of its excellent performance in terms of curtailing the separation of iron and copper. 3.3.1.6.2 Outline of Research (1) Research Periods

April 1990 to September 1993 The work schedule is as shown in Table 3.3.1-11.

(2) Unit Subject to the Research No. 1 unit of the Chita No. 2 Thermal Power Plant (Supercritical pressure conversion and the once-through type with the output of 700MW)

(3) Research Items a. To establish optimum water quality conditions b. To establish an optimum mode of switching from AVT and CWT and vice versa c. To establish optimum operation methods of condensate water and the desalination unit. d. Assessment test of impacts on other units

(4) Research Facilities a. Oxygen injection unit (Oxygen is fed from a cylinder to the outlet of a condensate water desalination unit

and of a deaeration unit.) b. Low pH ammonia injection unit c. Water quality monitoring system

3.3.1.6.3 Results of the Research In line with the regular inspection for FY1990 (February 4 to 6, a series of facility installation works was

conducted, including an oxygen injection unit. For approx. 1.5 months from the start of the units and after regular inspection, AVT was conducted to smoothly

transfer to CVT after the start of CWT. Subsequently, on August 15, the system was completely transferred from AVT to CWT, whereupon, a series of

tests was conducted, including optimum water quality conditioning tests, long-term running tests and impact assessment test to other units. (1) Iron and Copper Concentration in Feed and Condensate water

a. Iron Concentration

FY1990 FY1991 FY1992 FY1993

Facility Design and Construction (Regular Inspection)

Feasibility Test (Inspection of

Facility)

Analysis and Assessment

Test to determine optimum operating conditions

Long-term running test

Analysis

Intermediate assessment

Comprehensive assessment

Analysis

(Legend)

(Legend)

Oxygen Oxygen

221

1) When Converting to CWT

Just after the conversion to CWT, the iron concentration surged by more than 8 times or 24µg/l at the inlet of an economizer, compared to AVT. This transit phenomenon was also seen in copper concentration (tripled to 2µg/l). Both phenomena, however, disappeared within a few weeks.

The reason for this may be attributable to the fact that protection layers were not formed smoothly because of decreasing pH (AVT9.6→CWT8.5) and the fact that the injection of oxygen was performed simultaneously when the CWT was first started. In future, first oxygen should be injected to monitor the behaviors of dissolved oxygen (DO) and iron while reducing pH step by step. 2) Long-Term Running (pH: 8.5, DO: 100µg/l)

As shown in Fig. 3.3.1-18, compared to AVT, the iron concentration doubled to 8.8µg/l at the outlet of a condensate pump before injecting oxygen, due to the decreased pH, while after injecting oxygen, it reduced to between a third and a half (3.1µg/l→1.6µg/l at the inlet of an economizer). The CWT thus seems relatively effective in reducing the iron concentration of feed water and the volume of iron fed into boilers. 3) Relationship between pH and DO

As shown in Fig. 3.3.1-19, at the outlet of a condensate pump, as the pH increased, the iron concentration tended to decrease, with a level equal or slightly higher than AVT. The same trend was observed in low pressure feed heater drain, although no significant effects were observed in other systems.

There were also no significant relationships observed between pH and DO. b. Copper Concentration

The copper concentration in CWT was at the same level as AVT, i.e. 0.6µg/l. No significant relationships were observed between pH and DO as well. (2) Differential Pressure of Units a. Differential Pressure of Boilers

The boiler differential pressure refers to the difference in the pressure lost between the inlet of an economizer to

Outlet of a Condensate water Pump

Outlet of an Electromagnetic

Filter

Outlet of a Condensate

water Booster Pump

Outlet of an Economizer

Outlet of a Deaeration

Unit

Main Steam

Fig. 3.3.1-18: Shift of Iron Concentration under the Long-Term Running Test (pH: 8.5, DO: 100µg)

Fig. 3.3.1.19: Relationship between pH, DO and Iron Concentration

Outlet of a Condensate water Pump

Outlet of an Economizer

(Legend)During the research period

Diff

eren

tial P

ress

ure

(kg/

cm2 ) During the past AVT

period

Regular Inspection (Chemical cleaning)

Regular Inspection Regular

Inspection

Number of MonthsStart of CWT

Fig. 3.3.1-20: Shift of Boiler Differential Pressure

Chemical cleaning

a steam separation drain tank.

As shown in Fig. 3.3.1-20, the pressure surged by approx. 8kg/cm2 in 1.5 months for AVT, before the start of CWT. It then became a decrease one month after the start of the CWT, whereas in 9 months, the differential pressure had decreased to the same level as that after chemical cleaning was applied to the unit (30kg/cm2). The trend continued thereafter, before ultimately reaching differential pressure equivalent to that at the commissioning of the unit, i.e. 27.5kg/cm2).

Compared to AVT, the differential pressure showed a significant decrease to approx. 15 kg/cm2, which was the same level as that of a unit 15 months after chemical washing.

Due to such reduced differential pressure of boilers and other factors, the BFP outlet pressure was reduced. The volume of steam required for operating the BFP was reduced by 6.7 t/h for low pressure steam and 5.3 t/h for high pressure steam, compared to AVT, as shown in Table 3.3.1-12.

This effectively shows that CWT is relatively effective in reducing the differential pressure of boilers and the BFP power loss.

Table 3.3.1-12: Comparison of Steam Volume for Operating BFP (Unit: t/h) 1 AVT 2 CWT 1-2

Low pressure steam High pressure steam

125.3 9.2

118.6 3.9

6.7 5.3

Note 1: For AVT, the figures are the mean values from Jan. 1987 to April 1990. Note 2: For CWT. The figures are the mean values from Jan. 1991 to Jan. 1992.

222

(Legend)The unit used for this research No. 4 unit of the Ulsan Thermal Power Plant

Gen

erat

ion

Spee

d (m

g/cm

2 , 1,0

00h)

(2.5 years)

9.5 years

Fig. 3.3.1-22: Water Pipe Scale Generation Speed of the Unit used for This Research and the No. 4 Unit of the Ulsan Thermal Power Plant

Fig. 3.3.1-21: Scale Generation Speed of Water Pipes, etc.

4 years(1 year after the start of CWT)

(Legend)

Gen

erat

ion

Spee

d (m

g/cm

2 , 1,0

00h)

Upper part

Lower part

Lower part

Furnace side

Furnace side

Furnace material

side

Furnace material

side (Coal economizer)

Upper part

(Water pipe: front wall)

(Generation unit)(Water pipe: side wall)

223

. Differential Pressure of High Pressure Feed Heater

ater refers to the total differential pressure from the inlet of N

ssure of boilers and in 10 months, it came down to 7kg/cm2 , na

s oilers and turbine-related units were performed in line with

reere performed in three stages, the first involved inspecting the conditions of the AVT as a

ba

art (pipe observation test)

on showed a time course decrease after switching over to

at

(ii)s after switching to CWT showed a decrease in the same manner as the volume of scale

bThe differential pressure of a high pressure feed he

o. 1 unit to the outlet of No. 3 unit respectively. It shows the same trend as the differential premely, the same level as the unit immediately after chemical cleaning. Thereafter, the trend continued until the

differential pressure stabilized at approx. 6kg/cm2. (3) Corrosion and Scale Deposition on the Unit

Inspections for corrosion and scale deposition on bgular inspections. The inspections wsis of the assessment, and the second covered its conditions 1 year after the switchover to the CWT (CWT-1),

while the third was done approx. 2.5 years after the switchover to the CWT. a. Major Units Related to Boilers (a) Economizer and Evaporation P(i) Volume of Scales Deposited (Generation Speed)

As shown in Fig. 3.3.1-21, the speed of scale generatithe CWT. In approx. 2.5 years, it reduced by between two thirds to a half (1→0.7mg/cm2, 1,000h at the front wall of the water pipe on the furnace side). Compared to AVT, the decrease was relatively significant i.e. by a between a half and a third (1.7→0.7mg/cm2, 1,000h at the front wall of the water pipe on the furnace side). The trend for decreased scale generation speed is, as shown in Fig. 3.3.1-22, the same level as that shownthe No. 4 unit of the Ulsan Thermal Power Plant of Korea Electric Power Corporation. Based on this, the ultimate scale generation speed of a CWT is 0.5mg/cm2, 1,000h. Scale Thickness The scale thicknesdeposited. In approx. 2.5 years, it decreased to a level almost equivalent to that around a year after the switchover. Compared to AVT, it showed a significant decrease by approx. a half to a quarter (0.06→0.03mm at the front wall of the water pipe on the furnace side).

(iii) Surface Conditions of Scale As Picture 3.3.1-1 shows, for AVT, the scale was in a waveform of crystals of 10 to 20 µm in diameter. Around a year after the switchover to CWT, the surface conditions of the scale changed into a shape of fine powders of several µm in diameter, and no waveform shape scales were identified. After approx. 2.5 years, further progress was made in terms of miniaturization of the fine powder diameter.

Picture 3.3.1-1 Surface Condition of the Scales Deposited in Water Pipes

224

225

Resin Resin

(iv) Cross Section of Scales As shown in Picture. 3.3.1-2, in AVT, uneven distribution of scale thickness and many voids were observed.

The thickness, thereafter, was thinned and the surface was sleeker approx. one year after switchover to CWT and the same trend was observed after approx. 2.5 years. (v) Coarseness of the Scale Surface

Compared to AVT, the coarseness of the scale surface approx. one year after the switchover to CWT was miniaturized by half (77→35µm). (vi) Chemical Composition of the Scale

As shown in Fig. 3.3.1-23, the X-ray analysis indicated that in AVT, most of the scales were in the form of magnetite, while in CWT, they were a mixture of magnetite and hematite (α-Fe2O3).

It is considered that due to such qualitative and quantitative changes of scales in CWT, the differential pressure of the boilers was reduced. (vii) Solubility of Scales to Chemical cleaning Agents

A scale solubility test was conducted to confirm the solubility of scales generated in CWT into chemical cleaning agents. Consequently, scales deposited in CWT were totally soluble to agents used for the chemical cleaning of AVT (1.5% citric acid and 1.5% hydroxyacetic acid), even after approx. 2.5 years had elapsed.

The solubility of scale to chemical agents was the same for that deposited to CWT and AVT. No insoluble scales were generated some time after the use of CWT was commenced. (b) Stagnant Water (pipe observation test)

Generally speaking, in CWT, water containing oxygen must be there to supply oxygen. For this reason, there

Scale Scale

Pipe Wall

Pipe Wall

Picture 3.3.1-2: Cross Section of Scales Deposited in Water Pipes

Fig. 3.3.1-23: X-ray Analysis of Iron Compounds (Water Pipes)

was concern regarding corrosion in sections containing stagnant water.

In order to observe and assess the corrosion, a drain pipe close to the inlet of an economizer and close to the inlet pipe of a horizontally-set superheater header were chosen as two representative locations where water tends to be stagnant and oxygen is hardly supplied. Consequently, no significant differences emerged in the corrosion performance of CWT and AVT.

226

(Legend)

Thic

knes

s and

Vol

ume

of S

cale

s (C

WT/

AVT)

(Uni

t: %

)

Table 3.3.1-13: Results of Analysis on Scales Deposited on the Main Turbine (Unit: Fe: %, Others: mg/kg) AVT CWT-1 CWT-2

Medium-Pressure Unit #9

Low-Pressure Unit #15





Fe Cu Cr Ni Mo

SiO2 Na Cl

SO4

61.1 4900 12400 810

3100 3600 3300

7 210

64.1 9400 9000 1100 3300

13800 5400 120 450

62.1 4000 10000 2200 2600 4100 1100

4 250

64.1 8200 9400 1500 3800 5900 1500 <1 940

61.4 800

8400 330

2500 3800 500 24 93

62.9 5900 9500 1000 3600 4800 560 17 220

(c) Instrumentation and Control Valves A series of investigations were conducted on 11 types of spray control valves of a superheater. As for corrosion and erosion performances, no significant changes emerged between AVT and CWT except for

erosion, as explained in 3.3.3.1.6-(3)-d ‘Parts Using Stellite Materials’ observed on the spray control valves of a secondary superheater.

Scales deposited on the water side showed a change from black magnetite to red-brown hematite, while on the steam side, black magnetite was observed, although the volume had declined. b. Major Turbine Components (a) Main and BFP Turbines

The scales were colored from black or gray to a mixture of slight red or yellow-brown. Though some scales were seen deposited on various parts of AVT, in CWT they were rarely seen. As shown in

Fig. 3.3.1-24, compared to AVT, the scales deposited on CWT were reduced to two fifths to one fifth for the overall main turbine and to one tenth to one twentieth for the BFP turbines, as estimated.

The chemical composition of the scale was, as shown in Fig. 3.3.1-13, the same as those found in AVT, but a significant decrease was observed in NaSO4 and SiO2. The results of an X-ray analysis showed an increase of hematite, like the scale attached to the boiler generation pipes.

The form of the scale was, due to the use of CWT, observed in smaller crystal shapes. As for the corrosion and erosion of blades, rotors, nozzles and enclosures, there were no significant differences

when generally compared to AVT. There was also no abnormality in the non-destructive inspection (PT, MT, UT). (b) Turbine control valve

Main Turbine

BFP Turbine

High Pressure Water Supply Heater Flow Rectification

Tower

Strainer at the

Inlet of a BFP

BFP Rotator

Adjusting Valve of

High Pressure

Water Heater Drain

Fig. 3.3.1-24: Thickness and Volume of Scales Deposited on Major Components of a Turbine (Relative

Comparison)

An investigation was conducted on seven kinds of valves including the main steam stop valve. As a result, there

was no significant difference when compared to AVT generally. The maximum depth of erosion to a sub-valve of a main steam stop valve was 4 to 5 mm, which is equivalent to

that observed in AVT. (c) Instrumentation and Control Valves

A series of investigations was conducted to 9 types of valves, including BFP overheat prevention valves. Consequently, the drain control valve attached to a high pressure feed water valve No. 3 showed a significant decrease in the volume of scales (while in AVT, valve sticks were observed due to the deposition of scales), of which the thickness was one fifth or below compared to AVT (0.6 to 3.0 mm → 0 to 0.5mm). The scale was soft and easily removed and maintained.

As for corrosion and erosion performances, no significant changes emerged between AVT and CWT except for erosion, as explained in 3.3.3.1.6-(3)-d ‘Parts Using Stellite Materials’ observed on the BFP overheat prevention valves.

Picture 3.3.1-3: Scales Deposited on the Flow Rectifying Tower of the High Pressure Feed Heater

Good

Bad

Fig. 3.3.1-25: Performances of a High Pressure Feed Heater

Abl

atio

n Vo

lum

e (m

m)

Fig. 3.3.1-26: Wear Volumes of the Chrome Plating

227

228

(d) High Pressure Water Feed Heater The color of scales for both the inner water chamber and inner heating pipe, as well as the flow rectifying tower,

changed from black magnetite to red-brown hematite, while X-ray analysis also confirmed an increase in hematite materials.

Meanwhile, the volume of scales was also subject to decrease. As shown in Picture 3.3.1-3, the hardened scales that had been removed during regular AVT inspection were rarely observed. As shown in Fig. 3.3.1-25, the volume of scales in CWT was one tenth or below (1,520→150g/year).

Due to the decreased scale, the differential pressure of a high pressure feed heater was decreased and, as shown in Fig. 3.3.1-25, the heat transfer performance (TD: Temperature Difference at the End Parts) was increased by 0.3 to 0.5°C, compared to AVT.

As for the corrosion and erosion performances, no significant differences were observed between AVT and CWT and a similar trend was observed for other types of heat exchangers (condensers and deaeration units). (e) BFP

A series of investigations was conducted on a rotator and strainer. As shown in Fig. 3.3.1-24, the volume of scales attached to the rotator was found to have decreased to approx. one fifth of that to AVT (460→100g/year unit) and no waveform scales, as observed in AVT, were seen around the CWT periphery. As for the strainer, the scales were around one sixth that of AVT (250→40g/year unit).

Regarding corrosion and erosion, as shown in Fig. 3.3.1-26, the wear volume of the chrome plating layer of the driving part of the rotator showed an increase compared to that of AVT (0 to 0.06mm→0.02 to 0.35mm in terms of the maximum volume at each stage). It was presumed that feed water containing oxygen had slipped into the micro pin holes and micro cracks, causing corrosions in the gap between the base materials and their plating layers and leading to removal of the layer.

In order to reverse the situation, the following corrosion tests were carried out using an actual unit and chrome plating combined by electroless nickel plating was found to be effective. [Test Procedures]

Location: Suction strainer of BFP Materials: Base material (13 Cr) Cr plating Composite electroless nickel plating Cr plating + electroless nickel plating Thermal spraying of oxidized Cr Periods: July 1992 to Jan. 1993 (for 6 months)

c. Stress Corrosion Cracking Test of Turbine Materials

In order to confirm the effects of CWT on turbine materials, a stress corrosion cracking test was conducted in the following procedures. To compare the results, the No. 2 unit using AVT was subject to a similar test. [Test Procedures]

Location: Low pressure 16th stage extraction steam chamber (dry) Low pressure 17th stage extraction steam chamber (humidity: 2%) Low pressure air discharge chamber (humidity; 7.5%) Materials: Blade materials (12CrNiMoV steel) Blade materials (12CrMoV steel) Rotor materials (3.5CrNiCrMoV steel) Stress added: 40kgf/mm2

60kgf/mm2

Exposure periods: CWT: 761 days AVT: 674 days

The following results were obtained and no adverse impacts were considered to be imposed on the turbine materials by CWT:

1) No stress corrosion cracks observed for both specimens 2) No significant differences in stress corrosion cracks were observed between AVT and CWT.

d. Parts using Stellite Materials As there are some reports in Europe of CWT causing corrosions to stellite materials, a test was conducted to

confirm the corrosion performances of the stellite materials. (a) Final Stage Blade of Main Turbine

Stellite materials (#6) are used at the tip of the blade to shield it against erosions. Although erosion was observed there, the extent was not as significant as that previously found in AVT (0.05 to 1mm/year).

(b) Main Turbine Control Valve

There was erosion due to the past boiler scale observed at the skirt of a sub-valve of the main steam stop valve, but its degree was relatively insignificant, like those used in AVT (about 5mm). (c) BFP Overheat Prevention Valve

Some damage was observed at the seat of a turbine driving the BFP overheat prevention valve (stellite #12) approx. 1.5 to 2.0 months after the start of CWT. The condition was significant compared to AVT and the interval over which the damage occurred was significantly reduced. The phenomenon was considered to be due to the high differential pressure attributable to carbonized stellite materials and erosion due to the high flow rate. Meanwhile, the selective corrosion was considered to be attributable to the oxygen fed into the system.

To reverse the situation, the valve seat materials were changed from metal to Teflon seat. Thereafter, no abnormalities were observed and the system has remained in good operating condition.

See the intermediate report (5) attached to this journal for details of the valve damage conditions, structural comparisons and materials used. (d) Spray Control Valve of a Superheater

About two years after the start of CWT, the seat of a secondary superheater spray control valve was subject to some damage (stellite #6). This affected the corners of the stellite and extended all over the area surrounding the flow path. In AVT, however, such damage was restricted to the base material side without involving the stellite layer. The cause of the damage was considered to be the same as that affecting a BFP overheat prevention valve. Picture 3.3.1-4 shows the conditions of the damage and its mechanism.

Methods of preventing such damage have already been established and employed by other power companies. The author plans to employ multiple coatings on the stellite surface (ceramic coating) and will examine the method in detail, including other coating methods.

229

Selective Corrosion of Carbonized Cr-W due to Oxygen

(e) Other Instrumental and Control Valves

No other problems were observed, even during PT tests, except for moderately selective corrosion at the stellite part (#6) of the seat ring at the high pressure feed heater drain control valve.

Damage Conditions

Carbonized Cr-W

Co-Cr-W crystals

Microscopic Picture of a Cross Section of the Damaged Part

Carbonized Cr-W Co-Cr-W crystals

Erosion of Co-Cr-W crystals due to the High Flow Rate

Direction of Flow

Damages Caused by Erosion and Corrosion

Damage Mechanisms

Picture 3.3.1-4: Damages to the Secondary Superheater Spray Control Valve

e. Cautions in Operations (1) How to Operate a Deaeration Unit

Based on experiences in Germany, the vent valve of a deaeration unit was kept closed from the onset of switchover to CWT. This, however, caused an abnormal surge in the DO concentration (600 ppb or above) at the outlet of the deaeration unit during the unit operation at low load. This was because, due to the closure of the vent valve, the high concentration of oxygen, which was deaerated during high load operation, was stagnant in the upper part of the unit, before expanding in volume, being scattered away and then redissolved in water due to the decreased pressure inside the unit.

To reverse this situation, the vent valve was left open. This, however, caused deaeration and discharge of DO, making the environment the same as AVT with low pH and causing the iron concentration at the outlet of the deaeration unit to surge.

Due to such experiences, the vent valve was again left closed, and only reopened when the DO concentration surged (intermittent operation). (2) Increased Differential Pressure of the Electromagnetic Filter (EMF)

Once the use of CWF had commenced, the initial differential pressure after backwashing and regeneration of an EMF occurred and its post-regeneration operation life was shortened (in 7 months, the operation life was shortened to one tenth (or 2 days) compared to that used in AVT). The reasons for this are believed to include: (i) the fact that CWT tends to have higher iron loads than AVT, (ii) an increase in fine particles of FeOOH (the number of FeOOH particles of 1 µm tripled or quadruplicated), which split into the depth of the element and (iii) the fact that needle-shaped iron crystals reinforced the iron deposit layer, which could not be removed by a backwash and regeneration process.

To reverse such conditions, jet washing of elements was employed and the elements were replaced with new ones.

In order to implement permanent measures, the following items are subject to examination: - (a) Decreasing iron loads at the inlet of an EMF (pH to be increased to 9.0 : effectiveness confirmed) (b) Improvement of the regeneration methods (c) Improvement of elements

f. How to Start and Stop the Unit For several months after the use of CWT, the unit cleanup time tended to be longer than for AVT, due to

unstable hematite protection layers and for other reasons. This was successfully solved through measures to improve the cleanup process, such as stabilization of the protective layers, lapping of the boiler and pre-boiler processes and an improved flow rate and numbers of swinging, as shown in Fig. 3.3.1-14.

Fig. 3.3.1-14: Cleanup Time

Acceptance Assessment

Cleanup Time (h)

-10 -8 -6 -4 -2 0

Water Treatment

Method under Normal

Operation

Let-up Time (h)

Hydrazine Injected or

not

AVT 24.5 Injected Deaeration Unit Pre-boiler Boiler

CWT 31.0 Injected Circulation of Condensate water

Deaeration Unit Pre-boiler and Boiler

CWT 29.0 Not injected Circulation of Deaeration Unit Pre-boiler and Boiler

230

231

Table 3.3.1-15: Optimized CWT Operation Method Item Controlled Values and Operation Method

Normal Operation Item Stop

DO (µg/l) pH Cationic conductivity (µS/cm)

*50 to 100 *8.5 to 9.0

0.2 or below * All control values referred to above are for feed water at the inlet of an economizer.

Start-up and Stop Item Startup Stop

Water Treatment Method

Conventional method (AVT) Conventional method (AVT)

AVT ↑↓

CWT Switchover Period

Minimum loads or above, and Electric conductivity of 0.2µS/cm or below

At least 3 hours before the planned time for disassembly of the unit Reason: All water must be circulated once during the period between the switchover from AVT and before shutdown of the plant.

Storage Conventional method (AVT) Leakage of seawater When seawater leakage happens, CWT must be promptly switched over to AVT.

Deaeration unit CWT: The vent valve of a deaeration unit must be subject to intermittent operation. AVT: The vent valve of a deaeration unit must be kept open.

Condensate water desalination unit

‘H’ shaped operation

An operation test without using hydrazine was conducted by stopping the unit (WSS), but no significant

differences in cleanup time and water quality were observed. However, hydrazine is a reducing substance and its use renders the hematite protective layer unstable, which means it may be better to avoid this substance.

In addition, the unit was subject to a startup and stop test while CWT was used in DSS, and no deterioration in water quality was detected.

The author is determined to continue studies and tests to establish optimum system operation when it is subject to stop and startup.

Table 3.3.1-16: Problems Attributable to CWT and Measures Problem Cause Emergency Measure Permanent Measure

Damage to the BFP overheat prevention valve seat

Erosion caused by high differential pressure and flow rate attributable to the selective corrosion of carbonized stellite

- To change to a soft seat (However, a large-sized valve must be developed for Kawagoe Thermal Power Plants Nos. 1 and 2.)

Damage to the valve seat of a secondary superheater spray control valve

Same as above (erosion due to high flow rate)

The use of stand-by inner valves

The use of ceramic coating, etc.

Wear of Cr plating at the sliding part of the BFP rotator

Erosion caused by corrosion between a plate layer and its base materials, attributable to oxygen having slipped from the cracks, etc. of the plate layer

Recoating of Cr plating during the regular inspection period

The use of homogeneous and defectless electroless Ni plating between a Cr plating layer and its base materials

Surge of differential pressure of EMF (clogged elements)

High iron load and an increase in iron oxide fine particles having slipped into the depth of the element

Chemical cleaning of elements

Increase of EMF bypassing

Others

To decrease the iron concentration at the inlet of an EMS (pH: 9.0)

To improve elements Others

3.3.1.6.4 Assessment

Based on the results of these studies and findings in and out of Japan, an assessment was made. As a result of the assessment, it was found that CWT can be applied to actual units, and is more reliable and

economical than AVT. The assessment details are explained below. (1) Optimized CWT Operation Method

The author believes Fig. 3.3.1-15 shows the optimized CWT operation method. In this case, no hydrazine will be used in startup and stop, and neither will any switchover from AVT to CWT take place. (2) Technical Assessment Comparing to AVT a. Corrosion of Components

232

Similarly to AVT, the author believes CWT will not have any particularly adverse impacts on the corrosion and erosion of components.

Moreover, the author also believes that, due to the problems associated with parts using stellite, no basic issues arise that could deny the CWT applicability. Permanent measures against foreseeable problems are shown in Table 3.3.1-16, while measures for the seat of a BFP overheat prevention valve have already been established. For other parts, examinations are ongoing. b. Powers to BFP, etc.

The author believes that the time course increase of steam used for driving BFP found in the acceptance performance test of the unit using CWT (total amount of heat of low pressure and high pressure steam) can be substantially decreased to approx. a third compared to AVT, because of the reduced differential pressure of boiler and of scales deposited.

The author also believes that the functional loss of a high pressure feed heater can be reduced to approx. two fifths compared to AVT, because of the reduced deposition of scale. c. Frequency of Chemical Cleaning of Boilers

As shown below, as regards the frequency of chemical cleaning of boilers using CWT, the author believes, in the case of the No. 1 unit of the Chita No. 2 Thermal Power Plant, that this can be extended from the current 1.5 years to 9.5 years, while for other once-through boilers of supercritical pressure or above, the current 1.5 to 4.5 year period can be significantly extended to 10 to 15 years.

In line with the extension in chemical cleaning frequency, the author believes that regular inspection periods can be shortened and the disadvantageous transfer of loads eliminated. (i) Assessment in terms of Boiler Differential Pressure

The interval of chemical cleaning for the unit used in this study, which uses AVT, was 1.5 years due to an increase in the boiler differential pressure. After switching over to CWT, this increase was eliminated, which meant the interval of chemical cleaning was extended.

Fig. 3.3.1-27 shows the trend of boiler differential pressure experienced by the Ulsan Thermal Power Plant of Korea Electric Power Corporation after switching to CWT. The operation time till the pressure rose to the acceptable limit was 9.5 years, whereupon chemical cleaning took place.

It is considered that the trend of the boiler differential pressure surge of the unit used in this study tends to follow a path of gradual increase compared to that experienced in Ulsan Thermal Power Plant. However, with certain allowances taken into consideration, it is estimated that it will take 9.5 years for the unit used by this study to reach the allowable limit for differential pressure. The author, therefore, believes that the interval of chemical cleaning for the unit can be extended from 1.5 to 9.5 years. (ii) Assessment based on the Volume of Scales Deposited on Generation Pipes

The acceptance criteria for the chemical cleaning of boilers employed by this company is around 30 to 45mg/cm2 of scale deposited on generation piles unless other problems, such as abnormal boiler differential pressure, are observed.

As indicated in 3.3.1.6-(3) ‘Main Components of Boilers,’ the speed of the scales generated will be 0.5mg/cm2・1,000h.

Thus, the chemical cleaning intervals calculated from the scale generation speed are 10 to 15 years, based on the conditions of the unit utilization ratio of 70% (30 to 40mg/cm2 ÷ 0.5mg/cm2 1,000h × 365 days ×24 hours ×0.7).

Based on the above arguments, in the case of other once-through boilers of supercritical pressure or above, for which no boiler differential pressure need be considered, the author believes that the interval can be extended from the current 1.5 to 4.5 years to 10 to 15 years. d. Vibration of BFPs

In the case of the unit used in this study, BFPs were not subject to any vibration, even when using AVT, and anti-vibration measures remained unconfirmed during the switchover to CWT. However, in the absence of any waveform scales and the fact that the volume of scales was reduced to approx. one fifth compared to AVT, the author believes that CWT can represent the ultimate measure against the vibration of BFPs. e. Environmental Aspects

The use of CWT can eliminate chemicals used to treat feed water and regenerate condensate water desalination units as well as effluents generated from the chemical cleaning of boilers, thus reducing effluent contamination loads.

Based on these aspects, CWT can be considered an environment-friendly feed water treatment method.

(Legend)

No. 4 unit of the Ulsan Thermal Power Plant

The unit used in this study

Boi

ler D

iffer

entia

l Pre

ssur

e (k

g/cm

2 )

Chemical cleaning

233

Fig. 3.3.1-27: Differential Pressure of Boilers for the Unit used in this Study

and the No. 4 unit of the Ulsan Thermal Power Plant

Time

(3) Cost Performance Comparison with AVT

The author believes that the use of CWT can help reduce costs significantly for BFP power losses, and function on the chemical cleaning of boilers and chemicals for the treatment of feed water.

The time course trend of costs for the use of CWT used by the unit under testing is estimated to be around a quarter of that incurred when AVT was used, as shown in Table 3.3.1-17.

On the other hand, the ratio of profitability (annual earning × investment amount × 100%), an indicator used to assess investment results, is 86%, and an effective cost-benefit effect can be expected from the use of CWT. 3.3.1.6.5 Introduction Plan

Based on the results and the assessment of this study, CWT was officially used at Chita No. 2 Thermal Power Plant in October 1993 when this study was completed, and a plan for introducing CWT to other once-through boilers was formed. (1) Applicable Units

For the following reasons, there is scope to apply CWT to all 18 units of supercritical and advanced ultrasuper critical once-through boilers:

a. 14 out of 18 boilers show increased power loss of BFP, due to a surge of differential pressure and problems such as vibration of BFPs arising from iron scales, for which the CWT can be an anticipated solution.

b. The cost-benefit calculation of using CWT shows that for all units, an effective cost-benefit can be expected (Ave. ratio of profitability: 66%)

(2) Periods of Introduction CWT will be introduced starting from FY1994 for periods of 5 to 6 years in a phased manner, from units where

the technical and economical merits of its use will be considerable.

Table 3.3.1-17: Annual Cost of the No. 1 Unit of the Chita No. 2 Thermal Power Plant

Item AVT CWT

Economization Ratio

Remarks

Cost of Power Loss for BFPs 100 33

67

Incremental increase of steam used for driving the BFPs (actual)

Cost of Functional Loss for High Pressure Feed Heater

100 40

60

Deterioration of heat transfer performance (actual)

Cost of Installing a Boiler Chemical cleaning System

100 16

84

AVT: Once in 1.5 years CWT: Once in 9.5 years (estimated)

Cost of Loss Transferred in relation to the above

100 16

84

Transfer days: 4 days/time (actual)

Cost of Chemicals for the Treatment of Feed Water

100 17

83

AVT: Ammonia and hydrazine CWT: Ammonia and oxygen

Cost of Operating a Condensate water Desalination Unit

100 32

68

AVT: 56 times (actual) CWT: 21 times (actual)

Total 100 28

72

3.3.2 Water Control and Management of Thermal Power Plants

Since the water-steam cycle, the vital artery for a thermal power plant, does not stop instantly in the event of any abnormality, meaning any accident in the cycle tends expand to become a long-term problem for the system, it should be served carefully and meticulously while in static condition as well as during daily inspections.

Each type and purpose of use of a plant has its own control criteria, of which the water quality criteria and treatment methods are delineated in JIS B 8223 ‘Boiler feed water and boiler water,’ (hereinafter referred to as JIS) although different water treatments are required depending on the environment where a plant is situated and when it was installed. Each water control staff member strives hard to investigate, test and try to uplift technologies of the most appropriate water treatment methods.

This chapter describes the current status, challenges and future prospects of water treatment implemented in the fields based on those perspectives.

Table 3.3.2-1 Raw Water Quality Monitoring Items and Measurement Frequencies Frequency Analyzing Item

Daily Weekly MonthlyRemarks

Turbidity pH Conductivity Ca2+ Mg2+ Fe2+ Alkali ions (Na+ and K+) By

calculation Cl- SO4

2- HCO3

- CO3

2- NO3

- Free carbon dioxide (CO2) SiO2 Total iron Residual chlorine (Cl2) COD Colloidal silica Water temperature

3.3.2.1 Water treatment of Makeup Water 3.3.2.1.1 Water treatment of Raw Water Sources

Raw water sources used by a thermal power plant include potable water and industrial water, which originate from rivers, ponds, lakes and underground water. Depending on the location of the thermal power plant, a water conversion system is used to convert seawater to fresh water, while effluents are also sometimes collected and recycled as a water source. The types and concentration of impurities contained in raw water depend significantly on the particular source of the water is taken or the season in question. It is necessary to understand the quality of raw water, not only in order to design a makeup water treatment system, used to remove impurities in raw water and maintain water quality suitable for makeup water, but also to use the system in a stable manner. Table 3.3.2-1

234

shows a list of water quality monitoring items generally applied for maintaining and controlling the system, and their measurement frequencies. The frequency must be increased whenever significant fluctuation is observed in the quality of raw water, or a new water source is employed.

Table 3.3.2-2: Turbidity Assessment Indicators

Water Quality Indicator

Measurement Method Characteristics

MF

Refers to the time required to filtrate a liter of water using a 0.45µm membrane filter (HAWP 047 TYPE HA made by Millipore Corporation) under reduced pressure of 500mmHg.

Easy to use, but vulnerable to water temperature and considerable size differences in filters. No linear relationship with the volume of contaminants contained in sampled water

MFC

Calculated using the following formula after the measuring times t1 and t2 required to filtrate 0.5l and 1l of water respectively under the reduced pressure of 500mmHg, and using the same measurement instruments as those used to measure the time of MF.

⎥⎦

⎤⎢⎣

⎡−= 1

ttln2MFC2

1

Has a linear relationship with turbidity, can measure consistently from raw seawater to processed water, and is a relatively newly proposed indicator.

FI (=SDI)

Calculated using the following formula after pressurized filtration of 500ml of sampled water at 2.1kg/m2g using a 0.45µm membrane filter (HAWP 047 TYPE HA made by Millipore Corporation) to measure the filtration time (T1). The filtration time (T2) is then measured by continuing filtration for 15 minutes after measuring T1, whereupon the same filter is used to filtrate a further 500ml of sampled water.

⎥⎦

⎤⎢⎣

⎡−=

2

1

TT1

15100FI

Highly sensitive to detect low turbidity and known as an inlet water quality indicator for a hollow fiber reverse osmosis module. As the turbidity intensifies, its sensitivity tends to drop. For seawater, normally T2=∞, or FI=6.67.

PI

Calculated by using the following formula after measurement in the same manner as FI.

FI15TT1100PI

2

1 ⋅=⎥⎦

⎤⎢⎣

⎡−=

Has the same characteristics as FI. PI=100% means the filter is totally clogged, and 0% means it is entirely unclogged.

PN

Means the total water volume (l) filtrated under a certain stable pressure environment using a membrane filter.

Easy to use, but needs a statement that the values are subject to change depending on the measurement environment (filtration pressure and the type and size of filter used).

Turbidity Measured using a scattering light type or an integrating sphere type turbidimeter.

Can conduct rapid and inline analysis, but tends to be less sensitive for low turbidity.

SS Calculates the volume of impurities based on a change in the volume of sample water filtrated through a 0.45µm membrane filter and the weight of the filter.

Requires considerable sample water for low turbidity measurement, and is unsuitable for rapid analysis.

3.3.2.1.2 Control of Pre-Treatment Unit to Maintain its Performance

There are two aims of pre-treatment: one is to protect the ion exchange resin, reverse osmosis membrane and ion exchange membrane, etc., used for the desalination unit and another is to remove colloidal matters that cannot be removed via an ion-exchange reaction. The methods generally used for the unit include the coagulating sedimentation filtration method and the coagulation filtration method. In order to maintain the unit in good operational condition, it is essential to form a stable floc, for which aluminum coagulation agents are generally used. Although previously, aluminum sulfate was frequently used, quite recently, polyaluminum chloride (PAC) and electrolytic aluminum have come into widespread use, since they have excellent coagulation performance and are less inclined to decrease pH. As an auxiliary agent for floc formation, organic polyacrylamide polymers, with bentonite to make the floc heavier, are used. Floc formation has many elements to take into consideration, such as the turbidity of the raw water, alkali level, pH and water temperature. It is, therefore, necessary to determine optimum coagulation conditions, including the volume of coagulation agents used, selection of auxiliary agents and pH by implementing a jar test, etc. Inappropriate floc formation will cause residual impurities in the raw water, and colloidal matters and aluminum (residual coagulation agent) infiltrating in the filtrated water. Colloidal silica is also contained in the boiler makeup water like ionic silica. When it is channeled into the water and steam systems, it will become scales deposited onto the low-pressure turbine. Similarly, colloidal aluminum tends to be deposited on high-pressure turbines in the form of aluminum oxide or sometimes zinc aluminate scales, which cannot be easily washed away by boiler chemical cleaning. Colloidal silica is controlled by monitoring the silica concentration in boiler water, and in the event of any anomaly, a comprehensive inspection of the raw water, filtrated water and deionized water must be conducted.

Table 3.3.2-3: Troubles and Measures for the Pre-Treatment Unit

235

Phenomenon Cause Measures or Check Item

Inappropriate use of chemicals Coagulation agent Coagulation auxiliary agent pH adjuster

Conduct a jar test to check that the coagulation conditions are appropriate.

Operate the unit under the appropriate injection ratio and filtrate water coming out from the coagulation sedimentation tank and the coagulation tank using a No. 5 filter. If the filtrated water is normal, then the problem lies on the filtration unit, as explained in the next chapter.

Incomplete regeneration of a filtration unit

Check whether impurities remain in the filtration unit when the normal course of regeneration process is done. If necessary, extend the backwashing time.

Check whether there is any shortage in the backwashing air and water flow.

Incomplete water collection unit under the filtration unit

Check the air dispersion conditions when air backwashing is run. If necessary, open the lower manhole to check it.

Defects to the backwashing trough of the filtration unit Wear or decreased filtration materials Channeling of filtration layers Mad balls

Check whether effluents from the backwashing process are evenly collected.

Consider the replenishment of filtration materials. Check whether any impurities and organic slime are found in the filtration layers.

Repeat the water and air backwashing processes several times. Remove the mad balls and replace part of the filtration materials.

Det

erio

ratio

n of

the

proc

esse

d w

ater

qua

lity

Floc strength Check whether the differential pressure is appropriate for operation by examining the differential pressure surge from the start of sampling water and the quality of the processed water.

Dec

reas

e in

am

ount

of

obta

inin

g w

ater

Insufficient opening of the valves Slime inside the filter layers or accumulation of foreign matter Miniaturization of the filtration materials

Insufficient working air pressure Damage to valves Operate air backwashing using a filtration material cleaner. Remove foreign matter having accumulated in the upper part of the filtration

material, if it hinders the discharge of effluents from backwashing. Replace the filtration materials on the surface of the filter layers

A slurry circulation type coagulation sedimentation unit requires the maintenance of slurry concentration at an

optimum level, while a sludge blanket type needs the sludge blanket to be kept stable. For both types, the key is how to adjust the volume of chemicals used as well as that of the sludge discharged from the units. Even without any dramatic fluctuation in the turbidity and alkali level of raw water, the water temperature varies seasonally, with a lower water temperature leading to a deterioration in floc formation performance. Experience states that the threshold temperature is at 10°C or so. If the temperature descends from this level, an auxiliary agent should be increased to facilitate the floc formation. In the case of a coagulation filtration unit, high water temperature excessively increases the size of the floc formed and when this happens, the volume of the auxiliary agent should be decreased. When a separation membrane is used for the desalination unit, the water quality, including the turbidity and FI (Fouling Index) at the outlet of a pre-treatment unit, must be maintained within the criteria determined for them. Table 3.3.2-2 shows some examples of turbidity assessment indexes.

In many cases, the problems of a pre-treatment unit are mainly caused by insufficient regeneration of filter layers due to contamination, which can sometimes be the result of mad balls and mechanical failure of the unit. When problems involving the filtration unit surface, these can be identified by a deterioration in the quality of processed water and reduction in the cyclic sampling volume. Table 3.3.2-3 shows the major problems anticipated to occur with the pre-treatment unit and their measures. 3.3.2.1.3 Control of Deionized Water Generation Unit for Maintaining its Performance

As makeup water for a high pressure boiler requires highly pure demineralized water, the quality of the makeup water must be controlled. Water treatment items for this purpose include electrical conductivity and silica, and an indicator and recorder with an alarm are installed at the anion tower of a deionized water generation unit and at the exit of a polisher to continuously monitor these items. Table 3.3.2-4 shows some examples of water quality criteria for makeup water used at thermal power plants. The actual values measured by each company, as indicated in the table, are 1.0µS/cm and 0.01mg/l or below for conductivity and silica respectively.

The following items are examples of daily control items, which should be monitored daily at a fixed time if the values are measurable:

Operation cycles and collection volume per cycle Volume of regeneration agents used and stored Temperature when chemicals are injected to regenerate the anion exchange resin Inner pressure of each ion-exchange resin tower (before and after regeneration) Conductivity of water sampled at the inlet of a deionized water generation unit Conductivity of processed water

236 Silica concentration in processed water

Ion-exchange resin should be controlled once a year to measure the total volume of resin replaced and volume

of neutral salt decomposed. At the same time, check the volume of ion-exchange resin remaining in each tower and inspect whether the resin is finely-divided or not. As for the replenishment of ion-exchange resin, the volume of resin to be replenished for cation resin is 5% -10% per year, including those finely divided, provided that no swelling and/or contamination is observed. In the case of anion resin, the volume to be replenished comes to 10% to 20% per year in general, because it is subject to aging besides the pulverization.

The troubles a deionized water generation unit may encounter mainly include decreased collection volume and deterioration of the quality of processed water (incl. high conductivity and silica concentration). These problems are often associated with pre-treatment unit troubles. Table 3.2.2-5 shows the cause of the troubles and their measures as well as check items.

Table 3.2.2-4: Examples of Criteria set by Each Electric Power Company for Water Quality at the Outlet of a Circulation Boiler Makeup Water Desalination Unit

Company Name A B C

237

Pressure Category (kgf/cm2)

Item 100 Class 130 Class 170 Class 100 Class 130

Class 170

Class 100

Class 130

Class 170

Class

Conductivity (µS/cm) 5 or below 1.25 or below 5 or below 3 or below

5 or below

1.0 or below

Silica (mg/l as SiO2) 0.05 or below

0.015 or below

0.01 or below

0.1 or below

0.05 or below

0.02 or below

0.1 or below

0.015 or below

Table 3.2.2-5: Troubles (reduced collection volume and decreased purity of processed water)

of a Deionized Water Generation Unit, Their Causes and Measures Root Cause Cause or Phenomenon of Trouble Measure

Change in Raw Water Quality

Increased total ion volume in raw water Change in the percentage of Na, HCO3 and SiO2 Increased organic substances and total iron volume

Conduct a total analysis of the raw water and file the data every month (Conductivity must be measured and recorded every month).

Check the water sources. Adjust the ratio of water intake from various water sources.

Oversampling Inappropriate water flow rate Failure of a flow meter Slippage of a flow meter due to the small volume of water passing through

Deteriorated water quality Failure of a meter Failure of a communicator Sampling failure

Compare the data with that of an instantaneous flow meter. Operate by uplifting the water flow. Refer to the instruction manual attached to the instrumentation unit.

Compare the data with that of a portable water quality meter, etc.

Connect a resistance box attached to the unit to a cable in lieu of a communicator. If the values coincide with each other, then the communicator is damaged.

Refer to the instruction manual attached to the instrumentation unit.

Sampling valve is too far closed or totally closed. Sampling valve is too open. Damages or water leakage to the communicator case.

Incomplete regeneration

Insufficient regeneration level

Inappropriate concentration of chemicals (Insufficient volume or excessive dilute solution used) Insufficient dispersion of regeneration agent

Clogged or damaged chemical feed pipe Decreased chemical feed speed (dispersed unevenly)

Channeling of resin layers

Regenerate the volume specified in the instruction manual. (Or increase the level of regeneration)

Feed chemicals at an appropriate concentration. Repair the damaged pipe. Failure of a chemical feed pump Clogs of ejectors and nozzles Excessive decrease of the diluted water flow rate Backwash for more than 30 minutes. Check and remove clogs from the lower water collection and dispersion unit.

238

Root Cause Cause or Phenomenon of Trouble Measure

Insufficient extrusion

Measure the specific gravity of regeneration effluents (to confirm whether the extrusion force is sufficient or not)

Insufficient flushing Analyze the washing effluents (to confirm whether the concentrations of Cl― of the H tower and Na+ of the OH tower are the same as those in their inlet position)

Shortage in chemical injection time

Make the total time for chemical injection and for extrusion at least the same as those designated

Inappropriate temperature for chemical injection

Make the temperature of chemical injection as 35±5°C (if the temperature is lower than this, silica will leak out and if it is higher, then the resin performance will be deteriorated)

Fluidization of resin layers when chemicals are injected upward

Readjustment of chemical injection volume and slip water volume

Incomplete regeneration

In the case of a multiple-layer system, a mixture of mild/strong acid resins and basic resins

Replacement of the mixture of resins in the middle position

Flow-out of ion-exchange resin

Increased backwashing speed Flow out of resins due to excessive backwashing speed

Breakage of the lower water collection and dispersion unit

Flow out of resins to the outlet of an ion-exchange resin tower

Fractured resins due to oxidizing substances Fractured resins due to pressurization

Check the backwashing speed Check the water temperature

Repair the damaged parts

Conduct a functional test of resins (coarseness distribution, etc.)

Contamination of ion-exchange resin

Existence of iron oxides and manganese in raw water (This contaminates mainly cation resins.)

Existence of organic substances in raw water

Check the pre-treatment unit.

Channeling Compressed resin layers The raw water is highly turbid.

Pulverization of resins due to chemical fracture Pulverization of resins due to a high flow rate operation or the internal pressure surge of a tank during operation

Insufficient or failed backwashing

Failure of lower and upper distributors

Backwash thoroughly, or conduct air backwashing to completely remove turbidity from the resin layers.

Remove residual chlorine in the raw water. Operate at an appropriate flow rate.

Backwash thoroughly. (Backwash for approx. 30 minutes, and stop it just before the outflow of resins from the tank.)

In the case of a heavily uneven surface on the resin surface, check and improve the upper water distributor so that water can be distributed evenly.

Inspect and repair it. Low flow rate sampling (when only the purity of processed water is decreased)

The sampling water flow rate is 5m/h or below. This normally results in the leakage of ions from an anion resin tower deteriorating.

Operate the unit at a high flow rate as much as possible. Don’t operate the unit below the minimum flow rate.

Leakage from a valve (when only the purity of processed water is decreased)

Valve failure Inspect and repair or replace it.

Deterioration of the ion-exchange resin function

Even in the case that there are no hazardous substances in the raw water, the resin function generally tends to slowly deteriorate.

Excessive temperature of chemicals (Never raise the temperature above 45°C.)

Replenish resins as designated. Measure the degree of functional deterioration and replenish the resins.

Keep the chemical temperature at an appropriate level.

3.3.2.2 Boiler Water treatment 3.3.2.2.1 Objectives and Methods of Water treatment

The degree and types of corrosions caused by water and steam and damage inflicted by corrosion products on feed water systems, boilers and turbine systems vary depending on the materials used and the temperature. This chapter outlines the type of damage caused to each component and how to prevent it. (1) Condensate Water

Steam used in a turbine is converted back to water using a condensing unit and the loss is replenished by feeding ion-exchange water as makeup water. The condensing unit of most plants uses copper alloy cooling water pipes, due to their good heat conductivity and anti-corrosion properties. As Fig. 3.3.2-1 shows, copper solubility is lowest at a pH of about 9 and in the actual unit, the relationship is similar. For a unit using copper alloy for the feed water system, the pH of condensate water is around 8.8 to 9.0, which is a zone within which the dissolution of copper is mostly restricted. For a unit using steel pipes, the pH reaches 9.3 to 9.6, making it difficult to curtail the dissolution of copper. In an environment where dissolved oxygen exists, the dissolved copper forms complex copper ammonium ions, which means the dissolution persists. In order to prevent this, titanium pipes are used for

the air cooling zones of a condensing unit and nickel plated copper alloy cooling water pipes are laid around the exterior of the unit.

239

Iron

Sol

ubili

ty

Iron

Con

cent

ratio

n

Solubility

Iron Concentration in Feed Water

Temperature

Fig. 3.3.2-2: Behavior of Iron in Feed Water and Condenser

Saturation solubility of Cu(OH)2 in deionized water

Saturation solubility of CuO in deionized water

Solu

bilit

y

Fig. 3.3.2-1: Relationship between the Solubility of Copper and pH

(Iron)

(Copper)

Con

cent

ratio

n

(Nickel)

Fig. 3.3.2-3: Shift in Metal Ion Concentration due to a Change in the Feed Water pH (2) Feed Water

Feed water refers to the water run from the outlet of a condensed pump to the inlet of an economizer, between which lies a heat exchanger. The temperature moves from 20 to 260 and various materials are used in these areas, such as iron, copper and nickel alloys. In the process of feed water, the method used to prevent impurities in water channeled to the boiler, to prevent the generation of scales in the latter, as well as how to avoid corrosion are very important. Fig. 3.3.2-6 shows the JIS criteria. The level of dissolved salts and other impurities contained in

240

water is close to zero under normal operation, thanks to the upgraded performance of a deionized water generation unit. However, in the case of seawater leakage, dissolved salts slip into the system. Depending on which materials are used for the feed heater pipes, namely copper alloy or steel, the means used to control pH to prevent corrosion of the materials vary. In the case of copper alloy, the pH control mainly targets copper and the value is limited to 9 or so, because dissolved copper ions accelerate the corrosion of iron. The slight volume of ammonia generated by thermal decomposition of the hydrazine used as a deoxidizer is used to control the pH. Fig. 3.3.2-2 shows a behavior model of iron in condensate water and feed water systems. The higher the temperature rise, the more iron ions are generated. Around a high pressure feed heater, the iron ion concentration goes beyond the iron solubility curve meaning ion deposits are generated. The iron is then deposited in the high temperature zones of high pressure feed heaters, economizers, boilers and other units. In order to prevent such scales from being deposited, it is important to minimize the iron ion volume. Fig. 3.3.2-3 shows a change in the iron ion concentration in a course of a shift in the pH of feed water. As indicated, the pH should be kept high. Fig. 3.3.2-4 shows the solubility of magnetite (Fe3O4), a corrosion coating. In the high temperature zone, the curve bottoms out at a pH level of around 10. As iron ions tend to accelerate its oxidization under the existence of dissolved oxygen, hydrazine is added to the feed water to remove oxygen, so that the generation of dissolved oxygen can be minimized.

N2H4+ → O2+2H2O A unit using steel pipes is subject to a control pH at around 9.5. Ammonia is directly added to the outlet of a

condenser or deaeration unit as a pH adjuster, and hydrazine is added at the outlet of a deaeration unit as a deoxidizer respectively. (3) Boiler Water

Most substances dissolved in boiler water are separated out as the temperature rises, due to the low solubility and known as scale and sludge. This scale and sludge has low thermal conductivity, causing thermal efficiency to deteriorate and corroding the boiler generation pipes. With this in mind, the generation of scales and sludge must be avoided as far as possible. As shown in Fig. 3.3.2-3, the pH of the boiler water must be retained high to prevent corrosions of generation pipes. Silica contained in boiler water flows out to the steam side and is separated out on the turbine blades as scales, causing the efficiency of the unit to deteriorate. Therefore, silica concentration must be minimized as far as possible.

Table 3.3.2-6: Feed Water Quality

[Drum Type]

Cat

egor

y Max. Operating Pressure (MPa or kgf/cm2) Evaporation Rate of the Heat Transfer Surface (kg/(m2 h)) Types of Makeup Water

10 - 15 (100 - 150)

-

Ion-exchange water

15 – 20 (150 – 200)

-

Ion-exchange water

Feed

Wat

er

pH (at 25°C) Hardness (mgCaCO3/l) (mg/l) Dissolved Oxygen (mgO/l) Iron (mgFe/l) Copper (mgCu/l) Hydrazine (mgN2H4/l)Conductivity (µS/cm) (at 25°C)

8.5 – 9.6 0

0.007 or below 0.03 or below 0.01 or below 0.01 or above 0.5 or below

8.5 – 9.6 0

0.007 or below 0.02 or below 0.005 or below 0.01 or above 0.5 or below

[Once-through Type]

Max. Operating Pressure (MPa or kgf/cm2) 15 - 20 (150 - 200)

20 or above (200 or above)

Cat

egor

y

Treatment Method Volatile

Substance Treatment

Oxygen Treatment

Volatile Substance Treatment

Oxygen Treatment

Feed

Wat

er

pH (at 25°C)

Conductivity (µS/cm) (at 25°C) Dissolved Oxygen (mgO/l) Iron (mgFe/l) Copper (mgCu/l) Hydrazine (mgN2H4/l) Silica (mgSiO2/l)

8.5 – 9.6 0.3 or below

0.007 or below 0.02 or below

0.003 or below 0.01 or above 0.02 or below

6.5 – 9.0 0.2 or below 0.02 – 0.2


- 0.02 or below

9.0 – 9.6 0.25 or below 0.007 or below 0.01 or below 0.002 or below 0.01 or above 0.02 or below

6.5 – 9.0 0.2 or below 0.02 – 0.2


- 0.02 or below

241

Ir

on Io

n C

once

ntra

tion

(mol

/kg)

Fig. 3.3.2-4: Relationship between the Magnetite Concentration and pH

Fig. 3.3.2-7 shows the JIS criteria. For a drum type boiler, sodium phosphate is used in the boiler intermittently to control the pH of boiler water. Substances dissolved in the boiler water, such as silica and chlorine ions, are removed by blowing the boiler water, while in the case of a once-through boiler, all substances dissolved in the boiler water are separated and deposited as scales, since it lacks any air-water separation mechanisms. This means the pH of a once-through boiler is controlled by using ammonia, a volatile chemical, and by installing a desalination unit at the outlet of a condensate water generation system to remove dissolved substances from the condensate water.

Fig. 3.3.2-7: Boiler Water Quality Criteria

Cat

egor

y Max. Operating Pressure (MPa or kgf/cm2) Evaporation Rate of the Heat Transfer Surface (kg/(m2 h)) Types of Makeup Water

10 - 15 (100 - 150)

-

Ion-exchange water

15 - 20 (150 - 200)

-

Ion-exchange water

Treatment Method Sodium phosphate treatment


Sodium phosphate treatment


pH (at 25°C) 8.5 – 9.8 8.5 – 9.6 8.5 – 9.8 8.5 – 9.6 Oxygen consumption (pH at 4.8) (mgCaCO3/l) - - - - Oxygen consumption (pH at 8.3) (mgCaCO3/l) - - - - Total residue on evaporation (mg/l) - - - - Conductivity (µS/cm) (at 25°C) 60 or below 20 or below 60 or below 20 or below

Chlorine ion (mgCl－/l) 2 or below 1 or below 2 or below 1 or below Phosphate ion (mgPO4

3-/l) 0.1 – 3 0.1 – 3

Sulfurous acid ion (mgSO32-/l) - - - -

Hydrazine (mgN2H4/l) - - - -

Boi

ler W

ater

Silica (mgSiO2/l) 0.3 or below 0.2 or below

Fig. 3.3.2-8: Steam Quality Criteria Item Criteria

Conductivity (µS/cm) (at 25°C) Silica (mgSiO2/l)


Table 3.3.2-9: Facility Outline of Unit Nos. 1 to 4 Units of the Nishi Nagoya Thermal Power Plant Unit No. 1 Unit No. 2 Unit No. 3 Unit No. 4 Unit

Output 220 000 kW 220 000 kW 375 000 kW 375 000 kW Operation started in: July 1970 December 1970 July 1972 September 1972 Boiler type Reheating natural

circulation type Reheating natural circulation type

Reheating natural circulation type

Reheating natural circulation type

Turbine steam pressure 169 kg/cm2G 169 kg/cm2G 169 kg/cm2G 169 kg/cm2G

(4) Steam Impurities contained in steam are those carried over from the boiler water, which cause corrosion of superheater

systems and the deposition of scales onto turbine blades. Such impurities include chlorine ions, sodium, silica and copper. Fig. 3.3.2-8 shows their JIS criteria.

As success in steam quality control is significantly dependent on the boiler water quality, for a drum type boiler,

the conductivity of boiler water and its silica content, and - for a once-through boiler - those of feed water at the inlet of an economizer should be carefully monitored. 3.3.2.2.2 Water treatment of a Drum Type Boiler

Table 3.3.2-9 shows an outline of facility of unit Nos. 1 to 4 of the Nishi Nagoya Thermal Power Plant. The following are the explanation of water treatment of a drum type (reheating natural circulation) boiler: (1) Water treatment under Normal Operation

Fig. 3.3.2-5 shows the water treatment system for a drum type boiler. Table 3.3.2-10 shows the water treatment values.

a. Water Treatment of Condensate water and Feed Water The method of water treatment for condensate water and feed water depends on the materials used for a high

pressure feed heater, but as for the Nishi Nagoya Thermal Power Plant, its unit Nos. 1 to 4 use copper alloys and hydrazine is used for water treatment. Under normal operation, thin hydrazine (0.5% N2H4) is added to the outlet of the condensate pump in order to maintain the pH of the feed water (at the entrance of an economizer) at 8.8 to 9.0. Since, under normal operation, the conductivity at the inlet of a deaeration unit is commensurate with the pH at the inlet of an economizer, hydrazine is added based on the feed water volume after the measurement of the same. The conductivity at the inlet of a deaeration unit is 1.3 to 2.0 when pH is maintained at a level of 8.8 to 9.0 and the pump stroke comes to 15 to 20%. Under normal operation, the water quality of each system is maintained at the normal values as shown in Table 3.3.2-10.

242

To high pressure turbine

To medium- and low-pressure turbines

From high pressure turbine

Drum

Deaeration unit Condenser

Fig. 3.3.2-5: Water Treatment System Diagram of a Drum Type Boiler b. Boiler Water Treatment

Controlling the pH at the inlet of an economizer at 8.8 to 9.0 makes the pH of boiler water 8.7 to 8.9 or so. As the unit uses hydrazine only to control the pH, the boiler water lacks any alkali elements, while its lack of any hardness removers (sodium phosphate), also makes it vulnerable against the infiltration of corrosion products from feed water or impurities (Ca and Mg, etc.) brought into boilers from seawater leakage. Accordingly, for a unit frequently started and stopped, sodium tertiary phosphate (2% Na3PO4) is added to the boilers at the stroke of

Flush tank

Flush valve

Flush pipe drain tank

Deaeration unit water tank Conden

sate water pump

Ground steam

condenser

Makeup water tank

Makeup water pump

Flooding pipe of boiler Sampling point

30% to maintain the pH in the boilers at 9.0 to 10.0. c. Frequency and Method of Off-the-System Blows

Even under normal operation conditions, corrosion products generated from condensate water and feed water systems and pipes are concentrated in boilers. As there was a report that a generation pipe of a unit similar to this ruptured due to the detachment of scales containing zinc, for boilers in Nishi Nagoya Thermal Power Plant, an off-the-system blow process is applied to 20 to 25 t of boiler water every two days when the units have been in operation continuously for a week. The off-the-system blow is always performed as necessary, whenever the units are in continuous operation, when the concentration of corrosion products (Cu, Zn, Ni and Fe) in the system is high and when the conductivity of the boiler water is high. (2) Water treatment at Start-up and while the Unit is not used a. Stop Time Water treatment

Various methods of chemical injection and storage are used, depending on the necessity to apply anti-corrosion measures to the system, requirements for the early startup of power supply and cost performance.

Table 3.3.2-10: Water treatment Values at Normal Operations

243

Criteria Normal Operation Item Specimen

Control Item Unit Measurement Method


Phosphate Treatment

Value ANN Value

Makeup water

Outlet of a makeup water tank (Nos. 3 and 4) Conductivity µS/cm Measurement

instrument 1.5 > Same as left 0.5 – 1.0 1.5

Water in condenser Conductivity µS/cm Measurement instrument * * 0.15 0.5

pH * Anytime 8.6 – 9.0 Same as left 8.8 *

Conductivity µS/cm Measurement instrument 0.3 > Same as left 0.15 0.3

Total iron µg/l Once a year (10 >) Same as left 10 *

Water at the outlet of the condenser

Total copper µg/l Once a year (5 >) Same as left 3 * Water at the inlet of low pressure No. 3 feed heater

Dissolved oxygen µg/l Measurement

instrument 40 > Same as left 10 40

Conductivity µS/cm Measurement instrument * * 1 – 2 *

Total iron µg/l Once a year (10 >) Same as left 10 *

Water

Water at the inlet of the deaeration unit

Total copper µg/l Once a year (10 >) Same as left 5 * Dissolved

oxygen µg/l Anytime 7 > Same as left 2 *

Total iron µg/l Once a year * * 10 * Water at the outlet of the deaeration unit

Total copper µg/l Once a year * * 5 *

pH * Measurement instrument 8.6 – 9.0 Same as left 8.9 8.9 9.0

Conductivity µS/cm Measurement instrument 0.3 > Same as left 0.15 0.3

Total iron µg/l Once a year 10 > Same as left 10 * Total copper µg/l Once a year 10 > Same as left 5 *

Feed water

Water at the inlet of an economizer

Hydrazine µg/l Anytime (10 – 30) Same as left 10 *

pH * Measurement instrument 8.6 – 9.0 8.6 – 9.5 8.7 8.6 9.5

Conductivity µS/cm Measurement instrument 0.3 > 15 > 1.5 3

Silica µg/l Measurement instrument 0.2 > 0.2 > 0.05 0.2

Phosphate ion µg/l Anytime * 3 > 0 – 3 *

Total iron µg/l Once a year * (50 >) 20 – 50 *

Boiler water Drum water

Total copper µg/l Once a year * (20 >) 5 - 15 *

Conductivity µS/cm Measurement instrument 0.03 > * 0.1 0.3 Steam Steam at the inlet of

superheater Silica µg/l Anytime 0.02 > * 0.005 * Note 1: The figures in parentheses refer to the values that should be maintained. Note 2: The figures prefixed with refer to the value after the treatment of cation resins. Note 3: The pH level of the boiler water is 8.5 to 9.0 after all volatile treatment and 8.5 to 9.5 after the injection of sodium tertiary phosphate.

244

b. Startup Water treatment Just after the startup of units, water quality tends to be subject to considerable fluctuation and is under threats

such as seawater leakage and other water quality problems and measurement instrument failures. If these problems are left unattended, a serious accident will occur. For this reason, water quality targets, blowing procedures and chemical injection procedures are established for each stop time period. A standard startup command (operation) sheet shown in Fig. 3.3.2-6 is used to confirm the water quality for each stage to control the startup water quality, while the standard patterns for dissolved oxygen, silica concentration and conductivity of water are as shown in Fig. 3.3.2-6. As for the conductivity in particular, a standard pattern for each stop time period is formed, which is accessible when a recorder is located in the central control room. This pattern is subject to a comparison check with the current values by a power plant staff member so that the trend can be monitored and controlled. c. Off-the-System Blowing

This type of blowing is always implemented till the water quality at the inlet of an economizer and saturated steam descend below the criteria and till the conductivity shows a falling trend of 0.5 to 0.6µS/cm. (See Fig. 3.3.2-11) d. Control of Silica Concentration in Steam

It is known that silica deposits are hardly formed on turbine blades when its concentration is below 0.02mg/l. The silica concentration in steam can be controlled indirectly by managing the level in boiler water. After opening the system for regular inspections, dust, which is allegedly the source of dissolution of silica slips into the system, is often carried to a boiler via feed water pipes, where it is then deposited as ionic silica under an environment of high temperature and pressure. Thus, especially just before system startup after regular inspection, it is vital to implement the replacement of water and continuous blowing of boilers as practically as possible to reduce silica concentration in the system, so that silica can be purged completely and swiftly.

Up

load

s (22

0MW

)

Load

dis

patc

hing

ope

ratio

n

Wat

er q

ualit

y ch

eck

Load

dis

patc

hing

op

erat

ion

Up

load

s (10

0MW

)

Wat

er q

ualit

y ch

eck

Para

llel i

n

Switc

hove

r with

in

the

plan

t

Fig.

3.3

.2-6

: Sta

ndar

d St

artu

p C

omm

and

(Ope

ratio

n) S

heet

and

Sta

rtup

Wat

er Q

ualit

y

Prep

arat

ion

for t

he st

artu

p of

turb

ines

Para

llel i

n

Wat

er q

ualit

y ch

eck

Boi

ler w

ater

Star

tup

of tu

rbin

es

Turn

ing

on o

f a

disc

onne

ct

switc

h

Feed

wat

er

Satu

rate

d st

eam

Con

dens

ate

wat

er

Feed

wat

er

Con

dens

ate

wat

er

Prep

arat

ion

for t

he st

artu

p of

turb

ines

Star

tup

of tu

rbin

es

Boi

ler w

ater

Wat

er q

ualit

y ch

eck

Igni

tion

and

tem

pera

ture

/pre

ssur

e up

Up

of v

acuu

m

245

Command by the deputy manager Operation check Water quality pattern

Prep

arat

ion

for

star

tup

Prep

arat

ion

for i

gniti

on

Igni

tion

Tem

pera

ture

up

Pres

sure

up

Sw

itcho

ver o

f bur

ners

Feed

wat

er

vacu

um u

p

Che

ck it

em

Ope

ratio

n ite

m

Mai

n st

eam

pre

ssur

e

Prep

arat

ion

for i

gniti

on

Star

tup

of c

onde

nser

W

ater

qua

lity

chec

k St

art u

p of

the

feed

pum

p St

art i

njec

tion

of h

ydra

zine

[Dis

solv

ed O

xyge

n]

(Con

dens

ate

wat

er)

The

conc

entra

tion

of d

isso

lved

oxy

gen

falls

alo

ngsi

de th

e va

cuum

up

in th

e co

nden

ser.

(Fee

d W

ate r

) Th

e di

ssol

ved

oxyg

en c

once

ntra

tion

decl

ines

alo

ngsi

de th

e su

rge

of p

ress

ure

of th

e de

aera

tion

unit.

[Sili

ca]

Whe

n tu

rbin

es a

re st

arte

d, si

lica

scal

e de

posi

ted

on th

e lo

wpr

essu

re b

lade

s com

es o

ff. (P

art o

f the

silic

a sc

ale

com

es

from

the

dead

spac

e of

the

pre-

boile

r sys

tem

.) Th

e si

lica

conc

entra

tion

com

es d

own

as th

e bl

ow

oper

atio

n st

arts

.

[pH

] (F

eed

Wat

er)

Due

to th

e in

ject

ion

of h

ydra

zine

at s

yste

m ig

nitio

n, th

e pH

le

vel r

ises

. (B

oile

r Wat

er)

Due

to th

e in

ject

ion

of h

ydra

zine

at s

yste

m ig

nitio

n, th

e pH

le

vel o

f the

feed

wat

er ri

ses,

as w

ell a

s tha

t of t

he b

oile

r w

ater

.

[Con

duct

ivity

] (C

onde

nsat

e w

ater

) W

hen

the

syst

em is

subj

ect t

o a

para

llel o

ff, t

he

cond

uctiv

ity o

f the

con

dens

ate

wat

er ri

ses d

ue to

the

influ

ence

of c

arbo

n di

oxid

e ga

s con

tain

ed in

the

air,

but

falls

as t

he d

egre

e of

vac

uum

incr

ease

s. (F

eed

Wat

er)

Sim

ilar t

o co

nden

sate

wat

er, t

he le

vel f

alls

as t

he d

egre

e of

va

cuum

incr

ease

s. H

owev

er, t

he d

ownw

ard

curv

e te

nds t

o be

mod

erat

ed fo

r abo

ut 1

hou

r fro

m th

e pa

ralle

l in

poin

t be

caus

e of

con

tam

inat

ion

of th

e fe

ed w

ater

, con

dens

ate

wat

er a

nd th

e w

hole

syst

em.

(Sat

urat

ed S

team

) Fo

r abo

ut 1

hou

r fro

m th

e ju

xtap

ositi

on p

oint

, co

ntam

inat

ion

of b

oile

r wat

er is

con

spic

uous

. Due

to th

is,

the

leve

l of c

ondu

ctiv

ity te

nds t

o pe

ak d

ue to

a c

arry

over

an

d fo

r oth

er re

ason

s, be

fore

subs

eque

ntly

follo

win

g a

mod

erat

e do

wnw

ard

cour

se.

Fig. 3.3.2-11 Methods of Blow Except the Boiler Water System at Startup

Boiler water blow is implemented until water quality of the eco inlet, the boiler, and thesaturated steam is at the standard value or less, and the conductivity at the CP outlet shows a declining trend (0.5 to 0.6µS/cm).

S aConductivity oEoBoS

Conductivity 0CP outlet S

The silica concentration in boiler water sometim

sealed (steam sealing) after the unit shut down. The csealed, silica scale separates from the turbine bladesstarted. To reverse such situation, the following meaavoid boiler blow loss and gaining load up during thewhile the boilers are sealed, the silica concentration more, water is added to the condensate water after toimplementation of the measure, no abnormal silica co(3) Water treatment at Seawater Leakage

It is essential to detect any seawater leakage at an water quality depends on the seriousness and conditisought. For this purpose, conductivity is measured winlet of an economizer and in the boiler water. WheTwo lines of cation resin towers are installed at the foutlet of the condensate pump, so that the letup timemergencies such as seawater leakage. How to injeccontrolling water quality in the event of seawater leakcorrect action is decided. For this purpose, a seawatersuch items as ‘operation method,’ ‘actions done’ and ‘(4) Water treatment Values and Monitoring

Water quality during the normal operation timpre-determined control items and the measurement fCRT and recorders of the central control room, whesystem operation. In the case of an accident, an alarmdefined in ‘Water Quality ANN Messaging Procedure

As water quality requires monitoring of long-terstable operations of the unit are assured and good measurement data of iron and copper concentratioinspection and other measures to ensure that water is 3.3.2.2.3 Water treatment of Supercritical Pressu

There are two types of once-through boileronce-through boiler. The basics of water treatment In this chapter, the methods employed by unit Nos. Table 3.3.2-13 shows the basic unit configuration of t

ilic
co inlet
oiler water

aturated steam

r less

r less

r less
→Boiler water blow stop
(ANN of silica and conductivity

high have to be reset.)
.5 or less
hall be on a down note.

es exceeds the standard value (0.2mg/l) when the boiler is ause seems to be the fact that when the boiler is stopped and , etc., is channeled to the boiler and concentrated when it is sure is implemented while the boilers are sealed in order to early stages. When the level of water in a condenser goes up in condensate water is measured. If the value is 0.02 mg/l or tal blowing, whereupon the system is started. Following the ncentration up was reported.

early stage to implement measures. The means of controlling ons of the leakage, with appropriate water control measures ithin the condenser, at the outlet of the condensate pump, the n the value is found to be high, seawater leakage is present. ront stage of the condenser and the conductivity meter at the e for replacing the resin can be minimized, in the case of

t chemicals and how to blow boiler water are determined as age. Also, in a leakage, a process which requires prompt and leakage accident control sheet (Fig. 3.3.2-7) is used to cover restoration.’

e is monitored and checked for each specimen, using requency. All measurement values can be monitored by the re staff members are stationed to monitor during the normal is activated. Accidents are handled referring to the measures ’ (Fig. 3.3.2-12). m trends, a daily control sheet is formed. Appropriate and water treatment methods are established by collecting total n and in-house inspection results obtained from a regular appropriately treated.

re Through a Flow Boiler , a subcritical once-through boiler and a supercritical for both types of boiler are the same, with minor differences. 1 to 4 of the Sodegaura Thermal Power Plant are explained. he Sodegaura Thermal Power Plant.

246

247

4. Restoration measures

3. Causal investigation

2. Operation just after the accident

1. Occurrence of an accident

Restoration measures

‘Conductivity high’ ANN turns on. Causal investigationAccident handling

Point of leakage (1) Caustic silver check (2) Switchover of thermometer

takeout points Repair completed(1) Conductivity surges in the order of condensate water, feed water and boiler water.

(2) Check the conductivity of the confluence points of desalinated water.

(1) Boiler water blowing (2) Injection of sodium tertiary

phosphate (1) Condenser restored to

normal (2) Boiler steam flushing (S

Note 2) ee

(3) Restriction of load liftedSeawater temperature at the inlet of condensing water generation unit at 20°C or below (See Note 1) Vacuum of condenser at 690mmHg or above If these can be maintained: Load: 220MW

Seawater temperature at the inlet of the condensing water generation unit at 20°C or below (See Note 1) Vacuum of condenser at 690mmHg or above If these can be maintained: Load: 200MW of below

(1) Conductivity of condensate water, feed water and boiler water

(2) Degree of vacuum of condenser(3) Temperature of the air discharge

chamber (4) Conductivity deteriorated in the

order of condensate water, feed water and boiler water

One-side operation of condenserWater Quality Target of Boiler Water during Seawater

Leakage µS/cm of recovery

water Boiler water 0.3µS/cm or below Higher than 0.3µS/cm

(1) pH of boiler water: 8.0 or below

(2) Saturated steam conductivity: 1µS/cm or above

(1) Stop addition of sodium tertiary phosphate when the chlorine ion concentration in the boiler water goes below 0.2 ppm.

(2) Stop the blowing of boiler water

(1) Degree of vacuum of condenser (2) Temperature of the air discharge ch(3) Vibration of turbine, position of lid

amber s and

axis, difference in the temperature of metals Conduc(4) tivity deteriorated in the order of condensate water, feed water and boiler water

3 or below

Stop the unit. Conductivity (µs/cm) As low as possible 15 or below

Turbine E-STOP push button ON

Note 1: In order to avoid any stress corrosion cracks of SH and RH pipes, boilers should be subject to vapor washing when the unit is restored from seawater leakage (at the turbine rotation of 3,600 rpm).

Note 2: Considering the working environment while only one condenser is in operation, the vacuum is set as 690 mmHg or above (according to a test result in 1980).

Reference information: The conductivity of condensate water was once seen to surge due to pinhole damage of the exterior housing of a condenser water pump.

Fig. 3.3.2-7: Accident Handling Sheet for Seawater Leakage (1) Water treatment during Normal Operation a. Water treatment

As for the water control of a once-through boiler, in order to minimize the separation out of metals, due to the corrosion of materials used in the system, it is necessary that the pH and dissolved oxygen concentration should mainly and always be retained within the controlled values. For this purpose, ammonia and hydrazine are added continuously to the outlet of the condensate water pressure surge pump. The ammonia aims to control the pH to a level of 9.3 to 9.6 and for this purpose, the conductivity at the inlet of the deaeration unit is maintained at a constant value and the condensate water flow rate is controlled in proportion to the ammonia injection ratio. For hydrazine, the chemical is added at the rate proportional to the feed water flow rate to control the concentration of dissolved oxygen at 7µg/l or below by controlling the residual hydrazine concentration at the inlet of an economizer at 10µg/l or above. See Table 3.3.2-14 for the details of other water treatment criteria under normal operational conditions.

b. Analysis Items and Frequency of Analysis

Table 3.3.2-15 shows the analysis items and the frequency analysis. c. Control of the Condensate water Desalination Unit

A once-through boiler is unable to conduct the off-the-system blowing of boiler water under normal operational conditions. In order to remove impurities infiltrating the system from outside, a condensate water desalination unit is installed to remove corrosion products produced from within the system and other impurities originating from outside the system due to the seawater leakage of a condenser. The condensate water desalination unit is an ammonex type of the outside regeneration type. It functions quite similarly to a separate regeneration type system in reducing the regeneration time and reduction of nitrogen in effluents and is maintained and controlled considering the items shown below.

a). To check the area where resin layers are detached The water regeneration tank of a mixed bed system, most of the causes leading to the deterioration of water

quality are derived from the movements of resin layers detached, the tank is subject to checking after scrubbing.

248

Table 3.3.2-12: Check Sheet for Unit Nos. 1 to 4 Water Quality When ANN is Transmitted 3. Abnormal pH Level

Check Item Description Cause 1. Silica and Dissolved Oxygen * Whether silica and dissolved oxygen concentration tend to be high or not is checked by referring to the checksheet.

Feed water pH High (or low) b d

Check Item Description Cause High b d h

Boiler water pH Low a b d

Makeup water There is considerable conductivity of makeup water. c d

Silica concentration is high.

Makeup water

1.5µS/cm or above for the makeup water 10µg/l or above at the outlet of the makeup water desalination unit

c

Load Load surged (when silica is purged)

i

Chemicals Chemical concentration Failure of pumpsAddition of chemicals

Chemical concentration and type Switchover test of injection pumps Stroke and valve operations

b

Measurement instrument

The ANN of a measurement instrument is transmitted.

d

Cause and Measures

Cause Measures The dissolved oxygen concentration is high.

Load

Load decreased

e

Deaeration unit Check the inner pressure of the system.

f

Switchover of the condensate pump

Check whether any O2 is leaked or not by switching the condensate pump

g

Drain pump Check whether any O2 is leaked or not at the drain pump seal.

g


The indication does not change after switching the specimen water.

d

2. High conductivity/Condenser・High conductivity/Seawater

leakage at startup

Check Item Description Cause The conductivity of the condensate water and water in the system suddenly surges from the normal level, followed by feed water, boiler water, saturated water. The makeup water is intact.

a

a. Seawater leakage b. Excessive (too small)

hydrazine injection c. Abnormal quality of

makeup water Leakage of regeneration agent of a desalination unit and sampling after breakage

d. Fault of measurement instruments

e. Decreased loads f. Deteriorated deaeration

unit g. O2 leakage from

condensate pump and drain pump

h. Excessive addition of sodium tertiary phosphate

i. Surge of loads

a. See the section of seawater leakage.

See Accident Action Procedure. b. Adjust the injection volume. c. In the case of abnormal quality of

makeup water, blow the water in the makeup water tank.

d. If adjusting the flow rates and

temperature does not work, inform Chemical G.

g. Seal the leakage. h. Adjust the volume. i. Blow the boiler water. See the section explaining the

relationship of silica in boiler water and pressure.

Check the relationship of the following: Condensate water, feed water, boiler water, saturated steam and makeup water in the system

The conductivity of the makeup water surges, followed by others such as condensate water.

c

Chloride ions Adding caustic silver changes the water turbid in white. a


Their conductivity fluctuates separately. d

b). Regeneration Process Check

The entire regeneration process is subject to a detailed check once a year to confirm the conditions of the flow rate and the volume of chemicals added, to determine abnormalities at their early stage and to plan suitable countermeasures. c.) Confirmation of Functions of Ion-Exchange Resins

Performance of exchanging neutral salts, chemical reaction speed and other aspects are checked once a year.

Table 3.3.2-13: Facility Outline of Unit Nos. 1 to 4 Units of the Sodegaura Thermal Power Plant Unit No. 1 Unit No. 2 Unit No. 3 Unit No. 4 Unit

Output (MW) 600 1 000 1 000 1 000 Operation started in: August 1974 September 1975 February 1977 August 1979 Boiler type Reheated

once-through type

Same as left Same as left Same as left

Boiler capacity (t/h) 1 900 3 110 Same as left 3 170 Steam pressure (kg/cm2) 246/42.1 246/40.1 Same as left Same as left Steam temperature (°C) 538/566 Same as left Same as left Same as left Fuel used LNG Same as left Same as left Same as left

d. Replenishment of Resins Referring to the results of the performance check, resins are replenished to maintain the function of a

condensate water desalination unit at the appropriate level. The volume of resins replenished annually is 10% for

249

cation exchange resin and around 20% for anion exchange resin.

Table 3.3.2-14: List of Water Quality of Supercritical Pressure Through the Flow Boiler at Normal Times Specimens taken Analysis item Criteria

Makeup water Conductivity Silica concentration

< 0.5 µS/cm < 30 µg/l

Outlet of a condensate pump (CP out)

Conductivity

< 0.5 µS/cm

Outlet of a condensate water desalination unit (CBP out)

Conductivity Sodium concentration Total iron concentration Total copper concentration

< 0.15 µS/cm < 5 µg/l < 5 µg/l < 2 µg/l

Outlet of a desalination unit (Dea out)

Dissolved oxygen concentration

< 7 µg/l

Inlet of an economizer (Eco in)

pH Conductivity Total iron concentration Total copper concentration Hydrazine concentration Silica concentration

9.3 to 9.6 < 0.25 µS/cm < 5 µg/l < 2 µg/l < 10 µg/l < 20 µg/l

Table 3.3.2-15: Analysis item and Frequency Specimens taken Analysis item Frequency

Makeup water Conductivity Silica concentration

Once a month Once a month

Outlet of a condensate water desalination unit

Iron concentration Cooper concentration Sodium concentration

Once a month (once a week) Once a month (once a week)

Once a month (-) Inlet of a desalination unit

Conductivity Dissolved oxygen concentration

Once a month (once every three days) Once a month (once every two weeks)

Outlet of a desalination unit

Dissolved oxygen concentration Once a month (once every two weeks)

Inlet of an economizer pH Conductivity

Total iron concentration Total copper concentration Hydrazine concentration

Silica concentration

Once a month (once every three days) -

Once a month (once every three days) Once a month (once every three days) Once a month (once every two weeks) Once a month (once every three days)

Main steam pH Conductivity


- (once every two weeks) - (once every two weeks) - (once every two weeks)

Items in parentheses indicate the frequency of analysis within six months of the start of operation.

Table 3.3.1-16: List of Storage Methods when the Plant is Subject to Shutdown Feed water heater Stop time Boiler Deaeration unit Condenser

Steam side Feed water side Within 56 hours Hot banking In circulation mode Retains the ordinary

water level. Storage in vacuum condition or in steam sealing

Retain the shutdown status.

Within 72 hours Storage after filling water +N2 pressurization

(Hydrazine: 20 to 30 mg/l)

Normal water level + Steam sealing [or N2 pressurization]


Retains the ordinary water level.

N2 pressurization Storage by filling water


More than 72 hours

Dry storage after sealing N2

(RH: Dry storage)

Dry storage after sealing N2

Dry storage Dry storage after sealing N2 or dry storage

Storage by sealing N2 or by filling water

(Hydrazine: >300 mg/l)

(2) Water treatment at Startup and while the Unit is not used a. Water treatment while the Unit is not used

The most important thing in water treatment while the unit is not used is to minimize the inclusion of exterior air inside the system to prevent corrosion. The following three measures are conducted for this purpose:

(1) Hot banking that puts a boiler under a pressurization condition to eliminate the inclusion of exterior air (2) After a boiler has cooled down, a high concentration of hydrazine is infused to minimize the area

contacting with air, while also helping remove the dissolved oxygen from the contact area. (3) After a boiler comes to a stop, boiler water is purged and blown out by pressurized nitrogen while keeping

the boiler temperature at 100 or above to keep it in a dry condition.

250

Fig. 3.3.2-16 shows how to store the plant when it is not in use. The table categorized the storage method by the period of storage. In the case that the planned short time storage is subject to change in the longer storage period, the storage method for the plant must be changed. Sampling racks are stored by closing the valves and after filling with deionized water. b. Water treatment at Startup

Before starting the plant, it is subject to a cleanup process by dividing the system into 3 blocks of condenser, feed water system and boilers respectively. The controlled items for this purpose include, for a cold cleanup process, iron and mill scales, etc., with those that are generated during the time the plant is not in use analyzed. To check the iron concentration, two methods are used; the membrane filtration and automatic measurement methods. The former compares the color of the filter after filtration of sample water with the standard color, while for the latter, an iron meter of the particle counting method or the scattered light method is used. The control criteria of the iron concentration is set as 300 µg/l or below as a target, while Fig. 3.3.2-17 shows other water control criteria. In addition, the injection of ammonia and hydrazine, etc. is possible to combat corrosion of the system and maintain the quality of water in the systems at the appropriate level. In injecting chemicals, Mode PB on the sampling rack is selected. Fig. 3.3.2-18 shows a worksheet of a chemical feed unit.

In the Sodegaura Thermal Power Plant, a patterned operation is used for the blowing time and chemical injection in order to control iron concentration, etc. at the appropriate level. The pattern was formed based on the experience of the plant. More recently, more plants have been able to automatically control and operate water treatment and chemical injection using a computer. In the Sodegaura Thermal Power Plant, the iron concentration of feed water at the inlet of an economizer after ignition of the plant is controlled to 50µg/l or below as a target. The criteria for collecting drain water generated by each process is, 300µg/l or below for iron collected by a condenser, and 50µg/l or below for that collected by a feed water system.

251

Table 3.3.2-17: List of Water Quality of a Supercritical Pressure Through a Flow Boiler at its Startup Startup process Sampling at: Analyzing item Criteria

Surge of condenser vacuum - - Condenser Vacuum > 680 mmHg

Water feed to a condenser desalination unit

CP out Total iron concentration Hydrazine concentration

< 300 µg/l < 10 µg/l

Pre-boiler system blow stop CP out Total iron concentration Hydrazine concentration

< 300 µg/l < 10 µg/l

Water feed to boilers Dea out pH Total iron concentration

Dissolved oxygen concentration

9.2 to 9.6 < 100 µg/l < 50 µg/l

Boiler system blow stop WW out or SH out

Total iron concentration Hydrazine concentration

< 300 µg/l < 10 µg/l

Eco in pH Conductivity

Total iron concentration Total copper concentration

Dissolved oxygen concentration Silica concentration

9.2 to 9.6 < 1.0 µs/cm

< 50 µg/l < 10 µg/l < 10 µg/l < 30 µg/l

Ignition

WW out pH Conductivity

9.2 to 9.6 < 1.0 µg/l

Eco in pH Conductivity


Dissolved oxygen concentration Silica concentration

9.2 to 9.6 < 1.0 µs/cm

< 50 µg/l < 10 µg/l < 10 µg/l < 30 µg/l

Humidified circulation

WW out Conductivity < 1.0 µs/l Aeration to turbines - 1/2 loads Eco in pH



9.2 to 9.6 < 50 µg/l < 10 µg/l < 30 µg/l

Dra

in w

ater

co

llect

ion

Collection by condenser Collection by condensate water and feed water systems

- -

Total iron concentration Total iron concentration

< 300 µg/l < 50 µg/l

Conductivity: Cationic conductivity after going through cation resin. (3) Water treatment at the Seawater Leakage of Condenser a. Determination of a Leakage

In order to detect any leakage of seawater at an early stage, two methods are normally employed; a salinometer installed closer to a condenser, and checking the cationic conductivity of the condensate water by sampling it at the sampling rack (after it has passed through the cation exchange resin). However, the method of measuring the cationic conductivity tends to show a high rate, because of the condenser hot well water generated when a vacuum break occurs at the startup time and the influence of the carbonate ions generated by carbon dioxide dissolved in makeup water, the conditions of which are quite similar to that shown at seawater leakage. Accordingly, it was necessary to analyze the existence of chlorine ions. The shortcomings of this method are (i) the fact that the quantification limit is high, i.e. 0.01 mg/l, making it difficult to detect minute leakages and (ii) it takes about an hour for the analysis, meaning the method cannot be used for startup operation. Therefore, in this plant, a sodium meter (of ion electrode method) is installed at the outlet of a condensate pump to be used in combination with the salinometer for detecting seawater leakage.

Table 3.3.2-18: Chemical Injection Workflow Sheet

252

Fig. 3.3.2-19: Measures to be Taken upon Leakage of Condenser Pipes

Specimen

Conductivity (µS/cm) [Cationic

conductivity]

Chlorine Ion

(mg/l)

Determination of Leakage

How to Operate a Desalination Unit Measures to be Taken at Leakage

Outlet of a condensate pump

< 0.5 0.5 to 3.0

> 3.0

< 0.1 0.1 to 3.0

> 0.3

Normal Minute leakage

Significant leakage

H Type: 1 NH4 Type: 2 H Type: 2 NH4 Type: 1 H Type: 3

- Repair the leakage while operating only one condenser, after confirming the location of the leakage using a salinometer.

Stop the unit. Outlet of a condensate water desalination pump

> 0.15 > 0.1 Significant leakage

- Stop the unit

b. Operation of a Condensate water Desalination Unit

Table 3.3.2-19 shows the measures to be taken after identifying the occurrence of a seawater leakage to a condenser pipe and determining its severity. Immediately after detecting cationic conductivity exceeding 0.5µS/cm, a stand-by water feed tower of a desalination unit is switched on, one NH4 type water intake tower that is currently taking water is shut down under the system configuration of two H type water intake towers and one NH4 type water intake tower. When the leakage deteriorates further to exceed 3µS/cm, an examination starts to stop the unit while all towers are switched to H type water intake towers. By the time that cationic conductivity at the outlet of a desalination unit exceeds 0.15µS/cm, the desalination unit will have been in break condition (a status where the exchanging function is lost), the unit is stopped immediately.

When such abnormalities are detected, while the leakage is limited to 0.5 to 3µS/cm, the loads to the plant is decreased and one of the condensers is stopped to detect the location and seriousness of leakage and a repair process will start. The resins that are exposed to the seawater leakage contain salts in seawater and hence the ratio of sodium and chlorine contained in them are higher than in normal times. In such cases, the resins are subject to regeneration based on the degree to which such ions are absorbed.

Plant process Circulation of condensate water

Circulation of the deaeration unit

Circulation of the pre-boiler

Circulation of the boiler Normal operation Stop

Water treatment values Fe: 300 ppm or

below Hot banking N2H4: 10 ppm or

belowDO: 50 ppb

or below 10 ppb or above N2H4: 10 ppm or above

Normal storage and plant not in use:

Sampling rack mode

PB process

Circulation of

condensate water

Circulation of the

deaeration unit

Circulation of the

pre-boiler

Circulation of the boiler

Aeration of the main

steam pipe

Lamping end

High pressure heater

Low pressure heater

Normal operation

Delamping Parallel off Storage method

Hot banking

Ordinary storage

Stop

Kick signal

Injection of a high concentration of hydrazine

for 4 hours

Am

mon

ia p

ump

Stroke length control




RPM control

RPM control

Program control with the conductivity at the outlet of a demister as a preceding signal


Constant value control of conductivity at the inlet of a deaeration unit with the conductivity at the outlet of a demister as a preceding signal Injection of a

constant volume

In proportion to the condensate water flow rate


Constant value control of conductivity at the inlet of a deaeration unit with the conductivity at the outlet of a demister as a preceding signal Injection of a

constant volume

As above

As above




In proportion to the condensate water flow rate H

ydra

zine

pum

p

In proportion to the feed water flow rate In proportion to the condensate water flow

In proportion to the feed water flow rate In proportion to the condensate water flow

Injection of a constant volume (CONC)

Injection of a constant volume (CONC)

Che

mic

al in

ject

ion

loca

tions

Am

mon

ia

Open

Open

Hyd

razi

ne

Open

253

To th

e ou

tlet o

f H

PHTR

To th

e st

artu

p co

llect

ion

pit

Fig.

3.3

.2-8

: Wat

er T

reat

men

t Flo

w C

hart

of th

e N

o. 1

Uni

t of K

awag

oe T

herm

al P

ower

Pla

nt

To

the

star

tup

efflu

ent p

it

Dra

in ta

nk

LP.H

P H

TR D

rain

Stea

m A

an

d B

From

HP

1HTR

From

the

outle

t of

CB

P

From

HP2

HTR

To th

e bo

iler

blow

tank

To th

e bo

iler

blow

ta

n k

LP-H

TRD

rain

P

To th

e bo

iler b

low

tank

254

3.3.2.2.4 Water treatment of an Ultrasupercritical Pressure Once-through Boiler The method used by an ultrasupercritical pressure once-through boiler to control water quality is basically the

same as the conventional supercritical pressure once-through boiler. This chapter explains how water quality is controlled at the Kawagoe Thermal Power Plant. Fig. 3.3.2-8 shows the water treatment flow chart. (1) Water treatment in Normal Operation a. Frequency and Method of Chemical Injection

Aiming to keep the pH of the feed water at the inlet of an economizer at 9.6 as a target, ammonia is injected continuously and automatically to the outlet of a condensate water booster pump in proportion to the flow rate of the condensate water and with monitoring of conductivity (ammonia concentration) at the inlet of a deaeration unit.

In addition, in order to keep the hydrazine concentration of feed water at the inlet of the economizer at 20µg/l as a target, hydrazine is injected continuously and automatically to the outlet of a condensate water booster pump in proportion to the flow rate of the condensate water and with monitoring of hydrazine concentration of the feed water at the inlet of the economizer to retain it to 35µS/l. b. Operation Control of a Condensate water Desalination Unit

In this plant, the condensate water desalination unit is operated continuously without limiting the inflow of water into the unit. The unit consists of two prefilters (electromagnetic filtration system) and four mixed bed condensate demineralization towers (with one standby).

In principle, all condensate demineralization towers are of NH4 type and water collection stops when the towers are filled by a pre-designated volume of water and when water quality deteriorates. Table 3.3.2-20 shows the pre-designated volume of water and the water treatment values of this desalination unit. (2) Water treatment at Startup and while the Unit is not used a. Pattern of Stoppage and How to Store each Component

Table 3.3.2-21 shows the stoppage pattern and how each component is stored. b. Sampling Rack at the time of Stoppage and How to Store Chemical-Related Measurement

Instruments As soon as the systems of each sampling point stops, a shut-off valve installed at the inlet of the sampling rack

automatically closes. The chemical-related measurement instruments are in standby condition, ready to start measurement, just the same as when the system is in operation. No transfer of deionized water takes place in the sampling rack. c. Cleanup

Table 3.3.2-22 shows the scope of the cleanup and the water treatment criteria. d. Monitoring of Water Quality

Water quality is monitored by a continuous water quality measurement instrument. When each system component starts working, a valve installed at the inlet of the sampling rack automatically opens. As for the iron concentration, a scattered light/transmitted light ratio turbidity meter that can monitor colloidal iron is used by automatically switching the measurement point.

Fig. 3.3.2-20: Designated Water Volume of the Condensate water Desalination Unit and Water treatment criteria

Designated collection volume

NH4 Type H Type

384 000 (m3) 35 000 (m3)

Water treatment (outlet water quality)

Conductivity Sodium ion concentration

0.15 (µS/cm) 5.0 (µg/l)

e. Drain Collection

Drain is blown to the outside of the system at the same time with the startup of the feed water heater. Drain of the low-pressure feed water heater is collected to the condenser when the iron concentration becomes 500 µg/l or less. Moreover, when the iron concentration becomes 50 µg/l or less, it is collected to the condensate water pipe. Similarly, drain of the high-pressure feed water heater is collected to the condenser when the iron concentration becomes 50 µg/l or less. When the iron concentration becomes 50 µg/l or less, it is collected to the deaerator.

255

Fig. 3.3.2-21: Stoppage Patterns (Category) and How to Store each Component

Stop time Within 72 hours 72 hours to 1 week 1 week or longer Boiler stop condition

Normal stop Forced cooling Normal stop Forced cooling Normal stop Forced cooling Stoppage Category

Component Condenser vacuum

Retained Destructed Retained Destructed Retained Destructed Retained Destructed

Boiler body Hot bank When the pressure is decreased Pressurized sealing of N2

Same as left Pressurized sealing of N2

Same as left Hot bank After decreasing the pressure, the boiler is filled with 100mg/l of hydrazine for storage.

Storage after filling 100mg/l of hydrazine

Hot bank After decreasing the pressure, the boiler is filled with 200mg/l of hydrazine for storage.

After decreasing the pressure, the boiler is filled with 200mg/l of hydrazine for storage.

Pre-boiler Valve is closed like during normal operation.

Same as left Same as left Same as left Same as left Like the case of a boiler, the pre-boiler is filled with 100mg/l of hydrazine for storage.


Same as left Like the case of a boiler, the pre-boiler is filled with 200mg/l of hydrazine for storage.


Deaeration unit Hot bank When the pressure is decreased Pressurized sealing of steam or N2

Same as left Pressurized sealing of N2

Same as left Hot bank Like the case of a boiler, the pre-boiler is filled with 100mg/l of hydrazine for storage.


Hot bank Like the case of a boiler, the pre-boiler is filled with 200mg/l of hydrazine for storage.


Low pressure feed water heater Valve is closed like during normal operation.





From condenser to the inlet of the low pressure feed water heater

Cleanup Circulation continued





Shell side of feed water heater Retained under vacuum condition.

Pressurized sealing of N2

Retained under vacuum condition.

Pressurized sealing of N2

Same as left Same as left Same as left Same as left

Superheater and reheater Valve is closed like during normal operation.

Same as left Same as left Same as left Same as left Same as left Same as left Same as left

256

Table 3.3.2-22: Scope of Cleanup and Water Quality Criteria

Water quality criteria for approving the cleanup process Cleanup process Water quality measurement

point pH Cationic

conductivity (µS/cm)

Dissolved oxygen

concentration (mg/l)

Hydrazine concentration

(mg/l)

Total iron concentration

(mg/l)

Scope of cleanup

Remarks

Blowing Outlet of a condensate pump

9.8 1.50 or below

0.500 Condensate water

Circulation Outlet of a condensate pump

9.4 to 9.7

0.15 or below

0.200 0.050 (Outlet of a

condensate water booster pump)

From a condenser to the inlet of a low pressure feed water heater

Blowing Outlet of a deaeration circulation pump

9.5 to 9.7

0.50 or below

0.500 Deaeration unit

Circulation Outlet of a deaeration circulation pump

9.5 to 9.7

0.50 or below

0.10 or below

0.200 0.050

From a low pressure feed water heater to a deaeration unit

Blowing Pre-boiler cleanup pipe

9.5 to 9.7

0.50 or below

0.500 Pre-boiler

Circulation Pre-boiler cleanup pipe

9.5 to 0.7

0.50 or below

0.10 or below

0.200 0.050

From the outlet of a deaeration unit to a high pressure feed water heater

Blowing Outlet of a water separator

9.5 to 9.7

0.50 or below

0.500 Boiler

Circulation Outlet of a water separator

9.5 to 9.7

0.50 or below

0.10 or below

0.200 0.050

From the outlet of a high pressure feed water heater to a boiler

3.3.2.3 Water treatment When Condenser Pipes are Subject to Leakage

Most thermal power plant use seawater as the cooling water for their condenser and this is frequently channeled into the system through condenser pipes or joints between them and the pipe boards due to their corrosion and erosion (seawater leakage). The seawater, having infiltrated into the system, causes the following damage to each part of the unit:

(1) Corrosion of materials due to the decreased pH of the boiler water caused by magnesium chloride (2) Heat transmission failure due to the scales consisting of Ca and Mg on the generation pipes (3) Contamination and corrosion of superheater pipes and turbine blades due to the carryover of salts (4) Corrosion of SUS materials due to the chlorine ions contained in the temperature reduction spray water of a

superheater. As measures to avoid seawater leakage, several methods are possible, including the prevention of foreign

matter brought into a condenser, the installation of an electrolytic protection unit on the seawater side of a condenser, a coating of anti-ion and -oxide layers inside condenser pipes and the use of titanium pipes. As for protective maintenance measures against system aging, an eddy current flaw detector (eddiography) is used during regular inspections. 3.3.2.3.1 Confirmation of Leakage

A seawater leakage can be detected by the enhanced conductivity of condensate water. In the case of a drum boiler, it can be identified by increased chloride salt concentration, although in the case of a minute leakage, it is difficult to identify the leakage with such measures. Generally, the leakage is detected by transforming salts into acids via cation exchange resins. When salts are transformed into acids, the conductivity jumps up to 8 to 10 times. For this purpose, a cation exchange resin tower is installed before an conductivity meter of condensate water, basically. In order to detect seawater leakages as early as possible, most units take out condenser hot well water to measure its salt concentration using a salinometer, in combination with the conductivity meter, so that seawater leakage can be detected, even when the conductivity meter fails. The double monitoring system is relatively useful in detecting seawater leakage.

Fig. 3.3.2-9 shows an example of the salinometer installation. The conductivity meter shall be monitored constantly by installing the alarm system and the recorder because

the phenomenon of seawater leakage may be continuous but it can also be temporary, in which case it only lasts for a few minutes to over ten minutes.

3.3.2.3.2 How to Detect the Leakage Location in One Condenser Operation

It is desirable that the unit be shut down and damaged parts identified and repaired as soon as seawater leakage occurs. However, most cases involve the need for identification and repair of damaged parts in parallel with the unit in operation. As a unit installed with a salinometer and condenser hot wells can be selectable, in which a condenser system leakage occurs, condenser A or B can easily be identified by an operator by referring to its operation manual. There may be several ways to detect a leakage. A 1000MW class boiler has a large-sized condenser with a number of thin pipes installed on it. The first thing to do is to identify the location of the leakage. In this class, firstly, a circulation pump is stopped, and then a transparent vinyl tube is connected to the pump pressure detector so that it rises vertically against the top of the condenser. While checking the water level in the transparent tube, seawater is blown gradually to estimate the leakage part by referring to the indications of a salinometer and a sodium meter. After the total blowing, an operator enters into the seawater side system and pastes a thin polyethylene film to the surface of the pipes. The location of a pipe where the film is sucked has a hole as a source of the leakage.

In the case of a minute leakage, it is difficult to identify it via single condenser operation. In many case a leakage stops unknowingly to the operators. In this case, what is generally practiced is conducting a water pressure test after stopping the unit. Makeup water of a low concentration fluorescent solution is put into the condenser steam side and left for a period ranging from several hours to about one day. Subsequently, ultraviolet rays are cast from the seawater side to detect which part reacts against the rays. 3.3.2.3.3 Water treatment at Seawater Leakage

As explained above, seawater slipped into feed water due to a leakage can cause various damage to boiler and turbine systems, so it is important to stop the leakage and remove salts from the feed water.

Condensate water pump

Ret

urn

valv

e

Sampling valve

Sampling valve

Conductivity meter

Res

in to

wer

Res

in to

wer

Sam

plin

g

Sam

plin

g

Flow

met

er

Fig. 3.3.2-9: Example of Salinometer Installation

(1) Drum Type Boiler

Among the salts contained in seawater, MgCl2 can produce Mg(OH)2 sediments and HCl as shown in the following chemical formula, causing a significant deterioration in the pH level of boiler water and accelerating corrosion of the materials.

MgCl2+2H2O→Mg(OH)2+2HCl As soon as seawater leakage occurs, boiler blowing on a par with the volume of leakage starts and in order to

uplift the pH value slightly higher than the criteria of water quality, an appropriate volume of sodium phosphate is injected. The phosphate ions have the effect of preventing the deposition of hard scales of Ca and Mg, and also work to discharge such scales out of the system via the help of the boiler blowing. In the case of a unit where feed water is sprayed to a superheater and reheater to reduce their temperature, the volume of the feed water is reduced when seawater leakage is detected. When the chlorine ion concentration in condensate water rises to 0.5mg/l or above, the spraying is stopped. Table 3.3.2-23 shows the actions taken to a 175MW drum type boiler. (2) Once-through Boiler

When a condensate water desalination unit reaches its peak capacity, a once-through boiler stops its operation.

257

258

For this reason, when seawater leakage is detected, its remaining capacity is checked immediately. During normal operation, an ammonia type sampling is switched to H type sampling, on a par with the volume of leakage, and the standby tower starts operation. With these measures, loads are reduced as soon as possible to contain the volume of seawater slipping into the system. Table 3.3.2-14 shows the actions taken for a 1,000MW class boiler. 3.3.2.4 Water treatment During a Regular Inspection

A large-sized industrial use boiler tends to have a longer interval till the next regular inspection and its restoration will take also longer, from the process of filling water to restarting it. During that time, the water quality must be maintained in good condition. When such boiler is subject to a stop for regular inspection, a boiler is subject to blowing at as high a temperature as possible to remove any residual liquid in it, so that it can be stored in dry condition.

Generally speaking, 60% to 70% of the total process is subject to water pressure tests. Pre-restoration, each component is kept in such condition that pressure-related tests can take place. Firstly, a deaeration unit is flushed with water and then filled with water drawn directly from a makeup water tank, to which approx. 100mg/l of hydrazine is added. Subsequently, a commissioning test is conducted to a feed water pump. The next process involves filling the boiler with water and adding 100mg/l hydrazine. After the water pressure test, nitrogen is blown into the boiler to purge the water. The water containing hydrazine purged out from the boiler is then decomposed by adding sodium hypochlorite. The concentration of this chemical is also kept to a minimum for environmental purposes.

As for the storage of the unit till the next startup, the general practice is that the water used for the pressure test is blown out. If the unit is subject to an immediate start, boiler pipes are sometimes filled with water for storage.

Table 3.3.2-23: Example of Actions when Seawater Leakage Happens to a Condenser (Example) Surge of Conductivity

at the Outlet of CP (µS/cm)

Actions Taken

Water Quality Treatment 1/2 load Rated

load Actions for Operation Blowing Chemical Injection Others Cl-Concentration

of Condensate water

Lower than 6

Lower than3

1. To strengthen monitoring by monitoring instruments

2. Determination of the location and degree of leakage

3. Examination for starting operation and inspection plans

1. Continuous blowing a. To be done

immediately after detecting seawater leakage

b. Intermittent blowing depending on the leakage conditions (manual inspection, etc.)

2. Blowing of condensate water a. To be done

depending on the leakage conditions

1. To close the condensate water return valve connected to a distilled water tank

2. To conduct a water-pressurized leak check

-

6 or above 3 or above

1. One condenser operation (To stop damaged condenser and to operate the intact condenser continuously at the 1/2 load)

2. To confirm that the conductivity at the outlet of CP, etc., has decreased

0.3 ppm or above

10 or above

5 or above

1. Reduction or stop of the spray flow rate for a superheater and reheater

2. To reduce load

1. Continuous blowing To be done continuously

2. Rapid blowing To be done in the case that water quality check revealed it necessary (Openness: 10% to 15%)

3. Blowing of condensate water To be done continuously (Fully open in principle, but subject to adjustment depending on conditions)

1. To check and repair the damaged condenser while operating the intact one (Inspection using a vinyl sheet)

0.5 ppm or above

20 10

1. To stop the unit in principle

1. Total boiler blowing as necessary

1. Chemical injection to boilers a. To inject 10l of

sodium tertiary phosphate to the drum when continuous blowing starts

b. To inject an appropriate volume of sodium tertiary phosphate so that the pH level of boiler water can be maintained at around 9.5

2. Chemical injection to feed water a. To switch

hydrazine injection from automatic mode to manual mode (Target: pH of feed water to be around 8.8)

* As seawater elements causes an uplift of conductivity, it is not possible to control the pH level by adjusting it.

1. To check and repair the damaged condenser

1.0 ppm

3.3.2.5 Water treatment of Component Cooling Water

The component cooling water system (bearing cooling water system) can be roughly divided into circulation and non-circulation temporary cooling types. The former can further be divided into a open circulation type, in

259

which water heated up in a heat exchanging process is evaporated in a cooling tower to be cooled by discharging evaporative latent heat for use in recycling, and a closed circulation type, in which heated water is cooled down in a cooling water cooler using seawater. A large capacity power generation plant in Japan uses the closed circulation type, which can then be categorized into systems where bearing cooling water tanks are installed and those using stand pipes. Recently, the latter has been frequently used because of the ease of water treatment.

These component cooling water systems incorporate an oil cooler that cools down the lubricants used for turbine rotors, etc., a hydrogen cooler that is used for cooling generator and a coolant cooler. These units are made of aluminum brass. Controlling the water quality of such component coolants should involve consideration of the selection of an appropriate coolant circulation method and the use of steel, copper and copper alloys. Based on such views, the water treatment of component coolants involves the introduction of anti-corrosion agents into the coolants, in order to prevent the corrosion of the heat exchanger cooling pipes as well as other pipes, in turn, to prevent scale deposits on the heat exchanger and avoid deterioration of its heat exchanging function. 3.3.2.5.1 Temporary Cooling Type Cooling System

A temporary cooling type is used where river water is accessible. Cooling water containing heat is normally discharged into the river untreated, because the use of high concentration anti-corrosion agents is not practical for cost reasons. However, polymer phosphate and silicate anti-corrosion agents of 2 to 5 mg/l are sometimes used, mainly to prevent the generation of carbon steel rust and corrosion of the peripheral area arising from the same. In using these chemicals, it is necessary to secure a flow rate of at least 1 m/s to obtain favorable results.

Table 3.3.2-24: Example of Water Quality Treatments to a High Hardness Cooling Water System

Operation Condition of Cooling Tower

Volume of circulation water: 20,000 m3/h Water volume retained: 13,000 m3

Temperature difference in a cooling tower: 8°C Concentration : 2.5 times

Chemicals Used

Initial injection: Anti-corrosion agent: Kurizetto S370 (polymer phosphate series) 400mg/l Anti-scaling agent: Kurizetto T225 (polymer series) 200mg/l

Normal operation: Alkali treatment agent: Kurizetto S113 (phosphate - polymer series) 40mg/l Chlorine treatment: 0.5 to 1.0mg/l(Cl2) 3 h/day Slime control agent: Polyclin A496 (nitrogen compounds - polymer series) 50mg/l month

Partial Filtration of Circulation Water

Sand filtration: 3% (against circulation water volume)

Water Quality

Makeup water Circulation water Turbidity (degree) 2 5 pH (at 25°C) 8.1 9.0 Conductivity (µS/cm) 350 1 000 Calcium hardness (CaCO3 mg/l) 170 380 M alkali level (CaCO3 mg/l) 180 400 Chloride ion (Cl-mg/l) 20 63 Sulfate ion (SO4

2-mg/l) 31 79 Silica (SiO2 mg/l) 7 18

Treatment Periods 7 years Results of Effects using a Test Piece Corrosion rate (SPCC) [mdd] 3 to 4

Result of a Regular Inspection A small volume of scales and sludge was observed in several low speed heat exchangers, but other heat exchangers were in good condition without any corrosion scales, slime and sludge damage.

3.3.2.5.2 Open Circulation Type Cooling System

As the cooling water in an open circulation type cooling water system partially evaporates in a cooling water tower, the dissolved salts are concentrated in the circulation water. In order to obtain favorable functions of a coolant, it is necessary that water volume of a forced blowing be adjusted to control the concentration of salts in the circulation water, and that the quality of circulation water, as well as the concentration of chemicals, such as anti-corrosion and anti-scaling agents, be kept at a constant level.

In an environment with high calcium hardness i.e. 150mg/l(CaCO3) or above, the alkali level is sufficient and the pH level is high, i.e. 8 or above in cooling water, calcium phosphate anti-corrosion coating can easily be formed and the concentration of anti-corrosion agents in water can be retained to 5 to 6 mg/l(T-PO4) to perform their intended result. However in such an environment, where the calcium hardness, M alkali level and pH are all high, it is necessary to add polymer series anti-corrosion agent of a sufficient concentration to avoid calcium phosphate series anti-corrosion coating and calcium carbonate from forming scales in the high temperature area. As the quality of makeup water shows a relatively low hardness and low M alkali level, in most cases, this type of treatment is applied by high concentration operation of water cooling system (5 times or above). This works well in view of preventing environmental pollution because it can reduce the volume of blowing water and phosphorous discharged outside the system.

260

Unlike highly hard water, water with low calcium hardness tends to require an increased concentration of anti-corrosion chemicals and it is necessary to raise the concentration of anti-corrosion agent to 10 to 15 mg/l(T-PO4) under the calcium hardness of around 100 mg/l(CaCO3) and 15 to 20 mg/l(T-PO4) in the case of 50 mg/l (CaCO2). This is because the phosphate series anti-corrosion agent is influenced by dianoinic metals, such as calcium ions, and because the combined use of phosphate and zinc salts with a strong coating forming performance can achieve good anti-corrosion performance, even if the concentration of anti-corrosion chemicals is kept to a low level. In a low hardness cooling water system, scales such as calcium phosphate can be formed in a high temperature zone, and anti-scaling agents, such as acrylic acid series polymers and maleic acid series polymers, are generally used in combination with these chemicals. In an open circulation type cooling system, operation under a high concentration of chemicals to save the volume of water can thicken nutrients contained in the water and within such an environment, microbes can pullulate and slime be formed relatively easily. In order to prevent this, measures are taken by sterilizing the microbes and adding anti-slime agents that are effective in curtailing the reproduction of the same. For such purposes, the hypochlorites and cyanurates previously used have been recently replaced by carbonyl series compounds with no corrosion effects. Tables 3.3.2-24 and 3.3.2-25 show examples of water quality treatment at open circulation type cooling systems.

Table 3.3.2-26 shows the water quality analysis items and frequency of analysis that are usually used for the operating control of an open circulation type cooling system, while Table 3.3.2-27 shows the significance of these water quality analysis items.

Table 3.3.2-25: Example of Water Quality Treatments to a Low Hardness Cooling Water System

Operation Condition of Cooling Tower

Volume of circulation water: 5 000 m3/h Water volume retained: 2 400 m3

Temperature difference in a cooling tower: 12°C Concentration : 3 times

Chemicals Used

Initial injection: Anti-corrosion agent: Kurizetto S370 (polymer phosphate series) 400mg/l Kurizetto S611 (zinc salt series) 100mg/l

Normal operation: Anti-corrosion agent: Kurizetto S603 (phosphonates, phosphates and zinc salts) 50mg/l Anti-scaling agent: Kurizetto T225 (polymer series) 30 mg/l Chlorine treatment: Polyclin A411 0.3 to 1.0mg/l(Cl2) 3 h/day

Partial Filtration of Circulation Water

Sand filtration: 3% (against circulation water volume)

Water Quality

Makeup water Circulation water Turbidity (degree) 2 7 pH (at 25°C) 7.2 7.9 Conductivity (µS/cm) 100 254 Calcium hardness (CaCO3 mg/l) 24 72 M alkali level (CaCO3 mg/l) 23 47 Chloride ion (Cl-mg/l) 5 18 Sulfate ion (SO4

2-mg/l) 6 15 Silica (SiO2 mg/l) 6 15

Treatment Periods 8 years Results of Effects using a Test Piece Corrosion rate (SPCC) [mdd] 3 to 5

Result of a Regular Inspection Corrosion, scale and slime were hardly observed and the result was very good.

Table 3.3.2-26: Water treatment Items and Analyzing Frequency for Operation Control of an Open Circulation Type Cooling System (Standard)

Frequency of Analysis Analysis Item Makeup Water Circulation Water

Turbidity (degree) Once a week Once a week pH (at 25°C) Once a week Once a day Conductivity (µS/cm) Once a week Once a day M alkali level (CaCO3 mg/l) Once a week Once a week Calcium hardness (CaCO3 mg/l) Once a week Once a week Chloride ion (Cl-mg/l) Once a week Once a week Sulfate ion (SO4

2-mg/l) Once a week - Silica (SiO2 mg/l) Once a week - Total iron (Fe mg/l) Once a week - Residual chlorine (Cl2 mg/l) - - CODMin (O mg/l) Once a month Once a month Anti-corrosion agents (mg/l) - Once a day

There are several ways to monitor the effectiveness of an anti-corrosion agent, the representative methods of

which include (1) measuring the corrosion speed of a test piece, (2) measuring the corrosion speed using a corrosion measurement device using a polarization resistance method (an electrochemical method), (3) confirming the conditions of corrosion and erosion depth on the heat transfer surface of a pipe using a heat exchanger for

261

monitoring purposes. The use of a test piece or a corrosion measurement device to measure the corrosion speed cannot confirm the corrosion conditions on the heat transfer surface, in which case a heat exchanger for monitoring purposes can be used. However, this method is not generally used, because the conditions of pipes cannot be confirmed during test periods (normally 1 to 3 months) and because it requires significant plant and operation control costs.

Table 3.3.2-27: Significance of Each Water Quality Analysis Item Item Significance

pH (at 25°C)

Measured to obtain the trend of corrosion behavior and scale formation of water. The pH level of circulation water is normally controlled at 7.0 to 9.0. Where the level is decreased to 6.5 or below, it must be raised by the addition of alkali agents. In this case, the appropriate pH level is 8.0 to 9.0.

Conductivity (µS/cm)

Measured to determine the trend of salt concentration dissociated into water as ions. Generally speaking, water quality with high conductivity tends to be bad and is frequently a cause of corrosion damage.

Turbidity (degree)

Measured to determine the volume of suspended matter in water. Since the presence of such suspended matter in the system can cause deterioration of efficiency and erosion damage to a heat exchanger, the turbidity of the circulation water should be retained as low as possible.

M alkali level (CaCO3 mg/l)

There is a certain degree of connection between pH and the degree of alkali. The M alkali degree is an indicator of the trend of calcium carbonate forming scales.

Calcium hardness (CaCO3 mg/l)

This indicator is important to control the concentration of circulation water and to determine the trend for the formation of scales by calcium and other compounds such as calcium carbonate.

Chloride ion (Cl-mg/l)

This is generally used as an indicator for controlling the concentration of circulation water. For a system where chlorine treatment is performed, this indicator is used in combination with others, such as conductivity, calcium hardness and silica concentration, etc. Water containing high chloride ions tends to have strong corrosive performances.

Sulfate ion (SO4

2-mg/l) Water containing a high concentration of sulfate ions tends to have a strongly corrosive performance. As for HAVC coolants, the inclusion of sulfur acid gas contained in the air into the system causes a high concentration of sulfate ions and decreases the pH level, forming a highly corrosive environment.

Silica (SiO2 mg/l)

Silica is one of the causes of scale formation.

Ammonium ion (NH4

+/l) Water containing a high concentration of ammonium ions is highly inclined to generate slime. For a system using copper series materials, ammonium ions and copper react with each other to form a complex ammonium ion salt, which is a cause of corrosion.

Consumption of oxygen [CODMin ]

(O mg/l)

A system with a high consumption of oxygen tends to cause slime, meaning appropriate slime control measures must be implemented.

General microbe count (pcs/ml)

It can be an indicator to know the generation of slime. It can also be used to judge the effectiveness of the microbicide.

Total iron (Fe mg/l)

The total iron content in the circulation water includes iron ion and colloidal ion derived from the makeup water as well as other iron generated by corrosion of the system. The existence of iron can cause secondary corrosion, meaning the total iron concentration must be kept as low as possible.

Concentration of anti-corrosion agents and anti-scaling agents

(mg/l)

It is necessary to constantly maintain the concentrations of anti-corrosion and anti-scaling agents at an appropriate level. In the case of significant fluctuations, effective anti-corrosion and anti-scaling performances cannot be expected.

3.3.2.5.3 Closed Circulation Type Cooling System

A closed circulation type cooling system does not incorporate any cooling towers, and thus thickening of chemicals due to the evaporation of water cannot take place. The main problem affecting this type of cooling system is corrosion. In order to prevent corrosion, polymer phosphate series agents, nitrate series agents and molybdenum salt series polymer agents are used, as shown in Table 3.3.2-28.

As cooling systems with bearing coolant tanks have a large area, where the coolants come into contact with the atmosphere, and also generally tend to have significant cooling water leakage, substances that cause the pH level to fluctuate, such as carbon dioxide, can enter into the system quite easily from the air, and the volume of anti-corrosion agents consumed tends to be high. For such reasons, filtrated water is used as a coolant for cost saving purposes, and a polymer phosphate series anti-corrosion agent is used. The polymer phosphate reacts against copper to form dark green Cu2O CuPO3. Similarly to basic copper chloride, this compound does not form a continuous layer, has high solubility and can have significantly adverse effects to the system. Due to this, it is necessary to use benzyxotriazole and a derivative of trithio cyanuric acid in combination with the above agents, because these substances show a strong anti-corrosion performance against copper and copper alloys. The concentration of the polymer phosphate anti-corrosion agents is 100 to 200 ppm for initial injection. After the initial injection, the concentration is kept at 20 to 40 ppm. The concentration of anti-corrosion agents is controlled in such a manner that after understanding the relationship of the agent and conductivity, the agent is added by monitoring the conductivity of the coolant.

A coolant system incorporated with a stand pipe, where the coolant is circulated without being exposed to the air, meaning no carbon dioxide and oxygen will enter the system from the exterior air. The leakage of coolant is small and consumption of anti-corrosion agents to form anti-corrosion layers can be minimized. For these reasons, this type of system is superior to other types having a coolant tank, in terms of controlling water quality. Conventionally, filtrated water was used as a coolant. However, recently, deionized water has been used for this purpose due to the enhanced plant reliability. Nitrate salt is used to prevent corrosions, but sometimes hydrazine is

262

also used in a one-off manner for this purpose. As nitrate salt anti-corrosion agents cannot expect anti-corrosion performance equivalent to that of polymer phosphate series agents, a special agent to prevent corrosion to copper must be used in combination. In addition, as nitrite salts can cause decreased concentrations, due to the act of microbes (nitrification due to action of oxidative bacterium against nitrate salts such as nitrobactors), it is necessary to use an inhibitor in combination with the agents or to use deionized water containing less uncertain elements. In the case of the nitrate salt series agent, after the initial injection of 200 to 300ppm of this agent, the concentration is kept at a level of 60 to 130 ppm. In the case of hydrazine, the initial concentration is 20 ppm, following which the concentration is maintained at 5 to 10 ppm when the concentration reaches a stable stage, whereupon a good anti-corrosion performance can be obtained. The frequency of administering anti-corrosion agents ranges from once a week to once every two weeks to obtain good water treatment.

Table 3.3.2-28: Outline of Anti-Corrosion Agents for a Closed Circulation Type Cooling System Anti-Corrosion

Agent Applicable Water System Corrosion Rate Remarks

Polymer Phosphate Series

Retention period: Within 10 days

Makeup water: Industrial water Fresh water

Carbon steel: 5 to 20 mdd Copper and copper alloys:

1 mdd or below

As this agent facilitates the discharge of corrosion products outside the system, a corrosion speed of 5 to 20 mdd may be acceptable. Use polymer agents in combination with it to reduce the generation of scales on heat transmission surfaces.

Nitrate Salt Series Retention period: 10 days or more

Makeup water: Industrial water Softened water Deionized water

Carbon steel: 1 mdd Copper and copper alloys:

1 mdd or below

As this agent does not allow the easy discharge of corrosion products outside of the system, a chemical that ensure favorable anti-corrosion performance must be used. As the maximum temperature of bearing coolants is 40°C or below, which is within the optimum growing temperature for microbes oxidizing nitrate salts (generally 15 to 30°C), it is necessary to use its inhibitor in combination with it.

Molybdate Polymer Series

Retention period: 50 days or more

Makeup water: Industrial water Softened water Deionized water

Carbon steel: 10 mdd Copper and copper alloys:

1 mdd or below

When this agent is used in a system in which no or insufficient anti-corrosion treatment is performed, iron oxides existing in the system can be washed away turning the color of the coolant red. In such systems, it is necessary to flush it before starting injection of this chemical.

3.3.3 Future Prospects With unit size subject to rapid growth since the introduction of large-sized components after WWII, a daily

water treatment method has almost been established following the era of trials and errors. However, many problems still remain to be solved depending on the location of a thermal power plant, in order to further advance the method of water treatment.

The first involves how to raise the reliability of supplying sufficient energy. Considering recent advancements of society and economic growth, it cannot be said that sufficient energies are guaranteed. Under current conditions, where it is increasingly difficult to build a new power plant, currently available systems should be used as long as possible. The issues to be solved in order to ensure the patterns of water treatments, established based on the experiences of thermal power plants to advance into stricter and more assured means of water treatment, include raising the accuracy of the devices used to monitor water quality as well as measures against seawater corrosion of condenser pipes, in terms of materials.

The second issues to be solved include (1) a reduction of carbon dioxide emissions that is considered to be a cause of global warming, (2) shortening the system operation time to maintain and uplift the heat efficiency of boilers from the viewpoint of energy saving and the appropriate use of energies, (3) the introduction of such methods as oxygen treatment to prevent scale attachment to boilers and (4) reduction in the ammonia volume used for the removal of nitrogen and phosphorus from effluents.

The third issues include plant maintenance. Water treatment is a decisive factor in preventing the corrosion of components. With this in mind, it is important to occasionally conduct a water quality inspection to assess the quality of water in the system and to determine the volume of oxygen and metal ion concentration, in order to conduct appropriate anti-corrosion measures.

3.4 Turbines and Auxiliary Machines 3.4.1. Maintenance of Steam Turbines

Maintenance of steam turbines includes minor repairs (for example, retightening of a gland packing for a valve) during a patrol and major repairs during periodic maintenance, among which replacement or improvement of a faulty part may be included. There are two kinds of maintenance, one is daily repair to be effected whilst the turbine remains in operation or the electric generation is stopped and another is periodic maintenance where the electric generation is periodically stopped for a long time. 3.4.1.1 Daily Repairs

Daily repairs consist of (1) repairs and adjustments of faulty parts as regular repairs, (2) regular inspections, adjustment and investigation of equipment such as important instruments and of vibration of the rotating devices in advance as preventative maintenance, (3) improvement of a part to prevent recurrence of a failure that has often occurred. [1] Speed, [2] sureness, [3] safety and [4] low cost are taken into consideration when a repair operation is effected, and an elaborately prepared plan in terms of processes and application procedures should be prepared especially for a part that may have a critical effect on the operation of the unit and also when the repair is carried out with operation of the power unit stopped. 3.4.1.2 Periodic Inspections

Voluntary inspection of a steam turbine used for electric generation should be periodically performed in compliance with the Electricity Utilities Industry Law.

This inspection is called the “Voluntary Periodic Inspection”, and overhauling of a steam turbine is effected with operation of the power unit subject to a long time planned stop. This inspection became voluntary in 1995 and the regulation stipulates that the inspection should be effected every four years from 1999. However, this four year period may be extended depending on the operating conditions. 3.4.1.3 Content of Periodic Inspections

During this periodic inspection, major repairs and improvements that cannot be effected at other times are to be carried out besides overhauling of the main body and the major accessories in accordance with the established plan. Table 3.4.1-1 shows examples of the maintenance of the parts and the equipment.

Table 3.4.1-1: Maintenance of a Steam Turbine During Periodic Maintenance 1. Main body of a turbine

(1) Turbine wheel Cleaning by honing Detailed and precision inspection and

repair of the disk, the rotating blades and the shaft

Measurement of run-out and centering of the shaft

Inspection and repair of the coupling bolts (2) Ejection Holes and Partitions Cleaning by honing Detailed and precision inspection and

repair of the stationary blades and labyrinths

(3) Casing Measurement of the Cleaning of the

inside and the outside of casing, detailed and precision inspection

Measurement of the level of the horizontal flange

Measurement of the alignment of the casing

Maintenance of the bolts, hardness test (4) Bearings Adjustment of the contact of the white

metals Measurement of the bearing gaps

2. Equipment attached to the turbine body (1) Main Valves (MSV, CV, RSV, ICV, SMV) Maintenance and precision inspection

of the inside and the outside of the valves, the valve rods, the valve seats, and the valve casings

Measurement of bend and the gaps of the valve rods

Inspection of the bolts exposed to high temperature

(2) Speed Governor and Emergency Stopping Device

Inspection of the speed governor mechanism and the piping for the control oil

(3) Turning Device Detailed and precision inspection of the

gears and the bearings Inspection of the clutch mechanism

3. The turbine lubricating oil device (1) MOP, BP, AOP, TGOP and EOP Overhaul, repair and detailed and

precision inspection (2) Main Oil Tank and Oil Cooler Cleaning and oiliness test of the inside

of the tank Cleaning of the oil cooler piping and the

water chamber (3) Oil Cleaner Cleaning of the inside and replacement

of the filter Overhaul and repair of the attached

pump and the fan

263

264

3.4.1.4 Special Maintenance Accumulated operating hours of many of the then new and advanced thermal power plants in Japan that were

the motive power for the rapid development of the Japanese economy are reaching one hundred thousand hours or more. It is time for them to be thoroughly inspected in a systematic manner.

Desirable items to be inspected are listed below. (1) Rotors of high, medium and low pressure turbines a. Center hole......................... Visual inspection by means of a bore scope, magnetic particle test, liquid penetrant

test and ultrasonic test b. Surface of the rotors .......... Hardness test at the designated points, structural examination by means of a

microscope, liquid penetrant test of general surface, magnetic particle test, hardness test

(2) Blades a. Embedded portion.............. Inspection to check whether the roots of the first and second stage rotors of the high

and the medium pressure turbines that are exposed to high temperature have lifted Ultrasonic test of the rotating blades in each of the high, and the medium pressure stages

b. Shroud tenon...................... General inspection to check whether it lifted and how it lifted (3) Main steam check valve... Liquid penetrant test of the inside and outside surface, magnetic particle test,

ultrasonic test, hardness test and structural examination by means of a microscope (4) Turbine casing.................. Penetrant test of the inside and outside surface, magnetic particle test, structural

examination by means of a microscope 3.4.1.5 Content of the Periodic Inspection

Content of the periodic inspection for a steam turbine is summarized as follows.

Table 3.4.1-2: Major Items of the Periodic Inspection Operation : Inspection

Legend V⋅T: Visual Inspection P⋅T: Penetrant test M⋅T: Magnetic particle test

U⋅T: Ultrasonic Test R⋅T: Radiographic test

No. Place to be Inspected Inspection Method Number of Tested Samples

Inspection Frequency Remarks Illustration and Reference Point

1 Usual inspection (1) Casing a. Corners and the inside

surface of each pipe seat

V⋅T, M⋅T 100% inspection Every two years In our experience, cracks are liable to occur due to concentration of heat stress in the corners, the inside of a pipe seat and places where thickness of the material markedly changes.

b. Nozzle chamber (a) The inner and the

outer surfaces V⋅T, M⋅T, A mirror for inspecting the inside of a pipe

100% inspection Every four years (1) Removing the nozzle plate, carefully check the base part of the nozzle vane, the corners of the nozzle chamber, the welded part, etc. (2) Carefully check the shape of the internal threads of the nozzle plate fixing bolts.

(b) The weldedconnecting part

M⋅T, U⋅T 100% inspection Every four years It is a matter of concern that self-excited vibration of the nozzle chamber occurs with increased clearance between the casing and the nozzle chamber causing too much stress in the welded base part resulting in a crack there.

(c) Profile Dimensions(Deformed amount)

100% inspection Every four years It is necessary to monitor the total deformed amount of the nozzle chamber since it is one of the parts of a steam turbine that is subject to the severest conditions and creep deformation may occur after a long time operation.

(d) Fitted part Dimensions(Clearance)

100% inspection Every four years It is necessary to monitor the clearance between the casing and the nozzle chamber every year since it may increase due to repeated knocks caused by vibration generated by vapor flow.

(e) Vane Dimensions(Eroded amount)

100% inspection Every four years Increased eroded amount causes unfavorable influences such as decreased turbine efficiency, weakened strength of the rotating blades, etc. It is necessary to carefully monitor a turbine unit with which starts and stops are frequently repeated especially because it may suffer from outstanding increase in the eroded amount.

The inside of a pipe seat

A corner

A welded connecting part M⋅T, U⋅T

The inside of the nozzle chamber V⋅T, M⋅T, A mirror to inspect the inside of a pipe Deformed amount of the profile

Measurement of the dimensions a, b and c

Eroded amount

Measuring point

265

Operation : Inspection



c. The spot-faced part of the horizontal joint

V⋅T, M⋅T 100% inspection Every two years The corners and the spot- faced part are liable to cracking.

d. The inlet sleeves V⋅T, M⋅T, Dimensions

100% inspection Every two years Carefully observe tearing-off, scoring and cracks on the border of an area to which stelite is coated. The maintenance of dimensions of bearing seal ring and seal ring is also important.

e. Hole Plug for the balance hole

M⋅T, Dimensions, Shape of the screw threads

100% inspection Every two years The Hole Plug should be carefully inspected so that it may be easily removed and installed as needed. Inspection of the male and the female threads should be assuredly effected since its coming off during operation may result in a serious accident.

f. Balance tube M⋅T, U⋅T (Wall thickness)

100% inspection Every two years Cracks in the welded part and the extent of reduction in the wall thickness of the inner part of the vent should be controlled.

g. Welded part V⋅T, M⋅T 100% inspection Every two years Cracks are liable to occur in the welded part of the pipe base and also to the trace of the welding for repair effected when it was manufactured among the wide welded area. It is also desirable to concentrate the inspection on the welded part of the Low Pressure Casing Stay.

h. Key V⋅T, Dimension (Clearance)

100% inspection Every four years It is necessary to control the clearance to be appropriate at all the keys accessible for inspection so that expansion and shrinking of the casing may not be constrained.

(2) Casing Connecting Tube a. Welded outer surface of

the seat for the extraction tube

V⋅T, M⋅T 100% inspection Every four years Cracks occur because the shape is so complicated and the wall thickness is so variable that heat stress is caused.

b. Main Steam Inlet Piping Drain Pipe Base

The outer surface: V⋅T, M⋅T The inner surface: A mirror to inspect the inside of a pipe

100% inspection Every four years Remove Pipe Base and the welded part of Drain Pipe and observe the inside of Main Pipe and the welded part of Pipe Base by means of a mirror to inspect the inside of a pipe. Effect M⋅T on the outer surface.

Inlet Sleeve Spot faced portion

Hole Plug Stopper

Outer Casing

Hole Plug Hole Plug head clearance

Inner Casing

Inspect the male and the female threads of the Hole Plug

Inner Casing

Outer Casing

The welded outer surface of the seat for the Extraction Tube

Main Steam Inlet PipingDrain Pipe Base

266




(3) Rotor a. The R portion of the

outer surface V⋅T, M⋅T 100% inspection Every four years The small size parts near the inlets of the high

pressure and the medium pressure steam and the bottom of the dummy groove require inspection.

b. Shoulders of the grooves in the rotor of the medium pressure turbine where the blades are embedded

V⋅T, U⋅T 100% inspection Every two years In case the T root is used to embed the first stage blade of the medium pressure, check whether any cracking occurs in the inside by the outer surface(U.T) of the wall of the groove for the blade.

c. The outside diameter of Rotor (High and medium pressure turbine rotors)

Dimensions The area around the steam inlet

Every four years Whether creep deformation has occurred can be determined by change in the rotor outside diameter. In this method, measuring points should be fixed to grasp the yearly change in the outside diameter.

d. Hollows where the side entry turbine blades are embedded

V⋅T, M⋅T, U⋅T 100% inspection Every four years Apply the same method as that used for the root of the rotating blades.

e. Pump Shaft V⋅T, M⋅T 100% inspection Every four years It is desirable to remove the vane of the main oil pump and to measure the part with the least cross section and some other parts.

(4) Rotating Blades a. Tenon, Shroud Ring (1) V⋅T, M⋅T,

Measurement of the lifted amount of the shroud ring

100% inspection Every four years Lifting of the shroud may be caused by heat due to sheering or touching of the tenon This problem is common in rotating blades.

(2) Measurement ofthe eroded amount of the tenon

100% inspection Every two years In the event of serious erosion, part of the caulked tenon is lost. This problem should be observed. A unit that is subject to frequent starts and stops should be observed carefully.

b. Welded parts of the stubs

V⋅T, P⋅T 100% inspection Every four years In some cases, fine cracks occur in an area of the rotating blade of the low pressure turbine to which the metal is welded. Especially for an old blade where TIG welding was yet to be used, special observation is required.

c. Portions that make the profile

Measurement of the eroded amount

100% inspection Every two years The rotating blades near the main inlet and the reheat inlet are liable to be eroded. It is necessary to grasp secular changes in the profile by means of mold transferring using a standard gauge and compound as an additional means.

Rs

rotor

The shoulder part of the first stage blades of the medium pressure turbine The outside

diameter of the

Hollows in the rotor in which the entry blades are embedded

Cross-section of a blade

Erosion of the part making the profile

Cracks in the welded part of the stub

Cracks

Shroud ring cracks

Tenon

Rotating direction

Shroud

ErosionTenon

ShroudLifted amount

(An area in which the blades are embedded)

A crack

267




d. The blade root area of the side entry blade

V⋅T, M⋅T, U⋅T 100% inspection Every four years Both for the blade and the disk, the highest stress is experienced at the corner on the first tooth. The inspection is to be effected by means of M⋅T and U.T according to necessity. However, in case the inspection has to be carried out in a limited space such as the high or medium pressure stage, P.T will be used.

e. An area of the blade in the low pressure stage on which stelite is deposited.

(a) Eroded amount V⋅T 100% inspection Every two years According to the erosion, the condition is to be classified as follows and recoating should be effected according to a planned schedule. (1) The surface is somewhat rough. (2) The surface is pitted. (3) The erosion has reached the base metal. Special care should be paid to the leading blades of a group since they are liable to be eroded.

(b) Separation andcracking

P⋅T 100% inspection Every two years Immediately recoat any cracked blade or where separation has propagated to a wide area.

(c) Bonding condition R⋅T (or U.T) All the recoated blades

When recoated Make sure without fail that they have been well bonded because the bonded condition when the coating is effected is very important.

(5) Stationary blades

(Blades near the main steam inlet and the reheated steam inlet)

Measurement of eroded amount

100% inspection Every two years The end of the outlet for the stationary blade is liable to erosion because of steam oxidation scale and flowed-in drain.

(6) Major valves a. Inner corners, welded

area V⋅T, M⋅T 100% inspection Every time when

overhauled Cracks are liable to occur around the welded areas of the baffle plate and of the valve seat lip and the trace of the welding for repair when the unit was manufactured.

b. Welded outer surface V⋅T, M⋅T 100% inspection On and after the 5th year Every eight years

Cracks are liable to occur around the welded areas of the structural members and the trace of welding for repair when the unit was manufactured.

A crack in the blade root

The blade base metal

ErosionSilver solder

Separation A crack

An image taken during R.T showing incomplete fused spray of stelite

Erosion

Blade Ring

The

stea

m o

utle

t

Welded outer surfaces

Baffle Plate

The steam inlet

Valve seat

The point where hardness is measured Main Steam Stop Valve Hardness of the flange surface

268




c. Valve Seat V⋅T, M⋅T (P⋅T) 100% inspection Every time when overhauled

Cracks are liable to occur in a part coated by stelite. A stelite deposited part should be inspected by means of P.T.

d. Flange surfaces of the main body (Only for the Main Steam Stop Valve)

Hardness Representingpoints

On and after the 5th year Every two years

It is desirable to measure the hardness of the Main Steam Stop Valve that is exposed to the severest conditions among the valves and to monitor secular changes in order to grasp the tendency of its softening due to creep.

(7) Bolts exposed to high

temperature and high pressure

a. Bolts V⋅T 100% inspection Every time when opened

M⋅T (Fluorescent magnaflux)

About 1/4 of the installed bolts

Every four years

Hardness About 1/4 of the installed bolts

Every four years

U⋅T 100% inspection Every two years

Observe damage to the threads and cracks in the bottoms of the threads. Usually, the stud bolts must not be unscrewed.

b. Inner threads for the

stud bolts Hammering test 100% inspection Every time when

opened Carry out the hammering test before loosening the nut and immediately after the unit is opened to determine symptoms of damage to the inner thread. In the event that a bolt becomes shaky or the depth of a stud bolt in the inner thread becomes shallower over time, it is considered that the damage to the inner threads has worsened.

Dimensions About tworepresenting bolts per area

Every four years

Checking of the thread profile

- ditto - Every four years

For a unit of which the service time has exceeded 8~10 years, measure the internal and the pitch diameters every four years. For the profile of the threads, copy the profile by means of compound and periodically observe secular change of the cross section.

Checking of the profile of the female threads Standard profile Actual profile

269




2 Secular deterioration check (1) Rotors a. The center hole V⋅T, P⋅T, M⋅T, U⋅T

Dimension (inner diameter)

All the planes Once at 100,000 hours in service and every ten years thereafter

For a rotor where a blind hole is drilled in the center, another of which defect was found during the inspection of the center hole effected when it was manufactured and others of a unit that has been operated with frequent starts and stops, it is desirable to inspect in a short cycle of period.

b. Grooves for the blades (For the T root and the double T root types)

V⋅T, M⋅T (or by P.T) Rotors of the 1st to 5th stages in the high pressure turbine Rotors of the 1st and 2nd stages in the medium pressure turbine

Once at 100,000 hours or so in service

For a rotor having semicircle rotating blade fixing metals, it is desirable to sample some rotating blades and inspect them since it is a matter of concern that cracking will occur in the rotor side of the seating surface of the metal and also in the corner of the jaw in the groove for the blade in the rotor. Also for a rotor that is equipped with a flat fixing metal since it was manufactured, it is desirable to check its secular deterioration by an inspection method such as U.T from the outside (for the 1st stage of the medium pressure turbine) because cracking may occur in the jaw of the groove for the blade in the rotor.

Blades

Rotor

Caulking Piece

A crack

A crack

(2) Rotating Blades a. Blade root

(For saddle shape blades)

V⋅T, U⋅T, Measurement of lifted amount

100% inspection 8th to 10th year and every four years thereafter

Effect U.T and monitor the lifted amount of the blade root in order to control the creep deformation due to prolonged service under high temperature and the stress corrosion cracking of the stopper pin.

Rotating Blade

Stopper Pin

A crack

LiftRotor

b. Portion representingthe profile (Blades near the main steam and the reheat steam inlets)

Measurement of hardness (by means of X-ray diffraction or other methods)

Several representative rotors

8th to 10th year and every four years thereafter

It is desirable to control softening of the material used for the rotating blades every four years since the material may possibly be deteriorated after prolonged service under high temperature and this can be detected by measurement of the hardness. X-ray diffraction is available for the measurement of hardness as an inspection method without making any dent in the blade.

Measurement of hardness of the surfaces representing the profile

270




(3) Stationary blades(Blades near the main steam and the reheat steam inlets)

Measurement of inclination amount of a blade row

100% inspection 8th to 10th year and every four years thereafter

It is a matter of concern that creep deformation occurs to the stationary blades in the 1st stage of the high or the medium pressure turbine after prolonged exposure to steam of high temperature and high pressure so that difference in elongation between the rotor and the casing is constrained.

Steam

Amount of inclination

(4) Bolts exposed to high

temperature and high pressure

Destructive test One or two representative bolts per material

Every four years Select the place where the severest operational condition is realized and perform tests such as examinations of structural transformation, creep, low cycle fatigue, mechanical, etc.

271

3.4.2 Maintenance of Condensers 3.4.2.1 Inspection and Measures

Table 3.4.2-1 shows content of the maintenance and inspection to be effected at the periodic inspection.

Table 3.4.2-1: Content of the Maintenance and Inspection to be Effected at the Periodic Inspection

Item Purpose or Method Timing of Maintenance

Inspection Method Countermeasure/Improvement

1 The inside of the cooling pipes

• Inspection as to whether or not clogging of the pipe with foreign matter, corrosion or erosion has occurred

When the water chamber is opened

VI ET

• Clean it with a brush or something similar.

• Install a stop plug in pipes that water cannot pass through due to clogging.

• Effect anti-corrosion or anti-erosion treatment and install a stop plug as a precaution.

• Replace the clogged pipe with a new one.2 The outer surface

of the cooling pipe • Inspection of erosion and

damage When the main body is opened

VI • Install a stop plug in the damaged pipe. • Effect the anti-erosion treatment. • Replace the damaged pipe with a new

one. 3 The surface of the

pipe plate • Inspection as to whether or not

and how marine creatures and dirty matter adhere

• Checking of the connecting part of the cooling pipes


VI • Clean it with a plastic scraper, deck brush, etc.

4 The inside of the water chamber

• Inspection as to whether or not a swell, separation, damage or a pin hole has appeared on the rubber lining

• Inspection as to whether or not and how marine creatures and dirty matter adhere


VI PHT

• Repair the damaged part. • Clean it with a plastic scraper, deck

brush, etc.

5 The inside of the main body shell

• Inspection of erosion and damage caused by steam and drain attack, and inspection of the burned out part

• Inspection as to whether or not any scale or dust has been deposited.

When the main body is opened

VI PT

• Replace the eroded part with a new one, and install a protective cover.

• Clean the hot well.

6 The pressure-resistant portion of the main body shell

• Inspection as to whether or not any cracking has occurred in the shell plate, welded part, or fixing part of the nozzle stub


VI PT WT

• Repair the damaged part.

7 Connecting piece the rubber expansion joint

• Inspection of deterioration of the rubber expansion joint by viewing from the inside of the body

Once a year after the five cumulative years since the first steam extraction

VI ST WT

• Replace them with new ones after about ten years.

8 The extraction steam pipe expansion joint

• Inspection as to whether or not damage, breakage, adhesion. etc. has occurred

Once a year after the five cumulative years since the first steam extraction

VI PT

• Repair the damaged or broken part. • Establish a schedule to replace it with a

new one after 20 years.

9 The feed water heater outer cover (Lagging)

• Inspection as to whether or not damage, breakage, etc. has occurred


VI PT

• Repair the damaged or broken part.

V.I.: Visual Inspection E.T.: Eddy Current Test P.T.: Liquid Penetrant Test W.T.: Leak Test by Filling Water S.T.: Hardness Test P.H.T.: Pin Hole Test 3.4.2.2 Cleaning of the Water Chamber and the Surface of the Pipe Plate

Negligence in cleaning of the water chamber and the surface of the pipe plate allows marine creatures to adhere to them and strongly propagate on them so that the cooler pipes are so persistently clogged, requiring too much labor to remove them. Therefore, cleaning of the water chamber and the pipe plate shall also be scheduled whenever a planned shutdown or opening of the water chamber is expected.

272

3.4.2.3 Cleaning of the Inside of the Cooling Pipes Cleaning of the inside of the cooling pipes is important to maintain the performance of the condenser, and

Table 3.4.2-2 shows the method of cleaning. However, it is necessary to check the properties and condition of the scale deposited on the inside wall and to select an effective method since it may be different according to such properties and condition. When the pipe is clogged with foreign matter, remove it first, and then carry out the cleaning.

Table 3.4.2-2: Cleaning Methods for the Inside of a Cooling Pipe Cleaning Method Properties of the Scale Method and Procedure

1 Washing out with a nylon brush

Algae-containing scale

• Feed the nylon brush through the pipe by pressurized water from a pressure feed-type water gun.

• The water gun pressure should be about 0.6 ~ 0.8 MPa. • Attach a rubber guide to the tip of the water gun in order not to damage the tip

of the cooling pipe. • Wet the nylon brush in advance. • Feed the nylon brush against the direction of the flow of the cooling water. • Take a precautionary measure using a protective sheet so that the brush does

not directly touch the rubber lining in the water chamber on the opposite side. 2 Cleaning with a ball

purge-type cleaner

Slimy scale • Feed by pressurized water several granulated balls per pipe that are used by a ball cleaning equipment through the pipes with a water gun.

• Feed the balls against the direction of the flow of the cooling water.

3 Cleaning with a rotary tube cleaner

Scale containing a small amount of algae

• Feed by pressurized water a rotary tube cleaner with a water gun. • Wet the tube cleaner in advance. • Feed the cleaner in the direction of the flow of the cooling water at first, and

afterwards, feed it against the direction of the flow. 4 Cleaning with a

chemical detergent Hard scale • Use a neutral cleanser; never use a chlorine- or acid-containing detergent.

• Do not effect this treatment until it is ensured that the discharged rinsing water has been completely neutralized to be harmless to the environment.

3.4.2.4 Leak Test of the Cooling Pipes

In the event that leakage from a cooling pipe is found, it is necessary to exactly identify from which pipe among thousands or ten of thousands of pipes the leakage is occurring. Figure 3.4.2-1 shows the methods of checking.

Point of leakage Point of leakage

Rubber plug

Wrapping sheet

(4) Water-filling method (1) Wrapping sheet (very thin plastic film) method

Black lightRubber plug

Point of leakage Point of leakage Manometer

(5) Fluorescent agent method (2) Water manometer method

Point of leakageFoam

Rubber packing Rubber packing

Point of leakage Rubber plug Transparentacrylic resin

cap Foam

Application of soapy water A vacuum meter

(3) Foam method Vacuum-breaker

(6) Vacuum pump method Vacuum pump(an ejector) Check valve

Compressed air Figure 3.4.2-1: Methods for Leakage Check of the Cooling Pipes

273

(1) Wrapping sheet (very thin plastic film) method This method makes it possible for the unit to be operated with a single system. One end of the pipe plate is

blocked by a rubber plug, and the other end is covered by wrapping sheet or both ends are covered by the wrapping sheet, and leakage is detected by a hollow in the sheet. (2) Water manometer method

This method makes it possible for the unit to be operated with a single system. One end of the pipe plate is blocked by a rubber plug, and a water manometer is connected to the other end by pressure welding. Sucking up of a water column of the manometer shows that there is leakage from the pipe. This method can detect even a pin hole, but it takes time to check all the pipes since the test needs to be effected pipe by pipe. (3) Foam method(8)

This method makes it possible for the unit to be operated with a single system. Foam is sprayed on the surface of the both opened pipe plates from a fire extinguisher-type spray gun. If the foam sinks into the inside of the pipe, it means there is leakage from the pipe. This method can detect a relatively small hole, but has a little difficulty to detect a hole in the expanded part of a pipe. (4) Water filling method

This method can be used when operation of the unit is stopped and should be effected after it is ensured that the water-filling support is in good condition. Any leakage shall be checked after the main body is completely filled with demineralized water and left without being touched for 48 hours or longer. Moisture in the water chamber may cause dew formation, making detection difficult. This method can detect leakage from a pipe and a joint portion of a pipe. (5) Fluorescent agent method

The procedure for this method is the same as that of the water-filling method explained above, except that a whitener of the diaminostilbene-type fluorescent agent is added to the filling water. This agent decreases surface tension and increases osmosis of water to make it easier to detect leakage compared with the usual water-filling method. The leaking point radiates light in a dark room when black light is irradiated on it. It is required to neutralize and treat the liquid for discharging it after the test. (6) Vacuum pump method

This method can be used when the operation is stopped. One end of the pipe plate is blocked by a rubber plug, soapy water is applied over the other side of the pipe plate, and that side is capped by a transparent cap made from acrylic resin or a similar material. The air around the cap is sucked with a vacuum pump, and the leaking point is found by movement of the soapy water foam. Leakage can also be detected by monitoring a drop in the vacuum meter needle with the stop valve closed after having sucked the air. Leakage from the pipe or the connecting part of the pipe can be detected. 3.4.2.5 Checking of air leakage

Too much air leakage to the condenser lowers the steam-condensing capability and the degree of vacuum, resulting in inability to limit the load on the steam turbine or in inability of operation. Checking of air leakage can be effected with flon or helium gas as shown in figure 3.4.2-2. But flon is a regulated material from an environmental protection viewpoint. This method is suitable for checking of the many parts of the unit including the turbine and the piping. In the event that the degree of vacuum is lowered before or after the periodic inspection, parts dismantled and repaired in the meantime and connecting parts such as flanges should mainly be checked. The water-filling method explained above may be used when the checking is just for the condenser system.

Breakable diaphragm

Turbine casing

Flon or helium gas

Low-pressure feedwater heater

Rubber expansion joint

Connecting piping Connecting piping

Gas detector

Connecting piping Connecting piping

Air-cooling section

Vacuum pump

Figure 3.4.2-2: Method for Air Leakage Detection

274

3.4.2.6 Eddy Current Test (9)

The built-in-type eddy current test is a checking method to find damage to the cooling pipes. This checking should be effected at every periodic inspection, and the data should be compared and arranged in good order according to the passage of time. The eddy current test is a detection method that utilizes electromagnetic induction. There are two methods: synchronized detection and phase analysis. Both or only the latter should be effected. As shown in Figure 3.4.2-3, passing an alternative current through a coil situated near a metallic material causes an eddy current in it due to electromagnetic induction. Such defects as cracks or any variation in the material properties would change the eddy current, changing the impedance of the coil.

The cooled pipes

The AC coils

Figure 3.4.2-3: Built-in-type Eddy Current Flaw Detection Coils(15)

The condition of the cooling pipe can be known if this change in impedance is converted to voltage and

recorded using a suitable electric circuit. Figure 3.4.2-4 shows how the flaws are detected.

Dent in the outer surface Synchronized detection method

Synchronized detection method


Phase analysis method


A foreign metal adhered to the outside surface Corrosion of the outside surface





Figure 3.4.2-4: Detected Wave Form vs. Type of Flaw

Since this method can only detect variation in the volume due to damage, a sample pipe should be extracted corresponding to the wave form, and correlation between the wave form and any damage should be established. 3.4.2.7 Replacement and Blocking of a Cooling Pipe

Replace and block a cooling pipe from which leakage is found. Figure 3.4.2-5 shows the procedure for effecting the methods.

Pipe from which leakage occurred(1) Crimping of the end of the pipe Pipe

platePipe plate

Rubber plug, brass rod, etc.

Old pipe

Copper rod

Procedure for blockng a pipe (2) Extracting of a pipe Pipe plate Copper rod

Old pipe

Fix a brass plate with silver soldering. Pipe

plate(3) Inserting and expanding of a new pipe Pipe plate Expander A rubber plug

New pipe Procedure for expanding an d

Procedure for pipe replacement

Figure 3.4.2-5: Procedure for Replacing and Blocking a Cooling Pipe then blocking a pipe

3.4.2.8 Checking of a Connecting Piece Rubber Expansion Joint

The service life of a rubber belt is about ten years. So, the following checking and measurement shall be effected once a year after five years since the start of operation, and replacement shall be prepared after ten years. (1) Visual inspection

Check whether or not any cracks,swells, or irregularities due to aged deterioration are found.

275

(2) Measurement of dimensions Measurement of such dimensions as inclination, elongation, and shrinkage.

(3) Measurement of hardness

Replace it with a new one when the hardness reaches HS 80°. The hardness of the new one should be 65°±3°. (4) A leak test by filling water shall be effected when the rubber expansion joint is replaced with a new one. 3.4.2.9 Repair of the rubber lining

Figure 3.4.2-6 shows a typical method and procedure for the repair. Cut the damaged part off and grind the appropriate part and the surroundings with a grinder. Apply adhesive over it, attach a vulcanized rubber sheet, and shape them with pressure using a roller or some other tool. Make sure by visual check or by using a pin hole tester that there are no irregularities.

Application of an adhesive Vulcanized rubber

Metal part

Figure 3.4.2-6: Repair Method for the Rubber Lining 3.4.2.10 Replacement of a Bundle of Pipes

Copper alloy pipes may be replaced by highly corrosion-resistant titanium pipes when many copper alloy pipes among all of the pipes have been blocked after prolonged operation. As Figure 3.4.2-7 shows, there is a choice between the method of replacing only the cooling pipes and the pipe plates making use of the currently used support plate of the condenser, and the other method of replacing a bundle of pipes including the support plate as a module.

1. Removal of the copper alloy tubes Water chamber Inlet of the turbine bypass

2. Replacing of the existing pipes with titanium ones

(2) Cutting off of the copper alloy pipes, Removal of water chamber and pipe plate

(1) Current situation

(a) Method of replacing only the cooling pipes and the pipe plate

(3) Removal of the copper alloy pipes

The pipe plate Protective device for the turbine bypass Titanium pipe plate

The stakes

Welding of the titanium pipe

Newly fabricated water chamber(6) Installation of the pipe plates on

the side from which the pipes are inserted and installation of the water chamber, Welding of the titanium pipes, P.T. (Liquid penetrant test), The water-ftest

illing

(5) Insertion of the titanium pipes

(4) Installation of the pipe plate on the side opposite to the side from which the titanium pipes are to be inserted, the protective device for the turbine bypass, and the newly fabricated water chamber

Installation of the stakes Expansion of the pipes

(b) Method of replacing the pipe bundle module

Newly fabricated water chamber

Supporting plate Pipe bundle module

(4) Cutting off a notch at the end plate of the condenser, Removing the supporting plate and the inner structure

(5) Insertion of the pipe bundle module

(6) Recovery of the end plate, Installation of the water chamber

(7)Leak test by filling water

Figure 3.4.2-7: Procedure for Replacing of the Aluminum Brass Alloy Pipe Condenser by the Titanium Pipe Condenser

2.4.2-1 Maintenance of the Feed water Heater

Table 3.4.2-3, 3.4.2-4, and 3.4.2-5 show examples of the main failures occurring in the feed water heater, those of aged deterioration, and the main items of maintenance, respectively.

These items of maintenance and their frequency are just for reference, and they depend on the kind of a plant. Therefore, it is necessary to obtain information on other plants and to operate a plant in cooperation with the supplier.

276

Table 3.4.2-3: Examples of Main Failures Occurring in the Feed water Heater Part Name Item of Failure

Heating

［Monel metal］［Brass］［Steel pipe］［Stainless］

Stress corrosion cracking Ammonium attack Drain attack Inlet attack Stress corrosion cracking Drain attack Deposit of scale

Diaphragm of the cylindrical water chamber Fatigue damage Seal ring of the breech lock-type water chamber Cracking in the welded part Extraction steam nozzle Cracking in the welded part Body and the parts inside the body Erosion, thickness reduction, cracking due to thermal stress fatigue Diaphragm in the water chamber Erosion, thickness reduction, cracking in the welded part

Table 3.4.2-4: Examples of Aged Deterioration of the Feed water Heater

Phenomenon of Aged Deterioration Item Content

Measures for Control

Deterioration of heat transfer capability of the heating pipe

The inside and the outside of the pipe are becoming rusty (scale deposited).

Effecting the function test Renewing the equipment, Chemical washing, Jet water washing

Det

erio

ratio

n of

Fu

nctio

n

Deterioration of the structure and the function

Deteriorated function of the de-superheating system, the drain-cooling system, and the vent system

Checking in cooperation with the supplier

Damage to and deterioration of the heating pipes

Ammonium attack, Inlet attack,Corrosion, Erosion, Drain attack

Control by periodic inspection (Visual check of the heating pipe, Eddy current test)

Deterioration of the heating pipe mounting part

Loosening of the expanded portion of the pipe Erosion of the welded part

Effecting the leak test using pressurized water or air Eddy current test for the heating pipe

Deterioration of the body and the material inside the body

Erosion due to flowed-in drain and steam Local attack

Measurement of body wall thickness, Effecting inspection of the inside (Measuring of the body wall thickness using the ultrasonic measuring test)

Deterioration of the water chamber diaphragm and the corners of the pipe plate

Fatigue damage (hair cracks) Cutting off of the skin of the area where a hair crack appears Effecting the periodic inspection (Magnaflux particle inspection, liquid penetrantion test, Measurement of hardness, Visual inspection)

Deterioration of the diaphragm of the cylindrical water chamber made of forged steel

Hair cracks appear on the surface area where stress is concentrated.

Grasping of reliable information on the operation history (Replacement of the diaphragm with a new one) Effecting the magnaflux particle inspection and the liquid penetrantion test

Deterioration of the material used inside the water chamber

Fatigue damage of the diaphragm mounting portion

Periodically effecting the liquid penetrantion test

Det

erio

ratio

n of

Mat

eria

ls

Deterioration of the nozzle portion

Erosion and cracking due to thermal stress fatigue of the material around the steam inlet and the drain inlet nozzle

Effecting the ultrasonic measuring of the wall thickness

277

Table 3.4.2-5: Maintenance Items of the Feed water Heater

No. Maintenance Frequency Operation Mode Maintenance or Monitoring Item

1 Once every hour to once every few

hours In service

Level of the drain Opening angle of the valve Water quality control

2 Once a day In service Level of the drain Opening angle of the valve Water quality control

3 Once a day to once a month In service


4 Once a month to once a year In service


5 Once a year During a halt in operation

(during the periodic inspection)

Checking of the inside of the water chamber Replacing and checking of the packing Checking of the performance

6 Once every five to ten years

During a halt in operation (during the periodic

inspection)

Checking of the inside of the body, Measuring of the wall thickness Checking of the heating pipe, Measuring of the wall thickness (E.T.)

3.4.2-2 Maintenance of the Deaerator

Table 3.4.2-6 and Table 3.4.2-7 show an outline of the main failures and aged deterioration that may occur in the deaerator, respectively. Maintenance during operation is required in order to prevent and relieve these failures and deterioration and to enhance reliability.

Table 3.4.2-6: Examples of Main Failures of the Deaerator Part Name Item of Failure Nozzle exposed to high temperature Cracking due to thermal stress fatigue Bottom body plate Erosion and corrosion Spray nozzle Corrosion and abrasion Welded part Corrosion and cracking

Table 3.4.2-7: Examples of Aged Deterioration of the Deaerator

Phenomenon of Aged Deterioration Item Content

Measures for Control

Det

erio

ratio

n of

Fu

nctio

n

Deterioration of the heated deaeration system

Deteriorated performance of the heated deaeration system due to unevenness of the thickness of the water-feeding membrane caused by increased bending of the tray Deteriorated function of the vent due to damaged spray valve

Periodic control of the deaeration performance (Checking of the level of the tray, Inspection of the spray valve)

Deterioration of the materials of the body and inside the body

Erosion due to flowed-in drain and steam Local attack

Inspection of the inside Ultrasonic measurement of the body wall thickness

Det

erio

ratio

n of

M

ater

ials

Deterioration of the material of the nozzle

Cracking due to erosion of the materials around the steam inlet, the drain inlet, the outlet of the condensed water, etc. and thermal stress fatigue

Inspection of the inside Ultrasonic measurement of the body wall thickness

Table 3.4.2-8 shows the main items of maintenance. These items of maintenance and their frequency are just

for reference, and they depend on the kind of the plant. Therefore , it is necessary to obtain information on other plants and to operate a plant in cooperation with the suppliers in the same manner as explained later on the causes of and measures against representative examples of failures listed in Table 3.4.2-6.

278

Table 3.4.2-8: Maintenance Items of the Deaerator

No. Maintenance Frequency Operation Mode Maintenance or Monitoring Item

1 Once every hour to once every few

hours In service


2 Once a day In service Level of the drain Opening angle of the valve Water quality control

3 Once a month In service Level of the drain Opening angle of the valve Water quality control

4 Once a month to once a year In service


5 Once a year During a halt in operation

(during the periodic inspection)

Inspection of the inside Replacement and checking of the packing Inspecting the inside of the body

6 Once every five to ten years

During a halt in operation (during the periodic

inspection)

Checking of the functions Inspecting the inside of the body, Measurement of the wall thickness (Inspection of the tray and spray valve)

3.4.2-3 Maintenance of the Cooler 3.4.2-3.1 Maintenance and Control during Operation

The following maintenance and control shall be effected during operation of a power plant. (1) Monitoring of the attached instruments and the measured data

Monitor the data on the temperatures at the outlet and the inlet of the cooler measured by temperature meters or some other instruments to make sure that no irregularities related to function and performance occur.

Especially for seawater coolers, the pressure of the outlet and inlet of the cooler and, the opening angle of the temperature control valve (the lower the performance, the wider the opening) on the cooling water side shall also be monitored because its performance may be deteriorated due to deposited marine creatures clogging it and also because the cooling pipes may be corroded. (2) Checking of the appearance of each cooler

Check whether or not any leakage from the flanges is found and some other items. (3) Inspection of the attached equipment

If an electrochemical protection device supplied with off-device electricity is used, make sure that the corrosion control current and the corrosion control potential during operation are within the specified ranges. If a seawater strainer (seashell filtering device) is installed, check the pressure difference, washing frequency, etc. (4) Water quality control

When iron or chlorine is injected into seawater, the concentration of the solution to be injected and frequency of injection shall be controlled. (5) Storing of a device whose operation is suspended

If there is a spare seawater cooler or oil cooler, or in the event that the condenser cooler is bypassed in the summer season, its operation is suspended for a long time. In this case, seawater in the seawater system of the cooler shall be completely discharged, and the cooler shall be completely dried out for storing. 3.4.2-3.2 Periodic Maintenance

Table 3.4.2-9 shows the maintenance items to be effected during the scheduled suspension of operation or the periodic inspection. Many of the maintenance items for the seawater system of the cooler are the same as those for the condenser. However, for a cooler using seawater, which is different from a condenser as it is not usually equipped with a ball cleaning equipment, maintaining cleanliness of the cooling pipes is more difficult than in the case of a condenser. Therefore, it is desirable to increase the frequency of cleaning of the cooling pipe and to effect cleaning semiannually in addition to that in the periodic inspection of the plant by switching the operation to the spare unit.

279

Table 3.4.2-9: Maintenance Items to be Effected during Scheduled Suspension of Operation and Periodic Inspection

Name of Equipment Inspection Item Method Frequency Countermeasures Inspection of the inside of the cooling pipes

Inspect whether or not and how scale adheres and foreign matter is deposited. - Clean them when necessary (V.I.).

Once every six months to one year

Effect washing with a brush or something similar. Review the frequency of the cleaning.

Inspection of corrosion of the cooling pipes

Inspect corrosion of both the inside and the outside of the pipes using ECT.

Once a year Install a stop plug or replace the pipe with a new one.

Inspection of leakage from the expanded part and other parts of the cooling pipes

Apply water or air pressure to the inside of the body, and inspect whether or not there is leakage.

Once every two to three years

Re-expand the pipe. Install a stop plug. Replace the pipe with a new one.

Inspection of the surface of the pipe plate

Visually inspect corrosion of the pipe plate surface (V.I.).


Apply epoxy resin coating to the eroded area of the pipe plate.

Inspection of the inside surface of the water chamber

Inspect whether or not the rubber lining is damaged (V.I. and pin hole check). Inspect how the marine creatures adhere (V.I.).

Once every six months to one year

Repair/clean the lining

Seawater cooler

Inspection of the galvanic anode plate

Inspect the consumed amount of the anode material (V.I.).

Once a year Replace the anode plate with a new one.

Inspection of corrosion of the cooling pipes

Check whether or not any part of the outside or the inside of the pipe is corroded.


Install a stop plug.

Condenser cooler Oil cooler

Inspection of leakage from the expanded part of the cooling pipes

Apply water (oil) pressure or air pressure to the inside of the body, and inspect whether or not there is leakage (V.I.).


Re-expand the pipe. Install a stop plug.

V.I.: Visual Inspection ECT:: Eddy current test W.T.: Pressure test (Water, Air, or Oil pressure)

It is necessary to effect an eddy current test (ECT) at every periodic inspection for a cooler using seawater. However, the frequency of ECT for a condenser cooler and an oil cooler may be somewhat reduced provided that the water is appropriately treated to reduce the causes of cooling pipe corrosion.

A stop plug must be fitted in a pipe whose wall thickness is found by ECT to be too greatly reduced or when such a pipe must be replaced with a new one. Figure 3.4.2-8 shows the procedure for fitting a stop plug in a pipe.

Pipe plate Cooling pipe

Drive into the pipe(e.g. with a hammer)

Stop plug (e.g. brass bar)

Figure 3.4.2-8: Procedure for fitting a Stop Plug into a Pipe

280

3.4.3 Preventative Maintenance and Remaining Life Assessment techniques of Equipment and Components used in a Steam Turbine

3.4.3.1 The Main Body of a Steam Turbine 1 Typical Modes of Aged Deterioration of a Steam Turbine

Figure 3.4.3-1 shows causes and effects of the aged deterioration.

Causes Effects

Creep A creep deformationA creep rupture

Corrosion Corrosion fatigue Stress corrosion cracking

Embrittlement

Fatigue Low-cycle fatigue High-cycle fatigue

Erosion Solid particle attack Drain attack

Strength reduction

Material deterioration

Performance down

Increased possibility of

damage

Increased operating cost

Environment Start and stop Load fluctuation Corrosive

Stress

Temperature

Time

Aged deterioration (Aged deterioration

of quality)

Softening Abrasion

Figure 3.4.3-1: Causes and Effects of Aged Deterioration of a Steam Turbine

The inlet temperature of a steam turbine is as high as 500°C or more, and the steam at the final stage where it finishes its expansion is in the wet condition at a temperature of about 33°C and at a wetness fraction of about 10%.

Various kinds of deterioration occur due to the difference in the operating conditions mentioned above. 1.1 Creep

A material that is subjected to a load under high temperature gradually deforms and finally cracks and breaks. This phenomenon where a material gradually deforms is called creep, and the cracking or the breaking is called creep rupture. (1) Creep deformation

Deformation due to creep can be a cause of aged bending of the steam turbine rotor, aged deformation of the wheel casing, and aged lifting of the rotating blade. (2) Creep rupture

When creep enters an acceleration zone, the strain increases and the material finally breaks. A design to prevent creep rupture and cracking life assessment are effected based on creep rupture

characteristics (time taken until a rupture occurs under a certain temperature and stress). Macroscopic deformation of a material caused by creek can be detected by measurement of the dimensions and

lifted amount of the rotating blade and the shroud. (3) The mechanism of a creep rupture

The assumed process of a creep rupture is as follows. (1) A microscopic crack or a small void appears at the grain boundary during a creep deformation. (2) Such cracks or voids grow and combine. (3) Finally an inter-granular rupture occurs. Figure 3.4.3-2 systematically shows this process.

281

A start of a microscopic crack

Growth of a microscopic crack Inter-granular slide

Voids are combined and a microscopic crack appears

and is growing A void is formed and

growing A void is formed

Figure 3.4.3-2: The Mechanism of a Creep Rupture 1.2 Fatigue

Figure 3.4.3-3 systematically shows the process of formation and diffusion of a fatigue crack.

The direction of repeated stress

A crystal grain

A crack in the slip zone appearing at the initial stage

A cleavage crackThe direction of growth The inside

(The outside)The outside(The inside)

A ductility striation

A brittle striation that appears mainly under a corrosive environment

The final slant separation fracture The

first stage (Formation)

The second stage

(Diffusion of a crack)

Figure 3.4.3-3: Formation and Diffusion of a Fatigue Crack (1) Low-cycle fatigue

Low-cycle fatigue is fatigue in which the total number of stress cycles to cause a rupture is small and strong stress exceeding the normal proof stress of the material is applied to a point where stress concentrates or a certain point so that plastic deformation is caused and finally repetition of the stress causes cracking. (2) High-cycle fatigue

High-cycle fatigue is fatigue in which the total number of stress cycles to cause a rupture is very high, and it is very difficult in many cases to detect the symptom from the outside in a non-destructive way. (3) The mechanism of a fatigue rupture

The assumed process of a fatigue rupture is as follows. (1) Local plastic strain is repeatedly applied to a material at its surface or a point with an internal defect. (2) Under such a situation as described above (1), a slip line appears in the crystal grain and increases to form a slip zone. (3) Finally, microscopic cracking occurs along the slip zone and diffuses. 1.3 Embrittlement

Materials used for steam turbines are exposed to high temperature during operation for a long time, and their toughness and ductility are reduced. Brittleness is the result of this process and is progressive as time passes.

The turbine wheel casing, the rotor, the main valve, etc. that are used under high temperature are liable to suffer from this phenomenon.

Generally speaking, brittleness that appears after heating at high temperature for a long time is caused by segregation of such trace elements as phosphorus (P) and tin (Sn) reducing the grain boundary strength.

Figure 3.4.3-4 systematically shows the process of aged brittleness occurring to the materials used in steam turbines.

282

Acceleration of the graiundary segregatio

and embrittlement

n bo n

Diffusion of trace elements

Grain boundary segregation of trace elements

Acceleration of bulkinization and embrittlement

Condensation and bulkinization of carbides

Precipitation of grain boundary carbides

Figure 3.4.3-4: Mechanism of Embrittlement Phenomenon

Embrittlement occurs at temperatures of 350°C or higher, and does so in a relatively conspicuous manner in the temperature range between 450° and 500°C.

When a material becomes brittle, its resistance to unstable rupture (brittle fracture) and its ductility are reduced, and diffusion velocity of a crack is increased at the same time. For this reason, a big cracking may start from a point where stress concentrates such as a subsisting casting defect. It is necessary to extend the warming-up time in the rotor in order to prevent bursts that may occur in a cold start. 1.4 Corrosion

The stage of a turbine near the dry-wet alternating area that becomes wet with a heavy load and dry with a light load is an area where corrosion must be especially observed. In such an area, a phenomenon occurs where traces of corrosive substances dissolved in water droplets sometimes condense due to the alternation between a wet condition and a dry condition caused by load fluctuation or by starting and stopping.

And corrosion and pitting occur more or less in a steam turbine that has been operated for a long time because when the steam turbine stops, steam becomes droplets that attach to the metal surface even near the last stage where wet steam flows and in the higher stage where the temperature is higher. (1) Corrosion fatigue

Corrosion fatigue occurs when corrosion and repeated stress exist concurrently. The fatigue strength of a material is lowered under a corrosive environment. This phenomenon is

conspicuous especially in the case of high-cycle fatigue. The tenons and the shroud for the rotating blades in the wet area are liable to be affected.

In the case of corrosion fatigue, the higher the concentration of the corrosive substances and the longer the exposure to corrosive substances, the more the fatigue strength is lowered. The fatigue limit is outstandingly lowered compared with that in a dry environment. (2) Stress corrosion cracking (SCC)

The type of stress corrosion cracking is either a crystal grain boundary fracture or a transgranular fracture depending on the material, the stress, and the environment.

The feature of SCC is a delayed fracture under a specific stress, and it can occur under a stress that is only a fraction of the stress with which a material fracture occurs under a non-corrosive environment.

Under the same corrosive environment, SCC is more likely to occur with higher stress, and with the same stress, it is more likely to occur to a material with higher strength. 1.5 Erosion

Erosion that occurs to materials used in a steam turbine is mainly caused either by solid particles or by drain (water droplets). (1) Erosion by solid particles

Small solid particles of oxidized scale flying in steam from the boiler are in some cases the main cause of erosion of the nozzles and the rotating blades in the high- and medium-pressure stage. Erosion often occurs especially to the nozzle plate in the first stage, where stress working on the rotating blades in the first stage is increased, affecting the reliability of the rotating blades in extreme cases.

In the event that the area of the nozzle throat is increased or the profile of the effective part of a rotating blade is changed by the influence of erosion, the efficiency of the turbine is reduced. (2) Drain (wastewater) attack

Erosion occurs mainly to the rotating parts by wastewater produced in the wet area. Typical erosion of this

283

kind is seen on the rotating blades of the final stage. In another case, in the event that leakage occurs to a horizontal joint plane inside the wheel casing in a low-

pressure section, steam containing wastewater passing through the joint erodes the metal surface in certain cases. 1.6 Softening

The necessary strength and ductility of the materials used for the parts exposed to high temperature are maintained by heat treatment. However, various mechanical properties related to the strength are deteriorated due to the effects of temperature and stress caused by operation under high temperature for a long time. One of these phenomena is softening.

The higher the temperature and the stress, the more conspicuous the softening. Softening of a material used in a turbine can be detected by measurement of hardness.

1.7 Abrasion

Although the gasket installed between the rotor and stationary parts such as the nozzle is a non-contact type, light contact between them may occur due to deformation of the wheel casing during a thermal transitional period such as the start-up of the turbine.

Gaskets are gradually abraded due to light contact resulting in increased leakage of steam between two stages to cause aged deterioration of efficiency.

Journals of the rotor and bearings can suffer from abrasion and sliding scratches after prolonged operation. 2. Object Components and Areas in a Steam Turbine to be Assessed and Damage to Them 2.1 Object Components and Areas to be Assessed in a Steam Turbine

The rotor, the wheel casing, the rotating and stationary blades, and the main value are the objects of periodic maintenance and life control because they are the main components of a turbine.

Figure 3.4.3-5 to 9 show the object components and areas of each item of main equipment for which life assessment must be effected.

An enlarged view of the dummy part

The T-root type

The dummy groove The medium-pressure rotating blades

The place of the tenon The medium-pressure

rotating bladesThe high-pressure

rotating blades A shroud

The profile

The blade root

A rotor The center hole

An enlarged view of the blade groove

The side entry type The dummy part The outside surface

Figure 3.4.3-5: Object Components and Areas of the High- and Medium-pressure Rotor (reaction type)

for Which Life Assessment Must be Effected

The profile The place of the tenon

The outside surface The center holeA diskA disk

The blade root

A diskThe medium-pressure rotating blades

The dummy partThe high-pressure rotating blades The dub tail type


Figure 3.4.3-6: Object Components and Areas of the High- and Medium-pressure Rotor (impulse type) for Which Life Assessment Must be Effected

284

The profileThe low-pressure rotating blades A disk The part consecutively

connected by metal parts one by one

The blade rootThe center hole The outside surface

The side-entry type The fork type


Figure 3.4.3-7: Object Components and Areas of the Low-pressure Rotor for Which Life Assessment Must be Effected

A corner R portion A corner R portion A main steam nozzle

A bolt

The flat portion

Section A - A Figure 3.4.3-8: Object Components and Areas of the High-pressure Wheel Casing

for Which Life Assessment Must be Effected

A bolt

The valve rod

The valve casing

Figure 3.4.3-9: Object Components and Areas of the Main Valve (Steam-adjusting Valve) for Which Life Assessment Must be Effected

2.2 Use Environment of Main Equipment and Life Consumption Factor

Figure 3.4.3-1 shows causes of damage in components and areas of main equipment of steam turbines. Among above-mentioned agreed deterioration modes, softening and creep are mainly caused by temperature-

related factors, so the high-pressure and medium-pressure turbines and the main valve that are subjected to high-temperature and high-pressure steam are assessed.

As low-cycle fatigue occurs due to repeated thermal stress and centrifugal force caused by starting and stopping of the operation at the power plant, the corner portions where stress concentration occurs are assessed for the high-temperature and high-pressure, or rotating parts of the equipment.

Corrosion and drain erosion mainly occur in the wet-dry alternating stage and wet stage pf low-pressure turbines.

Solid particle corrosion caused by materials flying from boilers etc. occurs at the inlet portion of the high pressure and medium-pressure turbines. In many cases, the first stage nozzle plate is damaged.

285

Table 3.4.3-1: Main Parts of a Steam Turbine and Causes of Damage Cause of Aged Deterioration Test

Equipment Place/Component

Cre

ep

Fatig

ue

Embr

ittle

men

t

Cor

rosi

on

Eros

ion

Softe

ning

Abr

asio

n

Rem

aini

ng li

fe

asse

ssm

ent

UT MT PT VT

The center hole The blade groove The disk The dummy groove

The high and medium-pressure rotor

The outside surface The center hole The blade groove The disk

The low-pressure rotor

The outside surface The section of the shroud and the tenons The profile

The high and medium-pressure rotating blades The blade root portion

The part consecutively connected by metal parts one by one

The profile The low-pressure rotating blades

The blade root portion The nozzle The profile

The corner R portion The flat portion

The high and medium-pressure wheel casing The bolt

The valve casing The valve rod The main valve The bolt

UT: Ultrasonic Flaw Detection Test MT: Magnaflux Flaw Detection Test PT: Penetrant Flaw Detection Test VT: Visual Inspection 2.3 Example of aged deterioration of the steam turbine

Example of aged deterioration of the steam turbine in terms of four major cases of damage (creep, fatigue, corrosion, erosion) are described below. (1) Creep damage to the Tenon for a medium-pressure rotating blade

Photo 3.4.3-1 shows an example of creep damage that occurred to the tenon for a rotating blade near the inlet of the medium-pressure turbine.

A structure where the top parts of the neighboring rotating blades in the reaction stage are connected in the peripheral direction by the shrouds is adopted in order to improve the vibration characteristics. Traditionally, the blades and the shroud were clinched by means of a tenon.

Areas around the inlet of the medium-pressure turbine can become hot, and strong eccentric force works on the blades and the related parts in these areas because of the wide blade span, then such factors together with stress concentration on the tenon cause creep cracking.

Photo 3.4.3-1: Creep damage to the Tenon for a medium-pressure rotating blade

286

(2) Fatigue cracking damage to the base part of the high- and medium-pressure disk Photo 3.4.3-2 shows an example of low-cycle fatigue cracking damage occurring to the base part of the high-

and medium-pressure rotor due to repeated thermal stress caused by starting and stopping of the turbine. Thermal stress occurs due to temperature difference between the inside and the surface of the disk. This is

caused by the large heat capacity of the rotor, leading to disagreement in temperature between steam and the metallic part when the steam is introduced and also to inability of the inside of the disk to follow the rapid increase in temperature afterwards.

Moreover, since stress concentration occurs around the corner and the groove parts of the rotor surface, plastic deformation is repeated every time the turbine is started and stopped, leading to accumulated fatigue and finally to cracking.

Photo 3.4.3-2: Low-cycle fatigue to the base part of the disk

(3) Corrosion fatigue damage to the part of a rotor in which rotating blades are embedded

Photo 3.4.3-3 shows an example of the corrosion fatigue occurring to the part of a low-pressure rotor in which a rotating blade is embedded.

Impurities or corrosive substances in steam concentrated in the gap between the rotor and the rotating blades cause corrosion pitting. Starting from a pit, a fatigue crack develops and expands because the fatigue strength of a material is reduced under a corrosive environment.

A crackThe first hook

A specimen to be taken The second hook

The starting point The third hook (the inlet side) (The outlet side)

Photo 3.4.3-3: Corrosion Fatigue of the Part of a Rotor Where Blades are Embedded (4) Erosion of the nozzle of the first stage (1)

Chipped parts are often found on the first-stage nozzle of the high- and medium-pressure turbine, and Photo 3.4.3-4 shows an example of the chipped part. This is assumed to be caused by oxidized scale that has been separated from the boiler pipe flying into the turbine and finally crashes into the nozzle at high speed. This phenomenon is called solid particle erosion (SPE). Since there is concern that the developed erosion would reduce the internal efficiency and have a bad influence on the first-stage rotating blades, repair needs to be effected in a timely manner.

Photo 3.4.3-4: Erosion of the Nozzle of the First Stage

3. Techniques for Remaining Life Assessment of a Steam Turbine 3.1 The methods of remaining life assessment

Usually, methods for the remaining life assessment of major materials for steam turbines that have been operated under high temperature and high pressure for a long time are roughly classified into the following three.

287

(1) Remaining life assessment by a destructive test (direct assessment method) Among the test methods using destructive tests, one is where the specimens are sampled from the materials of

an actual turbine in service, and another is where the materials thrown away are utilized. (2) Remaining life assessment by a non-destructive test (direct assessment method)

Conventionally, the main non-destructive tests for periodic inspection include visual inspection (VT), the penetrant flaw detection test (PT), the magnaflux flaw detection test (MT), and the ultrasonic flaw detection test (UT). (3) Remaining life assessment by analytical calculation (indirect assessment method)(2)

The analytical calculation method is an assessment method where damage to the materials (consumed life) and the time until cracking occurs (remaining life) are calculated based on temperature and stress analysis effected by means of the finite element method (FEM) or of a kind of simple calculation, the operation history, and the data on the materials. 3.2 Techniques for remaining life assessment by means of non-destructive inspection

Here, we introduce techniques for remaining life assessment by means of non-destructive inspection in relation to each kind of damage or aged deterioration.

Table 3.4.3-2 shows examples of remaining life assessment techniques by means of non-destructive inspection now being applied to actual steam turbines, and an outline of each method is given below.

Table 3.4.3-2: An Example of Techniques for Remaining Life Assessment by a Non-destructive Test Cause of Damage Assessment Method Parameters to be detected Instruments/Measuring Method

Measurement of hardness Hardness A portable hardness tester Measurement of hardness together with analytical calculation

Hardness A portable hardness tester

Measurement of electric resistance Electric resistance An electric resistance-measuring device

By A parameters Creep void The replica method By the average length of the void Creep void The replica method By the area rate of the void Creep void The replica method By the mean area rate of the carbides

Carbides The replica method

Creep

By comparison of the structures (Area affected by welding heat)

Voids, Minute cracks, Structural change, Precipitated substances

The replica method

Measurement of microscopic cracks

Length of microscopic cracks The replica method

Measurement of hardness Hardness A portable hardness tester Measurement of hardness together with analytical calculation

Hardness A portable hardness tester Fatigue

Measurement of X-ray diffraction Half-value width An X-ray diffraction device Polarization Current density A polarization test device, A small-

size electrolysis cell Embrittlement

Chemical etching Surface roughness, Intergranular corrosion cleavage width

A surface roughness tester, The replica method

3.2.1 Creep damage (1) A technique by means of hardness measurement

A technique to assess remaining life by means of change in hardness taking place in a long period of operation under high temperature is widely used because it is easy to carry out and the quantitative accuracy of the assessment is relatively good. Removal of the surface layer and metallographic tests of the specimen are effected to avoid unfavorable influence on the surface layer whose quality has changed due to decarburization and machining. An Equotip hardness tester or a Shore hardness tester is used for field use.

Some methods of assessing remaining life based on the measurement of hardness are explained below.

288

(i) A method by means of measurement of hardness(2) ~ (6)

The hardness of the low-alloy steel for equipment used in steam turbines that is exposed to high temperature drops due to the change in the metal structure under high temperature (deformation resistance change). This drop in hardness occurs even with no load as shown in Figure 3.4.3-10 (heating with no load) and is accelerated by a load being imposed.

Heated material with no load

Vick

ers

hard

ness

(Hv)

Materials that receivcreep damage

Bef

ore

heat

ed

Temperature(°C)

Heated material

with no load

Materials that received creep

damage 450 ⎯⎯ 500 550 550

ed

[Cr-Mo-V steel]

Figure 3.4.3-10: Relation between the Temperature Time Parameter and Hardness

The hardness test method assesses the creep damage rate by means of the amount or rate of hardness drop measured in actual steam turbines. Figure 3.4.3-11 shows the relationship between creep damage rate φc and drop in hardness ∆Hv (difference in hardness between a material with a load applied and another with no load applied), and the creep damage rate is obtained from the measurement of hardness making use of the figure.

CrMoV forged iron

CrMoV cast iron

Dro

p in

har

dnes

s (∆

Hv)

Creep damage rate φc (t/tr)

[Cr-Mo-V steel]

Figure 3.4.3-11: Relation between Creep Damage Rate and Drop in Hardness (ii) A method of using hardness measurement together with analytical calculations (1)

This method assesses the creep damage rate by means of creep rupture characteristics after deterioration that are obtained from hardness measured in an area that received thermal aging (an area exposed to high temperature but only to low stress), as well as by means of the analytical calculations. Figure 3.4.3-12 shows the summarized test results of creep rupture characteristics represented by hardness and temperature time parameters. And creep rupture characteristics after aged deterioration are obtained from measurement of hardness on actual steam turbines and the calculation result of temperature and stress by Formula (1) that is induced from the figure.

T(C+logtr)=Σai⋅(logσ)i-1⋅Hv+Σbi⋅(logσ)i-1 ........................................................................... (1) where tr: Creep rapture time T: Absolute temperature C: Material constant Hv: Vickers hardness σ: Working stress ai⋅bi: Approximation constants

289

Stre

ss σ

(kg/

mm

2 )

(Estimation according to the formula (1))

Values obtained by experiments

Temperature time parameter (×10-3)

[Cr-Mo-V steel]

Figure 3.4.3-12: Comparison between Data on Creep Rupture Characteristics Obtained from Experiments and its Estimation Based on Measured Hardness

(2) The electric resistance method (3)- (5)

The electric resistance method measures structural change that occurred during operation at high temperature (cohesion bulkinization of carbides, drop in solid dissolved carbon in the matrix, and some others) by means of the potential difference method.

Like in the case electric resistance drops even by heating with no load applied, and the drop is accelerated with stress loaded. That is, the higher the creep damage rate, the greater the drop in electric resistance.

The electric resistance method assesses the creep damage rate by the ratio between the difference in electric potential measured on actual equipment and the same measured on a specimen that has not been subjected to deterioration (electric resistivity ratio).

Figure 3.4.3-13 shows the relationship between creep damage rate φc and the dropped amount of the electric resistivity ratio ∆Rρ (difference in electric resistivity between a material with a load applied and the same material with no load applied), and the creep damage rate is obtained from the measured electric resistance making use of this figure.

Dro

pped

am

ount

of e

lect

ric re

sist

ivity

ratio

(∆

Rρ)

Creep damage rateφc (t/tr)[Cr-Mo-V steel]

Figure 3.4.3-13: Relation between Creep Damage Rate and Dropped Amount of Electric Resistivity Ratio (3) The structure-observing method

The structure-observing method is where a metal structure is observed by means of a replica, and the extent of damage is assessed by the degree of structural change. Since the change in the structure due to creep damage itself can possibly be grasped, it is an important technique. Concretely speaking, the structure is transferred to a replica film after the portion of metal to be assessed is polished and etched. The vapor deposition of gold is applied to the replica, and the replica is observed by a scanning electron microscope.

Figure 3.4.3-14 and Photo 3.4.3-5 show the procedure for picking up of the replica and an example of observation of a creep void (a cavity), respectively.

290

Photo 3.4.3-5: An Example of an Observed Creep Void by Means of a Replica

(i) The replica

film (ii) Transfer of the structure

(iii) Picking up the replica

(iv) Vapor deposition of gold (to give the replica electric conductivity)

A crack or a void A carbide

The replica

A metal specimen (polished and etched) Figure 3.4.3-14: Procedure for Picking up of a Replica

For creep, there is a special feature where at first, formation of a creep void is recognized during the process of

damage and it develops, expands, and combines to become a crack through a microscopic crack. There is a correlation between the situation of void formation and the creep damage rate.

The following methods are proposed for quantifying the occurrence of a void. (i) The “A” parameter method (ii) The mean length of void method (iii) The area rate of void method (iv) The mean area of carbide method (v) The structure comparison method

3.2.2 Fatigue Damage (1) The microscopic crack measurement method (10)(11)

Early in life, many minute cracks appear on the surface of a material that received fatigue damage due to thermal stress fatigue, and they grow or repeat integration to constitute a major crack. The microscopic crack measurement method assesses fatigue damage by detection by means of the replica behavior of the growing microscopic cracks before they become a major crack.

Figure 3.4.3-15 shows the data. These data were obtained by means of a replica that detected the growing process of minute cracks with a high-temperature, low-cycle fatigue test that was interrupted at each loading condition.

The

max

imum

mic

rosc

opic

cra

ck le

ngth

(mm

)

Loading condition A material used for 140,000 hours

A virgin material

Fatigue damage rate φf (N/Nf)

[Cr-Mo-V cast iron]

Figure 3.4.3-15: Relation between the Fatigue Damage Rate and the Maximum Microscopic Crack Length

291

(2) A technique by means of hardness measurement (i) The hardness-measuring method (1)(12)

For high-temperature, low-cycle fatigue, the fatigue damage rate is obtained from the measurement of hardness making use of the relationship between the rate of the hardness after deterioration against that of the aged material and the fatigue damage rate. (ii) The method of using hardness measurement together with analytical calculations (1)(13)

In the same manner as that of the creep damage assessment, this method assesses the fatigue damage rate by means of the low-cycle fatigue characteristics obtained from the hardness of an area that received thermal aging (the absolute value of hardness), as well as by means of analytical calculations. (3) The X-ray diffraction method (1)(14)

Fatigue damage is one of the concerns regarding the bottom of the groove in the periphery of the rotor and some other portions. However, hardness measurement is difficult because of their narrowness. Therefore, the mean value width measurement by means of the X-ray diffraction method is applied. 3.2.3 Embrittlement (1) The polarization method (1)(12)(15)(16)

The polarization method assesses the degree of embrittlement by means of the relationship between the voltage and the current (polarization curve) appearing when electrolysis is caused in the electrolysis solution using a part whose embrittlement is to be calculated as the anode, as well as by means of the phenomenon where the natural electric potential varies according to advancement of embrittlement. (2) The chemical etching method (17)

The chemical etching method detects advancement of embrittlement by means of measurement of the depth of the grain boundary corroded groove (roughness and width of the grain boundary groove) to know the amount of segregation at the phosphorus grain boundary following selective corroding of a certain grain boundary by picric acid. 3.3 Application of Remaining Life Assessment and an Example of its Verification

Figure 3.4.3-16(4) shows an example of application of remaining life assessment to a rotor in a high-pressure turbine by non-destructive inspection. The reference part is the peripheral area of the coupling where the temperature and the stress are low, and the area to be assessed is the part in the first stage of the high-pressure turbine.

Assessment of a part in the high-pressure area is not always possible because measurement and inspection of actual equipment is required to effect assessment of the remaining life by means of non-destructive inspection as shown in Table 3.4.3-2. For example, for a rotor of a high-pressure turbine, assessment of creep damage to the center hole is necessary, and a device for this purpose has been developed.

Photo 3.4.3-6(1) shows a device to pick up a replica of the center hole of a rotor and measurement of hardness as an example.

Photo 3.4.3-6: A Device for Remaining Life Assessment of the Rotor Center Hole [MACH-I]

Figure 3.4.3-17(1)(18) shows the result of remaining life assessment of a high- and medium-pressure outer wheel

casing. (The cumulative operation hours is about 160,000 hours, the number of starts and stops is 370, and the temperature of the part to be assessed is 538°C.)

292

[1]Operation history Number of starts and stops: 370 Operated time: 161,000 hours Steam temperature: 538°C

The zone where cracks may occur

[2] Result of the life assessment

The lower half

Fatig

ue d

amag

e φ f

The reheated steam Figure 3.4.3-17: An Example of Actual Application of Remaining Life Assessment of a High- and Medium-pressure Outside Wheel Casing

Safe zoneThe part to be assessed

The main steam The part to be assessed Creep damage φc

[3]Observation Result of a Replica

Observation result by a scanning electron microscope (SEM) (×1000) Symbol Mark Inspection Item

Electric resistance measurement

Hardness measurement Microscopic crack

measurement

Electric resistance measurement

Hardness measurement

Microscopic crack measurement

Figure 3.4.3-16: An Example of Actual Application of Remaining Life Assessment of a High-pressure Rotor by Means of Non-destructive Inspection

It had been predicted that a crack would have occurred at the R portion of the base part of the main steam pipe and the re-heated steam pipe on the outside surface of the lower wheel casing due to accumulated creep damage. And a crack that was considered to be the result of the formation and combination of voids was detected during the inspection carried out the following year.

Figure 3.4.3-18(1)(19) shows the results of a creep rupture test of a specimen taken from a place very near a corner of the steam chest where accumulation of creep damage was predicted to have reached almost cracking point and of observation of the creep point of the place under discussion. The specimen was obtained from a main stop valve (subjected to about 90,000 hours of accumulated operation hours, about 800 starts and stops, and a temperature of 566°C at the assessed place) dismantled for the study. The creep damage experienced in the creep rupture test was near to the predicted damage, and the fact that the creep points had already been combined to make a minute crack verified that the predicted value was correct.

Becoming a microscopic crack

The data of the R portion of the steam chest

Formation of a void in the steam chest

(Combined voids)

(The initial stage) The creep

damage rate and the creep strain A void is

formed (an image on the replica).

The creep damage rate

[The main stop valve]

Figure 3.4.3-18: Situation of the Creep Damage Rate, the Creep Strain, and the Void

293

3.4.3.2 Heat Exchangers 1. Outline

Various kinds of heat exchangers are used in thermal power plants. In this chapter, we discuss measures for improvement including a new technology to enhance future reliability of the major types of heat exchangers.

Table 3.4.3-3 shows the main failures that occurred to the condensers, the feed water heaters, the cooling water coolers, the oil cooler and the gland steam condensers due to their aged deterioration, as well as the content of improvement and countermeasures. For typical examples among them, causes and their countermeasures are explained below.

Table 3.4.3-3: Improvements of the Plant Equipment and Countermeasures against Malfunctions Improvements of such major heat exchanging equipment as the condenser, the feed water heater, the oil cooler, the

cooling water cooler, and the gland steam condenser and countermeasures against major malfunctions Maintenance, Inspection, Improvement, and Replacement Name of

Equipment Item Purpose Method of Inspection Improvements and Countermeasures

(1) Replacement of the cooling pipes with new ones

Prevention of leakage due to aged deterioration, enhancement of performance and reliability

ET

Replacement of the cooling pipes with new ones Replacement of the cooling pipes with titanium ones

(2) Improvements of the base exposed to high temperature

Reinforcement of temperature-proof capability (To prevent cracking due to aged deterioration)

PT (1) Converting the base to a thermal sleeve type (2) Reinforcement of the welded part

(3) Improvements of other drains and the steam inlet base

Countermeasures against wall thickness reduction in the baffle and some others due to erosion caused by aged deterioration

PT DI

Increasing the wall thickness of the eroded part

(4) Total inspection of the inside of the condenser body

Inspection as to whether or not erosion, corrosion, or cracking occurs on the internal structure due to aged deterioration

VI PT

Repairing the damaged part

1. Condenser

(5) Replacement of the rubber belt expansion joint with a new one

Maintaining airtightness of the condenser (Prevention of cracking due to aged deterioration)

HT VI

Replacement of the joint with a new one

(1) Measures to prevent ammonium attack (2) Replacement of the steel pipes with new ones

(1) Prevention of ammonium attack (2) Prevention of erosion due to steam and drain attack

ET Replacing the heating pipes with new stainless pipes

2. Low-pressure Feed water Heater (3) Inspection of the parts

installed inside the body and the inside of the body plate

Checking whether or not erosion or wall thickness reduction occurs

VI UT

Replacing the set with a new one

(1) Improvement of the structure of the water chamber

(a) Prevention of leakage and sudden gush from the welded part at the pipe end Prevention of cracking at the pipe end

PT

(b) Prevention of cracking due to stress concentration on the corner of the water chamber

PT

(c) Prevention of the end of the heating pipes from being eroded VI

3. High-pressure Feed water Heater

(1) Improvement of the structure of the water casing and checking of inlet attack in the steel pipes (2) Adhering of scale to the steel pipes

(2) Prevention of scale adhering VI

(1) Improvements of the structure of the water chamber (a) Adoption of a new welding method for the pipe end (b) Increasing the corner R of the water chamber (c) Installing a tube-inserted pipe Totally replacing the existing water chamber with a new one in which the above improvements are integrated (2) Effecting water jet cleaning

4. Oil Cooler

(1) Replacement of the pipe nest with a new one (2) Modification of the water chamber

(1) Measures against aged deterioration (2) Recovering of function and performance (3) Simplification of maintenance and inspection

ET VI

(1) Replacement of the pipe nest with a new one (2) Installation of a cover that is to be tightened by flanged bolts on the water chamber

5. Cooling Water Cooler

(1) Replacement of the cooler with a new one (2) Modification of the water chamber

(1) Measures against aged deterioration (2) Recovering of function and performance (3) Simplification of maintenance and inspection

ET VI

(1) Replacement of the cooler with a new one (2) Installation of a cover that is to be tightened by flanged bolts on the water chamber

6. Gland Steam Condenser

(1) Replacement of the blower with a new one (2) Modification of the blower to a separately placed type

(1) Measures against aged deterioration of the impeller shaft (2) Measures to prevent vibration of the blower

VI

(1) Replacement of the blower with a new one (2) Installation of the blower in a separate place (3) Modification of the distribution valve located around the blower

Meaning of the acronyms: ET (Eddy Current Flaw Detection Test) PT (Penetrate test) DI (Dimension Inspection) VI (Visual Inspection) HT (Hardness Test) UT (Ultrasonic Flaw Detection Test)

2. Technology for Preventative Maintenance and Measures to Strengthen the Deterioration-proof 294

Capability of the Condensers 2.1 Corrosion of the Cooling Pipe

Aluminum-brass pipes have been traditionally used for the cooling pipes of condensers. Inlet attack and deposit attack (the impingement attack) are corrosive and erosive attacks from the inside of the pipe. Ammonium attack is a corrosive and erosive attack from the outside. In particular, corrosion and erosion from the inside, such as deposit attack, can sometime pierce the wall in a short time to cause water leakage.

Such measures as the injection of iron sulfate, electrochemical protection, the injection of chlorine, ball purge, and counter flow washing are traditionally taken to prevent corrosion and erosion from the inside, but additional daily elaborate operation control is also important.

A periodical E.T. (eddy current flaw detection test) is effective as preventative maintenance for aluminum-brass pipes. And nowadays, automatic control is adopted to arrange the data of the E.T. in order and to control the remaining wall thickness.

Recently, there is a tendency to take the safety measure whereby all aluminum-brass pipes of a condenser already installed are replaced by titanium pipes. The merit is a great reduction in the risk of seawater leakage and the omission of the E.T. to be effected at periodic inspection and daily maintenance to protect the cooling pipes. On the other hand, it becomes necessary to shorten the space between the cooling pipe supports as a vibration-proofing measure so that titanium pipes with thin wall thickness can be used because titanium is somewhat inferior to aluminum-brass in terms of heat conductivity. 2.2 Cracking in the Hot Nozzle

Hot fluid such as main steam wastewater whose temperature exceeds 400°C flows into the condenser. In some cases, too much thermal stress occurs in the nozzle into which hot fluid flows due to a big difference in temperature between the nozzle and the body where the temperature is about 33°C. Table 3.4.3-4 shows examples of cracking that occurred in the part where the nozzle and the body are welded due to thermal fatigue caused by repeated heating and cooling under operation in DSS mode.

Table 3.4.3-4 Damage to the Base Portion of the Condenser Exposed to High Temperature and Examples of Their Countermeasures

Damage situation of the base portion of the condenser exposed to high temperature and examples of the countermeasures in the shape of the base

Shape of the Base

No.

Plant Output (at the

opening of the plant)

Operation Mode

Name of the Base

Time of Damage

OccurrenceDamage Situation

Original Shape Countermeasures

1 250 MW (1967) DSS

High-pressure Drain

Manifold (150 A)

Seventeen years after

the operation started

A crack of 105 mm in the body and another crack of 30 mm in the nozzle

3 250 MW (1974) DSS

Medium-pressure Drain

Manifold (100 A)

Twelve years after


A crack of 115 mm in the body and another crack of 70 mm in the nozzle

4 350 MW (1970)

At a constant load

(entered in an emergency)

SSR steam inlet

(150 A)

Nineteen years after


A crack of 178 mm in the body

5 600 MW (1973)

At a medium load

Turbine Lead Pipe Drain

Inlet (50 A)

Ten years after the operation

started

Three cracks of 80 mm max. length in the peripheral direction occurred on the welded part of the thermal sleeve and the body

Cracks(hatched area)

Fillet welding


The thermal sleeve

Fillet welding


Groove welding

Fillet welding

Cracks

Meaning of the acronyms: DSS (Daily Start-Stop) WSS (Weekly Start-Stop) SSR (Steam Seal Regulator)

Cracking can be prevented by the countermeasure where the hot side, the nozzle, and the cold side, the body, are connected through the thermal sleeve and the point of injection is chosen so that the hot fluid does not point the body to relieve steepness of the temperature gradient between the nozzle and the body.

295

3. Technology for Preventative Maintenance and Measures to Strengthen the Deterioration-proof Capability of the Feed Water Heater

3.1 Inlet Attack of the Steel Pipes Inlet attack is a phenomenon where the protective film on the inside surface of a pipe is destroyed and eroded

by water flow, and it is often seen in high-pressure feed water heaters whose temperature of feed water is as high as between 150°C and 200°C. Erosion and corrosion due to the influence of the pH and temperature of the feed water are the causes besides vortices and too high a velocity of the fluid flow. = Countermeasures =

Installation of an inserted pipe and a flow-smoothing bell mouth as illustrated in Figure 3.4.3-19 are effective.

Feed waterFeed water

Leakage

A flow-smoothing bell mouth (1.25Cr0.5Mo Steel plate)

A flow-smoothing bell mouth (1.25Cr0.5Mo Steel plate)

An inserted pipe (SUS304TB)

An inserted pipe (SUS304TB)

Protrusion welding type Intrusion welding type Figure 3.4.3-19: An Inserted Pipe and a Flow-smoothing Bell Mouth

3.2 Adhering of Scale on the Inside Wall of a Steel Pipe

In some cases, black consistent scale adheres to the inside wall of the steel pipes and some other places in the high-pressure feed water heater where steel pipes are used. This scale is iron oxide called magnetite. Too much magnetite adhered to the surface may lead to damage to the partition in the water chamber due to increased pressure loss, overload of the BFP, imbalance of the flow of wastewater, and lowered heat conductivity. = Countermeasures =

Such measures as periodical removal of scale, complete protection against rusting during storing of the unit in the case of suspension of operation, complete control of water quality when the unit is restarted, etc. can be countermeasures. Water jet washing and blast washing are usually used for the mechanical removal of the scale, and acid washing with monoammonium citric acid is used for the chemical removal of the scale. The former is superior in terms of cost, while the latter is superior in terms of removal performance. 3.3 Drain Attack of the Outside Surface of Steel Pipes

In some cases, steel pipes are attacked by erosion from the outside surface of the pipe, and leakage occurs around the area near to the inlet of the heated steam flowing into the feed water heater or near the drain. In either case, high-speed heated steam involves wastewater and blows against the outside surfaces of the front most row of the pipes so that the erosion gradually penetrates into the inside. = Countermeasures =

To keep the space between the main body and the row of the pipes so that the local flow speed of heating steam does not become too fast, adopt an arrangement or location of the nozzle that prevents the heating steam and the wastewater from interfering with each other. In addition to this measure, utilization of stainless pipes for two or three outside rows is also effective. 3.4 Ammonium Attack of the Copper Pipes

For a feed water heater in which copper-type alloy pipes such as aluminum-brass pipes are used as heating steam and the chamber in which the pressure is always negative, the pipes are eroded in some cases near the area where non-condensed gas is extracted and at the support plate and the pipe plate near to the steam inlet impact prevention plate where the concentration of ammonium is high. = Countermeasures =

Replace some of the tubes located where any parts are liable to receive ammonium attack by stainless pipes that have good corrosion resistance. Such measures as modification of the structure so that gas does not stay around parts from where non-condensed gas is extracted or a change of operational parameters such as setting of a vent amount also deserve consideration.

296

3.5 Non-destructive Inspection of the Heating Pipe Carbon steel, stainless steel, aluminum-brass, monel metal, etc. are used for the heating pipe of the feed water

heater. But the necessary frequency and type of inspection vary according to the material used. The necessity of inspection is high for carbon steel pipes and brass pipes in order to prevent leakage due to corrosion or erosion of the heating pipe, while the necessity is low for stainless pipes that are highly corrosion resistant.

On the other hand, austenite stainless and copper alloy material that necessitate less inspection can receive E.T. contrary to no necessity of inspection because they are non-magnetic materials, while checking of damage to such ferromagnetic materials such as carbon steel pipes by means of E.T. is not so easy.

The following are inspection techniques to be used for the inspection of carbon steel. (i) Magnetic saturation eddy current flaw detection method

This is a method where the influence of magnetism is removed in order to improve detection capability by magnetically saturating part of the heating pipe whose flaw is to be detected. (ii) The ultrasonic test method

This method can differentiate reduction in the thickness that occurred to the inside of a pipe from that to the outside, and is used mainly for inspection of reduction in wall thickness using a submerged rotating probe. (iii) The remote field eddy current flaw detection method

This is an inspection method where a transmission coil and a receiving coil are placed approximately a few times the pipe diameter apart from each other, where magnetic flux is induced by the transmission coil penetrating up to the outside surface of the pipe, and where the receiving coil receives the magnetic flux propagated along the axis of the pipe.

Leakage from the heating pipe can also be checked by the opening angle of the drain valve, the drain level, the temperature difference of the feed water, and abnormal noise. Although it depends on severity of the damage suffered, the most popular method is to open the water chamber and install stop plugs in the leaking pipe and other pipes suffering from the secondary damage. It is recommended to replace the feed water heater as a unit when plugs are installed in 10% of the pipes. Replacing the heating pipes with stainless ones is also an effective measure. 3.6 Damage to the Diaphragm in the Cylindrical Water Chamber of the High-pressure Feed Water

Heater There is a structure where a diaphragm is used to obtain water tightness of the water chamber of the high-

pressure feed water heater (Figure 3.4.3-20). In this structure, the high pressure in the water chamber is supported by a water chamber cover made of a thick plate and is sealed by the diaphragm. The diaphragm is not often damaged.

The outlet for the feedwater

The body

The water chamber coverThe partition cover

The diaphragm

The heating pipe

The sheer pieceThe pipe plate The water chamber

The inlet for the feedwater

Figure 3.4.3-20: The Structure of the Diaphragm

Since there are more cases nowadays where the welded part at the corner of the pipe plate where the partition plate is attached and the welded part of the covering plate for the γ hole are cracked (fatigue cracking), it is better to totally modify it.

297

= Countermeasures = It is required to replaced the diaphragm every two years taking the DSS mode of operation into consideration.

The old one should not be used but should be replaced it with a new one at such an opportunity as opening of the water chamber when the old one is broken. 3.7 Erosion of the Inside of the Low-pressure Feed water Heater Body

Reduction in the wall thickness of the parts located inside the heater such as the body plate and the pipe-supporting plate due to aged deterioration has often been experienced recently.

A study revealed that the places where reduction in wall thickness had been experienced are those where the flow of steam is relatively fast in the heater or places where there was movement of wastewater and where erosion is liable to be induced. This attack is a phenomenon due to erosion and corrosion occurring in a specific temperature range.

The corrosion speed of an iron or steel material is greatly influenced by the environmental temperature, and there is a tendency for a material to increase its corrosion speed at a specific temperature. The number of cases of the phenomenon where plates inside the body of the low-pressure feed water heater are corroded and their wall thickness is reduced around the above-mentioned temperature range has increased. = Countermeasures =

Such measures as padding by welding on the area whose wall thickness has been reduced, backing the area by a metal stripe and/or partially replacing with a newly fabricated part made of SUS material are effected. 4. Technology for Preventative Maintenance and Measures to the Strengthen Deterioration-proof

Capability of the Deaerator 4.1 Deaerator (1) Cracking in the hot nozzle

Strong thermal stress is generated in structural members when there is a steep temperature gradient or a big change in temperature takes place in an area into which hot steam flows (in the case where the temperature of the fluid itself changes, or in another case where very rapid changes in the temperature like thermal shock take place due to wastewater flowed into the hot portion). The generated thermal stress may cause fatigue damage when concentrated at the point of discontinuity and stress concentration is repeated in DSS operation mode = Countermeasures =

Check whether or not there is any incorporation of wastewater from the upper stream in the piping (e.g., failure in discharge of the wastewater due to deterioration of the wastewater trap), and take necessary measures to improve the root of the incorporation if such incorporation exists. Adopt the thermal sleeve-type structure for the nozzle as shown in Figure 3.4.3-21 in order to relieve the temperature gradient between the piping and the body-side plate. Also use full welding to avoid the occurrence of stress concentration.

The nozzleThe breast plate The breast plate

The nozzle

The thermal sleeveThe reinforcement plateFigure 3.4.3-21: Examples of Modifications to the hot Nozzle of the Deaerator

5. Remaining Life Assessment of a Heat Exchanger

As for heat exchangers, countermeasures against aged deterioration have traditionally been effected focusing on preventative maintenance. However techniques for remaining life assessment have recently been developed and have partly being applied to feed water heaters and deaerators.

Remaining life assessment is carried out by means of theoretical analyses, destructive tests, and non-destructive tests, and utilization of these techniques has altogether improved its accuracy.

298

3.4.3.3 Pump 1. Preventative Maintenance for Pumps

As for preventative maintenance of pumps, there are two types of pumps. The first one includes such pumps as the boiler feed water pump and the boiler circulating pump whose life is decided by fatigue under high temperature and high pressure. The second one is a circulating water pump whose life is decided by seawater corrosion. Moreover, there are another classification of the condition, that is, whether it is with or without a spare unit, continuously operated or not, etc. It is important to effect preventative maintenance suitable for each pump. In this section, the boiler feedwater pump and the circulating water pump are discussed. 2. Daily Inspection and Periodic Inspection

Scheduled maintenance includes maintenance by means of daily inspection and another one by periodic inspection to be effected every two years or four years. Tables 3.4.3- 4 to 5 show the items for daily inspection, and Tables 3.4.3-6 to 7 show the items for periodic inspection. The following points shall be observed when each pump is inspected.

Table 3.4.3-4: The Content of Daily Maintenance and Inspection of BFP Operation Record Discharge pressure/Suction pressure/Amount of feed water/Feed water temperature/r.p.m./Motor

current/Bearing temperature/Lubricating oil temperature/Lubricating oil pressure/Vibration/Temperature of the returned seal water (for the bushing seal type)/Temperature of the flushing water (for the mechanical seal type)

Inspection Items Abnormal noise and vibration/Leakage from the piping, the gland and the coupling/Vibration of the small-size piping/Opening angle of the sealing water control valve/Differential pressure of the strainers

Table 3.4.3-5: The Content of Daily Maintenance and Inspection of CWP

Operation Record Discharge pressure/Opening angle of the variable blade/Motor bearing temperature/Electric current/Vibration/Noise/Gland temperature/Differential pressure of the lubricating water strainer

Inspection Items Abnormal noise and vibration/Leakage from the piping/Abnormal noise and vibration/Leakage from the gland/Differential pressure of the strainer/Abnormality in the bearing lubricating water

Table 3.4.3-6: The Content of the Inspection on the Periodic Inspection of BFP

Item Content of the Inspection The sliding ring • Clearance

• Cracks (P.T. inspection) The main shaft • Measurement of the bend

• Cracks (P.T. inspection) • Measurement of the dimensions of the gland and the journal • Visual inspection

Corrosion, Abrasion, Fretting, Threads on the shaft, Key way Rotating Component

The impeller • Dimensions and run-out of the sliding part • Scale • Cracks (P.T. inspection) • Visual inspection

Cavitation corrosion, Abrasion, Erosion, Dents, Movement The outer body • Erosion of the inside surface

• Cracks in the stainless padding (P.T. inspection) • Dimensions of the joint part and scratches on the surface • Damage to and abrasion of the threads of the tightening bolts

Casing The inner body • Erosion of the mating surface

• Scale • Cracks (P.T. inspection) • Damage to the mating surface of the joint to the outer body • Loosening of bolts • Scratches and cracks • Inner body bushing/Erosion of the water extraction pipe/Deformation

The radial metals • Contact • Clearance • A crack in and separation of the padding metal (P.T. inspection) Bearing

The thrust metals • Abrasion of the thrust shoe and the disk • Damage • Checking of the thrust shoe movement

Others • Checking of the end play of the rotating components • Alignment check after dismantling and assembling

299

Table 3.4.3-7: The Content of the Inspection on the Periodic Inspection of CWP Item Content of the Inspection

The main shaft (1) Measurement of the bend (2) Dimensions of the outside diameter of the bearing sleeve (3) Dimensions of the gland gasket sliding sleeve (4) Visual inspection

Corrosion/Cracks/Abrasion/Looseness of the key/Bolt hole The impeller (1) Measurement of the outside diameter of the wearing

(2) Visual inspection • Corrosion • Cracking • Dents/Contact of the tip of the vane with the liner • Abrasion of the base of the vane entrance • Looseness of the key/Contact of the joint part

(3) Cracking in the boss and the base of the vane (P.T.) The coupling cover (1) Visual inspection

Corrosion (the general part/the flange surface/the portion for the O-ring)

Rotating Component

The shaft sleeve (1) Visual inspection Corrosion (the general part/the joint/the portion for the O-ring)

1. The bearing (1) Dimensions of the internal diameter (2) Visual inspection

• Deterioration of the rubber • Boundary separation of the rubber from the shell

Bearing

2. The shaft case (1) Visual inspection Corrosion (the spigot joint/the boundary between the case and the shaft shell)

1. The pumping-up pipe of the discharging body

(1) Visual inspection Corrosion (the spigot joint on the flange surface/the general part)

2. The stuffing box (1) Visual inspection Corrosion (the spigot joint on the flange surface/the gasket inserting part/the general part) Casing

3. The suction bell for the guide vane

(1) Dimensions of the internal diameter of the liner ring (2) Visual inspection

Corrosion (the shaft case inserting part/the joint portion with the vane entrance tip and the inner pipe/the spigot joint on the flange surface/the general part)

(1) The feed water pump

In some cases, magnetite adhered on the impeller due to poor quality control of the feed water (AVT) may increase the r.p.m. vibration for the same output. It is necessary to make sure by daily inspection that there are no changes in tendency regarding discharging pressure, r.p.m., vibration, etc.

Since corrosion problems of the hard chrome plating applied to the wearing occurred due to the quality control of the feed water (CWT) that was recently introduced, visual inspection of these parts is necessary at each overhaul.

Note 1. CWT is the acronym for Combined Water Treatment and is a kind of water treatment for boilers where combined injection of oxygen and ammonium are carried out.

2. AVT is the acronym for All Volatile Treatment (deaeration treatment) where a protective film of oxide iron (magnetite) is formed with the help of hydrazine under a deaerated condition to lower the oxygen concentration in the system to the lowest possible minimum in order to make the object corrosion resistant.

(2) The circulating water pump

In the event that operation of a circulating water pump is suspended for a long time as it is installed, it is necessary to operate it once a week or once a month to prevent pitting and crevice corrosion. 3. Measures to Strengthen Deterioration-proof Capability

In the previous section, we discussed preventative maintenance where periodic inspection is expected to identify deteriorated parts and where necessary parts are repaired or replaced by new ones.

However, the following changes in the situation are now known. (1) Power plants that are operated in DSS or WSS mode have increased.

300 (2) The percentage of power plants that have been operated for 20 years or more since their start of

operation has increased.

(3) The interval for periodic inspection of pumps has been extended. So, it is desirable that deterioration-proof capability of the pump itself be extended. Table 3.4.3-8 ~ 9 show

the measures taken by the industries to strengthen deterioration-proof capability. Table 3.4.3-8: Measures Taken to Strengthen the Deterioration-proof Capability of the Boiler Feed water Pump at

Hitachi, Ltd.

No. Phenomenon that Occurred Cause Measures for Improvement

Cavitation erosion Increased operation time in the low flow rate range due to increased opportunity of intermediate-load-range operation such as DSS and WSS

• Padding of a corrosion-resistant metal on the first-stage impeller

• Improvement of the shape of the inlet channel and the first-stage impeller

Increased vibration Unbalanced force vibration due to locking of the tooth flank of the gear coupling caused by a sudden change in the load

• Adoption of a diaphragm coupling with better flexibility

Increased vibration Increased operation time outside the designed flow rate due to increased opportunity of intermediate-load-range operation such as DSS and WSS

• Adoption of rotors with high rigidity

Damage to the shaft Fatigue started from corrosion pitting caused by deterioration of feed water quality due to leakage of seawater or a certain other reason.

• Detailed inspection of the shaft and removal of corrosion pits

Occurrence of self-excited vibration

Increased clearance due to aged deterioration

• Adoption of a vibration-damping-type balancing drum

Corrosion damage to the chrome plating

Deterioration of the corrosion-proof environment due to increase in DO in the feed water caused by CWT operation

• Replacing the current one with the new one to which improved corrosion-proof chrome plating is applied

• Improvement by adoption of a material not necessitating chrome plating (adoption of a shaft without chrome plating/a shaft seal part without chrome plating)

Damage to the part where stress was concentrated

Too much stress due to thermal deformation of the casing repeatedly working on it when the turbine is started and stopped

• Replacing the current casing with a new one whose deterioration-proof capability is strengthened

• Changing the control method to temperature control from the seal water control to prevent too much cold water from flowing in

Erosion of the inner surface of the discharging nozzle

Erosion corrosion caused by aged deterioration of the material and developed under an environment where there is steam with high-speed flow

• Applying padding of austenite stainless steel after effecting welding of carbon steel for the repair

301

Table 3.4.3-9: Measures Taken to Strengthen the Deterioration-proof Capability of the Circulating Water Pump at Hitachi, Ltd.

No. Phenomenon that Occurred Cause Measures for Improvement

Reduction in wall thickness of the loose flange due to corrosion

Crevice corrosion due to seawater Aged deterioration of the material

• Adoption of a corrosion-proof technique to coat the inside surface in contact with water with a ripoxy lining

Corrosion of the mating surface of the intermediate coupling


• Adoption of a technique to prevent seawater from penetrating by means of epoxy resin putty filled in the joining surface

A malfunction of feeding of the seal water Abrasion of the shaft-sealing device

Clogging due to adhering of marine creatures and accumulation of adult creatures in the feed water piping and the feed water channel in the pump

• Adoption of seal-waterless bearings

Corrosion of the flange surfaces of the bearing bracket and the column pipe


• Adoption of a corrosion-proof technique to coat the inside surface in contact with water and the flange surface with a ripoxy lining.

Corrosion of the bolts, the nuts, and some others


• Strengthening corrosion-proof capability by applying a crevice corrosion inhibitor (RFC)

Reduction in wall thickness due to corrosion of such parts as the casing, the column pipe, etc.


• Effecting a periodic assessment of corrosion by means of an ultrasonic test

Leakage of oil in the impeller boss (only for the circulating water pump for the rotating vanes)

Crevice corrosion of the part of the variable pitch vanes due to seawater

• Neutralizing the possibility of oil leakage by adopting oil-less bearings for the variable pitch control mechanism for the vane to eliminate the necessity of boss oil.

302

4. Assessment Techniques for Equipment 4.1 Assessment Techniques at Mitsubishi Heavy Industries, Ltd.

The following are explanations of the assessment techniques for the area of components that are possibly subject to fatigue taking remaining life assessment of the outer casing as an example to prevent escalation to secondary damage to other equipment than the boiler feed water pump in the event of failure. (1) Selection of objects of assessment and assessment methods

They had an experience where cracking occurred to the suction-side gasket seat of the outer casing (Figure 3.4.3-22 ) of a turbine unit whose operation had started about 15 years ago when the failure occurred; it was repaired by welding, and the outer casing was replaced. Welding of stainless padding was carried out on the part under discussion to strengthen corrosion and erosion resistance. That part was subjected to stress fluctuation according to starting and stopping of the unit. It was sealed against differential pressure between the suction part and the discharging part by a gasket, and the thrust force from the inner casing due to differential pressure worked on that part and caused stress concentration on the corner.

Figure 3.4.3-22: A Structural Drawing of the Feed water Pump of the Mitsubishi MDG 267 Boiler

This life assessment method(Figure 3.4.3-23) is an analytical method, and they are studying the replacement of

the component when cracking due to fatigue has occurred, development afterwards, and when the maximum allowable depth of a crack has been reached.

Cracking due to fatigue

Stre

ss S

According to the S-N curves for carbon steel and stainless steel in Notification 501 Number of Occurrences

of Cracking N Life assessment

The max. allowable depth

Development of a crack

Dep

th o

f a C

rack

FEM Stress analysis Max. allowable crack

depth according to the crack development curve for stainless steel Number of Starts

and Stops N(Cracking in NDI) Remaining Life

Predicted future operation mode

Remaining life assessment

Figure 3.4.3-23: Remaining Life Assessment Method

Even if such a crack is found in the periodic inspection and repaired by means of welding, its deterioration resistance capability would not be high enough and cracking would recur and continue its development. For safe operation of a plant, it is recommended as the measure to strengthen the deterioration-proof capability of the pump that “occurrence of cracking” should be deemed the end of its life. (2) Results of remaining life assessment

The results of this analytical remaining life assessment were analyzed by means of an FEM model. Table 3.4.3-10 shows the results of the assessment in an actual plant. The part under discussion was assessed, and it was found that it had more than 30 years of remaining life, also taking into consideration the fact that no cracks had been found in a non-destructive inspection that had been separately carried out for a similar part of the unit.

Table 3.4.3-10: Results of the Life Assessment

303

Part Name Name of Portion Consumed Fatigue Life Remaining Life

1. The corner area (the suction nozzle side) 4.0% More than

30 years

2. The corner area (in the horizontal direction)

2.5% More than 30 years

The seating surface for the outer casing suction-side gasket

3. The corner area (on the side opposite to the suction nozzle)

2.4% More than 30 years

The suction nozzle

The outer casing

The suction nozzle

Welding

The outer casing

304

3.4.4 Corrosion of Steam Turbines and its Countermeasures The conditions inside a steam turbine continuously change from high temperature and high pressure to a

pressure lower than atmospheric pressure and low temperature depending on the conditions of steam. For a recent typical power plant, the steam temperature at the inlet of the turbine is about 540°C to 570°C, the

pressure is about 250 kgf/cm2 and the exhaust gas temperature is about 30°C. The change in steam conditions agrees with improvement of the materials used for each part of a turbine and

of the turbine structure. Development of heat resistant materials having great strength at high temperature is indispensable to achieve high efficiency in power generation.

The design of a steam turbine is usually achieved for the high temperature section and the low temperature section separately. Table 3.4.4-1 shows typical materials used in each section of a steam turbine. Low alloy steel containing a low percentage of chromium and molybdenum is used for the rotor, the casing, the steam valve and the tightening bolt all of which are used in the high and medium pressure section. And low alloy steel that has high tensile strength such as 3.5NiCrMoV steel is used for the low pressure rotor. In this material, chromium contributes to the oxidation resistance and resistance to the graphitization, and molybdenum contributes to the high temperature strength. The 12Cr heat resistant steel is used for the rotor, the blade, the nozzle and the bolt all of which are used in the high and medium pressure section. For the material for the rotor whose temperature increases to about 580°C, 12Cr steel reinforced with molybdenum and vanadium is used with tantalum and niobium or nitrogen added and with fine carbide and nitride precipitated. For use in the higher temperature, strength at the high temperature is improved by the addition of tungsten or molybdenum.

Table 3.4.4-1: Material used in main parts of the thermal turbine Parts Representative steel types

The rotor 1Cr-1Mo-1/4V steel 12Cr-Mo-V-Ta-N steel 12Cr-Mo-V-Nb-N steel 12Cr-Mo-V-W-Nb-N steel

The blade The nozzle

12Cr-Mo-V-W steel 12Cr-Mo-V-Nb-N steel 12Cr steel Ni-based superalloy

The casing The steam valve

1Cr-1Mo-V steel 1Cr-0.5~1Mo steel 11/4~21/4Cr-0.5~1Mo steel 12Cr steel

Hig

h an

d M

iddl

e pr

essu

re tu

rbin

e

The tightening bolt 1Cr-1Mo-V steel 12Cr-Mo-V-W steel 12Cr-Mo-V- Nb-N steel Ni-based superalloy

The rotor 3~3.5Ni-Cr-Mo-V steel The blade The nozzle

12Cr steel 12Cr-Mo-V steel 12Cr-Ni-Mo-V-N steel 17-4PH Titanium

Low

pre

ssur

e tu

rbin

e

The casing Carbon steel 3.4.4.1 Corrosion in the High Temperature Zone and its Countermeasures

For corrosion in the high temperature zone exceeding 400°C, it is usually only a question of an oxidation reaction of the target substance with a gas phase substance, and there is no involvement of a liquid phase. In a thermal power plant, especially for the boiler, this countermeasures for the high temperature corrosion is an important issue. This phenomenon is related to the decrease of thickness in many cases, therefore, the selection of materials and setting of corrosion control such as coating are decided from this perspective.

As for the materials for the components of a turbine, problems directly related to the high temperature corrosion have not often occurred. However, issues to be considered such as sticking of the major valves like the main steam stop valve caused by products of corrosion deposited on the valves and erosion occurring due to the flowing in of oxide particles still remain.

High chromium steel such as 12Cr steel is effective to improve corrosion resistance. And hardening of a material by applying nitride treatment to the material surface and cladding welding of stelite that is a cobalt base alloy are effective measures to strengthen corrosion resistance too.

305

Erosion is a phenomenon where decrease of thickness occurs due to high speed fluid and the reduction is accelerated when solid substances included in the fluid abrade the wall. For example, we experienced a phenomenon whereby the nozzle plate in the control stage at the turbine inlet is eroded by oxide particles. This phenomenon mainly occurs at the start-up of a turbine, damaging the end part of the steam outlet of the nozzle finally resulting in lowered turbine efficiency. Boronizaion treatment is one measure to prevent this erosion from occurring. This treatment diffuses boron(B) applying its vapor deposition to the surface of the material to harden it.

Figure 3.4.4-1(2) shows the effect of the treatment. This is the data of effectiveness of the boronization treatment applied to an actual turbine unit. We obtained the result whereby the nozzle plate with a thickness of 80 µm of the treated layer could realize such durability that almost no erosion occurred in an operation even exceeding 15000 hours. And also another measure is one in which the shape of the nozzle is structurally changed to enhance erosion resistance.

With no treatment

Eros

ion

Rat

e (%

)

With boronization treatment(thickness of the surface layer: 40 µm)

With boronization treatment (thickness of the surface layer: 80µm)

Time (h)

Figure 3.4.4-1: Erosion Resistance of the Nozzle Plate with Boronization Treatment 3.4.4.2 Corrosion in a Low Temperature Zone and its Countermeasures

Corrosion in a turbine usually means corrosion under a wet steam environment at 200°C or lower. Corrosion in this temperature zone includes stress corrosion cracking (SCC), corrosion fatigue, and erosion corrosion. The SCC and the corrosion fatigue are generally considered to be caused by condensed water in the space between turbine components where impurities such as Cl- and SO4

2- contained in steam are dissolved and condensed. In fact, these types of corrosion often occur around the boundary zone between dry steam and wet steam in a low pressure turbine.

The SCC is cracking that occurs in a material when corrosion happens under a static tensile stress. This cracking suddenly causes a brittle fracture, therefore, it presents a serious problem for a structural component.

On the other hand, corrosion fatigue is a rupture of a material when an alternate stress is imposed on a material, being caused by a phenomenon whereby the fatigue strength is lowered in a corrosive environment. For a low pressure turbine, it occurs at the base of the blade and in the tenon in some cases.

And in the wet steam zone, erosion occurs in some cases. The erosion is caused in a process such that a water film formed on the surface of the stationary blade is carried away by steam flow to become droplets and they crash onto the rearmost rotating blades. Stelite welding or some other measures are applied to the rearmost rotating blades to prevent this from happening. 3.4.4.2.1 SCC Sensitivity of a Material Used in a Thermal Power Plant

Here, some recent studies on the influence of impurities’ ions and dissolved oxygen on the SCC of materials used in a thermal power plant are introduced.

The influence of impurities’ ions on SCC under an AVT (Volatile Matter Treatment) environment is being studied. The test conditions were 90°C and pH 9.5 with deairing (7 ppb or lower O2) achieved. Figure 3.4.4-2(11) shows the results of investigation on the influence of Cl- concentration on the depth of the maximum crack that occurred in a constant strain SCC test.

306

3.5NiCrMoV steel12Cr steel 17-4PH steel

Max

. cra

ck d

epth

(µm

)

Cl- Concentration (ppm)

Figure 3.4.4-2: Influence of Cl- Concentration on Behavior of Various Types of Steel Related to SCC

For the 3.5 NiCrMoV steel, pitting occurs at 100 ppm of a Cl- concentration and SCC with 1000 ppm of a Cl- concentration. While, for 12Cr steel and 17-4PH steel, no SCC occurs when Cl- is increased up to 1000 ppm. SO4

2- has so little acceleration function compared with Cl- that no SCC occurs when the Cl- concentration is increased up to 1000 ppm though pitting occurs at the same Cl- concentration. And for Na+, neither pitting nor SCC occurs up to 1000 ppm of Cl- concentration. However, the SCC occurs both in 12Cr steel and 17-4PH steel where the concentration is as high as 10% at a temperature of 200°C or higher.

Thus, we show the influence of dissolved oxygen. Figure 3.4.4-3(12) shows the situation in the case of 3.5NiCrMoV steel. This test was conducted in conditions such as a temperature of 90°C, a Cl- concentration of 10 ppm and range of dissolved oxygen concentration between 7 ppb or less and 1.63 ppm. The test results show that the SCC occurs when the dissolved oxygen concentration exceeds 10 ppb and 10 ppm Cl- concentration. Our results showed that the SCC sensitivity to the dissolved oxygen has the same tendency as that of the low alloy steel as shown in Fig. 3.4.4-4(12) in the case of 12Cr steel used in the material for blade, but the speed of crack development is slow. And the SCC resistance of the 17-4PH steel is more superior than the two types of steel mentioned above as shown in Fig. 3.4.4-5(12).

1000 hour test1500 hour test3000 hour test5000 hour test

5000 hour test (320 µm on average)

Max

. cra

ck d

epth

(µm

)

3000 hour test

1500 hour test

Dissolved oxygen concentration (ppm)

Figure 3.4.4-3: Influence of Dissolved Oxygen Concentration on Behavior of 3.5NiCrMoV Steel Related to SCC

1000 hour test 1500 hour test 3000 hour test 5000 hour test

Max

. cra

ck d

epth

(µm

)

3000 hour test 5000 hour test(75 µm on average)

1500 hour test


Figure 3.4.4-4: Influence of Dissolved Oxygen Concentration on Behavior of 12Cr Steel Related to SCC

307

1000 hour test 1500 hour test 3000 hour test 5000 hour test

Max

. cra

ck d

epth

(µm

)

3000 hour test 5000 hour test (19.5 µm on average)

1500 hour test


Figure 3.4.4-5: Influence of Dissolved Oxygen Concentration on Behavior of 17-4PH Steel Related to SCC 3.4.4.2.2 Corrosion Fatigue

Generally, the stronger the corrosiveness of the environment, the more the corrosion fatigue strength of a material is reduced. For example, the corrosion fatigue strength is reduced with the existence of NaCl, the corrosiveness becomes more severe in conditions of a smaller pH number and increasing concentration of the dissolved oxygen(13). As shown in Fig.3.4.4-6(14), there is also a report asserting that the number of cycles at which a material is broken in the NaCl solution is influenced by temperature and the life is shortest around 150°C.

Num

ber o

f cyc

les (

N)

Temperature (°C)

Figure 3.4.4-6: Influence of Temperature on Corrosion Fatigue in NaCl Aqueous Solution

Recently, the method of feedwater treatment in thermal power plants is being changed to CWT (treatment by oxygen) from conventional AVT. Some of the power plants have adopted the CWT. Influence of this change in water treatment to the material of turbine is being investigated and no significant difference in the fatigue strength of 3.5NiCrMoV steel and 12Cr steel has been found in any conditions of AVT (pH9.5, 7 ppb O2) and CWT (pH8.0, [I] 50 ppb O2 and [II] 200 ppb O2) according to Fig. 3.4.4-7(15).

3.5%NiCrMoV steel 12%Cr steel

Stre

ss a

mpl

itude

(kg/

mm

2 )

Stre

ss a

mpl

itude

(kg/

mm

2 )

Repeated number at which the material is broken Repeated number at which the material is broken

Figure 3.4.4-7: Influence of Water Treatment Conditions on Corrosion Fatigue Strength 3.4.4.3 Properties of Steam

All the types of corrosion damage explained above are greatly influenced by impurities contained in the steam. Therefore, it has become a very important issue to completely grasp the properties of the steam quantitatively.

308

3.4.5 Corrosion of Heat Exchangers and Piping of Turbine Systems and its Countermeasures 3.4.5.1 Introduction

The system configuration of a thermal power plant using steam turbines has become more complicated due to improvement in the steam condition and increase in the capacity of a single unit. Figure 3.4.5-1 shows as an example the main system diagram(1) of a power plant having steam turbines of the 1000 MW class. The main facilities of the power plant are classified as follows.

(1) The turbines and generators (2) The water treatment unit for water for the condenser and feedwater (3) The condenser unit (4) The feedwater heater unit (5) The feedwater pump (6) The piping unit for each component system

Approximately 3000 tons/h

The generator

The high pressure turbine

The medium pressure turbine

The low pressure turbine (B)

The

boile

r The low pressure turbine (A)

The condenser

The make-up water

The condenser pump The turbine

The demineralizer for the condenser

Approximately 280°C Gland steam

condenser

The condensate booster pump

The

deae

rato

r

The feedwater heater

The drain pump The boiler feedwater pump

The feedwater booster pump

Figure 3.4.5-1: Main System Diagram(1) of a Power Plant Having Steam Turbines in the 1000 MW Class

Among these main facilities, equipment and components, of which corrosion is one of the main concerns, are the low pressure turbines, the condenser in the condensing and water feeding system, the deaerator, the feedwater heater and the piping, all of which are placed under wet conditions during operation of the turbine, is well as other components and piping that utilize seawater for cooling fluid, all of which handle or use water. In this chapter, we discuss corrosion occurring in the heat exchanger and the piping for the turbine system and also its countermeasures. 3.4.5.2 Examples of Corrosion Occurring in the Heat Exchangers and the Piping used in the

Turbine System and its Countermeasures 1 The Heat Exchangers Used in a Power Plant Using Steam Turbines 1.1 The Condenser

Figure 3.4.5-2 shows types of corrosion occurring in the condenser cooling pipes and protective measures against it(2).

309

The outside of a pipe

Corrosion of and leakage from a

copper alloy pipe

The inside of a pipe

Type of corrosion and leakage Protective measures

Inlet attack

Deposit attack Formation of a protective film by the injection of ionized iron

A chlorine injection device

A shell remover A counter-flow washing device

Improvement of the structure

Change of the material of the pipes to titanium types

Erosion by drain

Ammonium attack

Inspection of the water intake gate

A ball purge device

Corrosion by polluted seawater

Abnormal impingement damage

Sand erosion

An electrochemical protection device

Type of corrosion and leakage Protective measures

Galvanic corrosion

Hydrogen absorption Reinforcing the water chamber for

which electrochemical corrosion protection is not provided by the

addition of a rubber lining

Full-titanium pipes

Making the electric potential of the electrochemical corrosion protection device appropriate

(-0.45 ~ -0.6V (SCE))

Applying epoxy coating on the pipe plate

Copper alloy pipes + titanium pipes

Copper alloy pipes + titanium pipes

Improvement of the structure Erosion by drain The outside of a pipe

Corrosion of/ leakage from

a titanium pipe

The inside of a pipe

Figure 3.4.5-2: Corrosion and Leakage of Condenser Cooling Tube and Its Countermeasures

Here in this chapter, we introduce the types of corrosion occurring in the cooling pipes facing the outside (steam side) and the protective measures. (1) Ammonium attack of the copper alloy pipes.

Part of the ammonium that is used as a feedwater treatment agent is discharged in the form of non-condensed gas to the outside of the circulating system through the condenser cooler unit by the air extraction device. The air cooler and the surrounding area are always exposed to ammonium concentrated in the condensed water and especially the pipes near the supporting plate along which the condensed water drips down suffer from corrosion.

Photo 3.4.5-1 shows an example of the above mentioned corrosion. The countermeasure is replacement of the existing pipes by titanium types that have superior corrosion resistance.

Photo 3.4.5-1: An Example of the Ammonium Attack of an Aluminum-brass Pipe

(2) Erosion by Drain (Droplet erosion)

In some cases, droplets accelerated by steam that is discharged from the low pressure turbine collide with a pipe located outside of the pipe arrangement and erode its outer surface making it resemble a matte finished type(5).

Figure 3.4.5-3, Photo 3.4.5-2 and Photo 3.4.5-3 for example show the position where the erosion occurs and the condition of an eroded aluminum-brass pipe and of an eroded titanium pipe, respectively. A protective element and protective pipes are installed for the countermeasures as shown in Figs. 3.4.5-4 and - 5.

310

The low pressure turbine

The connecting body The low pressure feedwater heater The connecting

body

A group of the cooling pipes

Erod

ed p

ortio

n

Eroded portion

Eroded portion

Eroded portion

Figure 3.4.5-3: Erosion of the Outer Surface of a Condenser Cooling Pipe (at a place where erosion occurs)

Conditions of the outer surface of a cooling pipe

The outer surface

The inner surface

Photo 3.4.5-2: Erosion of an Aluminum-brass Cooling Pipe

Conditions of the outer surface of a cooling pipe

The outer surface

The inner surface

Photo 3.4.5-3: Erosion of a Titanium Cooling Pipe

311

The cross section of the protection element

The supporting plate

The cooling pipe protection element

A bundle of pipes The supporting plate

Figure 3.4.5-4: Prevention of Droplet Erosion by Installation of Protection Elements


A bundle of pipes The protection pipes


Figure 3.4.5-5: Prevention of Droplet Erosion by Installation of Pipes to Protect the Peripheral Portion of the Bundle of Pipes

1.2 Deaerator

In some cases, the inner surface of the deairing chamber body is partly damaged in a plant where the pH of the boiler feedwater is smaller than 9.0 and the water includes a relatively large amount of dissolved oxygen. The area receiving the damage is limited to the area on which the feed water drops from the deairing tray or against which the dropping water is blown by the influence of heating steam. The cause of the damage is erosion corrosion caused by the feedwater that drops from the deairing tray, is accelerated by heating steam entering from the bottom center part of the tray which directly collides with the deairing chamber body. And when there is a certain distance between the deairing tray and the body wall, erosion corrosion is sometimes caused by free fall of the feedwater regardless of the existence of the heating steam.

Figure 3.4.5-6 shows an example of the damage. The protective plates made of stainless steel having strong erosion corrosion resistance are attached to the inner surface of the deairing chamber body as shown in Fig. 3.4.5-7 to prevent erosion corrosion damage of the deairing chamber body.

Deaerating chamber

The tray Heating steam

Erosion

Figure 3.4.5-6: Example of Corrosion Inside the Main Body of the Deaerating Chamber

312

The deairing chamber

The tray Heating steam

Protection plate (SUS304)

Figure 3.4.5-7: An Actual Measure Against Erosion of the Inner Surface of the Deairing Chamber Body 1.3 The Oil Cooler

In most cases, corrosion of the oil cooler actually experienced occurs on the cooling water side of the copper alloy cooling pipe. Causes of the corrosion are mainly (i) pitting and (ii) stress corrosion cracking. (i) Pitting

When pitting occurs, it occurs almost evenly along the entire length of the cooling pipe, and the speed of progression is of such an extent that the pitting pierces the wall with a thickness of approximately 1.2 mm in about 6 months in the fastest case. There is a special tendency whereby pitting is liable to occur more in the cooling pipes of a spare unit fully filled with water than in the pipes through which water is running.

It is considered that the following combination of factors is the cause of the pitting. 1) The corrosion inhibitor added in the cooling water is not effective in corrosion prevention, or the

concentration of the corrosion inhibitor is insufficient. 2) A corrosion factor exists in the cooling water. 3) Local electric potential difference or difference in the concentration of dissolved oxygen is liable to

occur because the liquid in the pipe has almost no velocity when the operation is suspended (in a stand-by situation).

4) Dissolved oxygen is consumed by microbes to make the atmosphere anaerobic, and the pipe is corroded by ionized sulfur produced due to the propagation of sulfate salt reduction bacteria.

(ii) Stress corrosion cracking

On the other hand, there is the fact that a stress corrosion crack starts from a pit and occurs within a limited area approximately 500 mm apart from the plane including the pipe plate in most cases. This crack advances towards the outside surface of a pipe from the inside. Photo 3.4.5-4 shows an actual stress corrosion cracking starting from a pit occurring in an aluminum-brass pipe that is dipped in the water (a sectional photo).

Photo 3.4.5-4: Stress Corrosion Cracking in an Aluminum-brass Pipe Dipped in the Cooling Water.

This is considered to be caused by the following two factors combined.

1) Existence of a corrosive catalyzer in the cooling water (ionized ammonium, ionized sulfur, etc.) 2) Existence of residual stress (those in the raw material itself, generated during the operation or

generated during assembling process) Prevention of pitting is possible especially for the oil cooler whose operation is suspended (in a stand-by

situation) by providing a fluid speed fast enough (faster than 0.3 m/s) to equalize the difference in the local concentration of ions that may form a corrosion battery. However, since it is not possible to nullify a corrosive catalyzer, it is necessary to select and add an appropriate corrosion inhibitor from a practical view point.

As for measures against stress corrosion cracking, firstly, add the most appropriate corrosion inhibitor to suppress pitting, the starting point of cracking. Secondly, reduce the residual stress in the cooling pipe as much as possible. For this purpose, improvement of the processing method such as expansion of the cooling pipe and assembling is important.

313

2 Piping for the Turbine Plant 2.1 The Steam Pipes

The damage to a main steam pipe or a hot reheat steam pipe is creep damage caused by the effect of internal pressure during the steady operation and erosion and low cycle fatigue damage due to repeated thermal stress caused by load fluctuation due to start and stop or some other factors.

Places where wall thickness reduction is liable to occur are the main stop valve (MSV), the governing valve (CV), the lower outlet of the drain valves located before and after the valve seat of the combined stop and control reheat valve (CRV), the elbows, the caps and the orifices. Leakage due to reduction in the wall thickness was experienced in these places and components. Figs. 3.4.5-8 and 3.4.5-9 show the location of the damage and the actual damage to the drain pipe elbows before and after the valve seat of the CRV, and the actual damage to the main steam lead pipe warming orifice, respectively.

The primary valve FlowThe secondary valve

Reduction in wal ickness l th (The drain pipe before and after the valve seat of the MSV valve)

Figure 3.4.5-8: An Example of Damage to a Drain Pipe Elbow Before and After the Valve Seat of the MSV Valve

From the lead pipeThe main steam lead pipe

To th

e hi

gh p

ress

ure

casi

ng

Erosion

To the condenser Oxidized

powder scale

Erosion

To the condenserLeakage

Figure 3.4.5-9: An Example of Damage to the Main Steam Lead Pipe Warming Orifice

These types of erosion are caused by the collision of the drain jet and flowing- in of the oxidized scale existing in the main pipe. The countermeasures are periodic measurement of the wall thickness of the places in the system where reduction in wall thickness is liable to occur and replacement of those components for which reduction in wall thickness has been advanced.

And the orifice used to be replaced by one that had a shape more favorable in terms of erosion resistance. 2.2 Pipes of the Condenser and the Feedwater Systems

Damage to the pipes of the condenser and the feedwater systems is reduction in the wall thickness due to erosion corrosion.

Generally speaking, erosion corrosion is a phenomenon of reduction in wall thickness that is caused by the interaction of erosion, a mechanical action and corrosion, a chemical action.

Conditions of the fluid (temperature, wetness fraction, pH, dissolved oxygen), flow speed, material properties and shape of the part are considered to be factors contributing to erosion corrosion.

Figures 3.4.5-10 to 13 show the influence of fluid speed, wetness fraction, content of alloy elements and temperature on reduction in weight due to erosion corrosion, respectively.

Red

uctio

n in

wei

ght d

ue to

ero

sion

co

rros

ion

(mg/

cm2 )/1

4 da

ys

Temperature: 158°C Wetness fraction: 11% O2 concentration: 16 ppm Carbon

steel

Ni-Cr-Cu steelCr-Mo steel

Steam speed (m/s)

Figure 3.4.5-10: Influence of Steam Speed on Erosion Corrosion

314

Red

uctio

n in

wei

ght d

ue to

ero

sion

co

rros

ion

(mg/

cm2 )/1

0 da

ys

Temperature: 171°C Velocity: 200 m/s O2 concentration: 16 ppm Carbon steel

Ni-Cr-Cu steelCr-Mo steel

Wetness fraction of steam (%)

Figure 3.4.5-11: Influence of Wetness Fraction of Steam on Erosion Corrosion

Eros

ion

vs. c

orro

sion

ad

vanc

ing

spee

d ra

tio

Temperature: 150°C Velocity: 90 m/s Wetness fraction: 11%

Content of alloy elements (weight %)

Figure 3.4.5-12: Relation Between Erosion vs. Corrosion Advancing Speed Ratio and Content of Alloy Elements

: Carbon steel : Cr-Mo steel

Red

uctio

n in

wei

ght d

ue to

ero

sion

co

rros

ion

(mg/

cm2 )

Velocity: 90 m/s Wetness fraction: 11% O2 concentration: 16 ppm

Figure 3.4.5-13: Influence of Steam Temperature on Erosion Corrosion

Steam Temperature (°C)

In the case of carbon steel, reduction in weight increases in proportion to the square root of the steam speed,

and rapidly increases with wetness fraction up to approximately 7%. But the increase rate in the reduction becomes very slow even when the wetness fraction increases. And the reduction in wall thickness tends to decrease with the increase in the amount of alloy elements in the material. In the case of a low alloy steel, the reduction in wall thickness is not much affected by the wetness fraction or steam speed. As for the influence of temperature, there is a tendency for the amount of reduction in wall thickness to become its maximum in a certain range of temperature (around 180°C) and to decrease outside this temperature zone either above or lower. This phenomenon happens because the major dominant factor is the speed of corrosion action in the temperature zone lower than the temperature at which the speed of reduction in wall thickness reaches its peak while a protective film of magnetite (Fe3O4) is formed on the surface of the metal in the higher temperature zone. Difference in the quality of water need not be considered taking into account the difference in speed of the reduction in wall thickness occurring in each part because the quality of water is controlled according to the requirement of standard values separately set for the feedwater heater and the boiler. Therefore, the parameters that cause the difference in speed of reduction in wall thickness are the temperature and the shape of the channel for fluid. The places and portions where the reduction in wall thickness and leakage were experienced are bends and branches where the water temperature was between 150°C and 250°C, the outlet of the booster pump for the boiler feedwater pump (BFP), the curved pipe at the outlet of the BFP, the outlet of the adjusting valve and the outlet of the orifice.

Photos 3.4.5-5 and 3.4.5-6 show an actual case of reduction in wall thickness occurring in the outlet of the booster pump for the boiler feedwater pump (BFP) and also another actual case of that occurring in the curved pipe at the outlet of the BFP, respectively. Erosion corrosion is prevented by means of periodic measurement of the components and the portions where the reduction in wall thickness is liable to occur and replacement of the component by a corrosion resistant material such as a CrMo steel pipe or a stainless steel pipe if the reduction in wall thickness has already progressed.

315

The BFP connecting pipe

Photo 3.4.5-5: An Example of Reduction in Wall Thickness of the BFP Booster Pump

The high pressure feedwater pipe

Photo 3.4.5-6: An Example of Reduction in Wall Thickness of a Curved Pipe at the Outlet of the Boiler Feedwater Pump

2.3 Heater Drain Piping

In many cases, damage to the piping in this system is caused by erosion corrosion and the damaged area has been limited to the lower side of the adjusting valve. It means that conditions of the drain in the drain piping in terms of pressure and temperature are very near to the saturated temperature and the saturated pressure of the upper side of the adjusting valve becomes lower than the steam pressure in the lower side of the adjusting valve due to pressure reduction caused by the throttling effect of the adjusting valve. This creates a situation in the pipe whereby it is flushed and bubbles are generated in the drain so that flow in the pipe becomes a gas liquid two phase flow. As a result, reduction in wall thickness occurs in the lower side of the adjusting valve due to turbulence in the two-phase flow and high speed flow of the drain. Many of the old power plants adopted the piping design around the adjusting valve as shown in Fig. 3.4.5-14 and experienced a reduction in wall thickness occurring in a short pipe in the lower side of the adjusting valve. Erosion corrosion is prevented by means of periodic measurement of the components and the portions where the reduction in wall thickness is liable to occur and replacement of the component by a corrosion resistant material such as a CrMo steel pipe or a stainless steel pipe if the reduction in wall thickness has already progressed.

The short pipe

The adjusting valve The reducer

The flow direction

Figure 3.4.5-14: The Typical Piping Form Around the Heater Drain Pipe Adjusting Valve 2.4 Control System of Reduction in Wall Thickness in the Piping

Until recently, wall thickness was periodically measured to check the reduction in wall thickness and experienced engineers assessed and controlled the reduction in wall thickness based on extensive measurement data on the reduction in wall thickness. However, such manual work on the numerous numerical data has reached its limitation in terms of efficiency and accuracy. Therefore, a control system of reduction in wall thickness by means of automation and computerization of the measurement has been developed.

316

3.4.6 Corrosion in the Equipment of the Seawater System and the Piping and its Countermeasures 3.4.6.1 Preface

Seawater is considered to be a typical substance that produces a strong corrosive environment since it contains large amount of strong corrosive chloride ions and has high electric conductivity. Therefore, the equipment and the piping used in a seawater system and their components often suffer from noticeable corrosion damage unless they are used based on specific technologies of each component and material. This report explains these situations of damage, their features, and countermeasures. 3.4.6.2 Corrosivity in a Seawater Environment

Generally, the temperature, pH, and concentration of dissolved oxygen of clean natural seawater are 20˚C or colder, approximately 8, and 5 to 10 ppm, respectively. They are approximately the same level as those of freshwater.

However, the fact that it includes a concentration of 19,000 ppm of chloride ions, which corrodes metal surfaces, and its electrical conductivity is 27 Ω⋅cm (both are 100 times or more those of freshwater) has important meaning.

In other words, since a layer of rust formed on the material surface becomes porous in the presence of a high concentration of chloride ions, the corrosion rate of carbon steel, cast iron, and niresist cast iron (austenite Ni cast iron) in seawater increase approximately linearly (3) when the flow speed increases to 5 to 15 m/s although the gradients of the speed increase are different from each other. This is decisively different from the behavior of carbon steel in freshwater where the corrosion rate rapidly decreases due to a delicate film forming with a flow speed of 1 m/s or faster (2)(4). This is one of the strongest reasons why carbon steel cannot be used for seawater hydraulic machines and equipment without any surface treatment. However, since some niresist cast iron (type III) and copper alloys (BC2, BC6, AlBC2, etc.) are low in corrosion rate as well as in flow speed dependency, these materials show superior corrosion resistance unless they are used in high-speed flow or in polluted seawater (3).

The next fact is that highly concentrated chloride ions electrochemically destroy a passive state film. And this makes the solution in this area acidic by pitting, hydrolysis of metallic ions dissolved inside the structural crevice, and concentration of chloride ions. As a result, active dissolution occurs to a metal used as a raw material like stainless steel and copper alloys and continuously advances so that various kinds of local corrosion such as pitting, crevice corrosion, selective corrosion, intergranular corrosion, stress corrosion crackling, hydrogen embrittlement, etc. are caused.

Figure 3.4.6-1 shows an example (5) of the relationship between the corrosion of stainless steel and the environmental conditions. It may be understood that pitting and stress corrosion cracking that do not easily occur in freshwater (the concentration of chloride ions is 100 ppm or lower) can occur in low-temperature zones in solution of richly concentrated chloride ions carried by seawater. Concentration of C1- of

seawater equivalent

Tem

pera

ture

(°C

)

Concentration of C1- (ppm)

O: No corrosion S: rusted P: Pitting C: Stress corrosion crackling

Figure 3.4.6-1: Each corrosion damages on SUS304 stainless steel in solution of neutral chloride

317

The next fact is that high electric conductivity of seawater has an important influence on the occurrence and progression of corrosion damage. In other words, corroding action caused by an electrochemical mutual action called bimetallic corrosion (also called galvanic corrosion) occurs when component and materials that have different corrosion potential from each other (for example, a noble metal Cu and a base metal Zn) electrically make contact and are immersed in a solution. This phenomenon is corrosion damage that a base metal dissolves at first (as a sacrifice) against noble metal (anodic dissolution). In this case, higher electric conductivity means a longer distance over which an electrochemical mutual action can occur. In seawater, where the affected area is simply calculated as the square of 100 times that in freshwater, the speed of corrosion damage is accelerated according to the increase in the affected area.

Therefore, special care is necessary to prevent this galvanic action (selection of materials and design of parts configuration) in order to decrease corrosion damage to hydraulic machines and some others that consist of various components and materials (1).

In certain cases, the corrosivity feature of seawater environment works moderately to the same degree as freshwater does depending on the material used and the environmental conditions. On the other hand, cautious observation is required because it is not unusual for a new kind of corrosion phenomenon to appear, greatly increasing the speed of corrosion propagation and often leading to severe corrosion damage when the environmental conditions change or a particular material is used(6)(7). 3.4.6.3 Corrosion Damage to Various Parts and its Countermeasures 1 Parts Made of Stainless Steel

Figure 3.4.6-2 shows an example of a configuration and materials of parts of a seawater pump(8).

The casing (Carbon steel coated heavy corrosion control) rubber bearing

Level of seawater

shaft casing (SUS316) shaft (SUS316)

shaft sleeve (17-4PH)

rubber bearing bell casing (SCS14) runner ring (17-4PH) impeller (SUS14)bell mouth (Cast iron coated heavy corrosion control)

Figure 3.4.6-2: Structure of a Vertical Shaft-type Seawater Pump and Examples of Materials Used in the Components

Such local corrosion as indicated below likely occurs in seawater that contains a high concentration of chloride

ions although stainless steel is used for various parts as shown in the figure. We explain the corrosion damage listed below and countermeasures.

(1) Pitting (2) Crevice corrosion (3) Selective corrosion along metallographic structure (4) Intergranular corrosion (5) Stress corrosion cracking and hydrogen embrittlement cracking (6) Cavitation erosion (7) Corrosion fatigue

318

1.1 Pitting Damage and its Countermeasures Figure 3.4.6-3 shows the relationship between the chemical composition of various kinds of stainless steel that

have been immersed in seawater in Japan for three years and the maximum depth of pitting. It is understood that the depth of erosion is very thin and that it takes considerable time for a large-sized hydraulic machine made of a thick plate to be corroded although it depends on the metallographic structure and the alloy composition.

γ typeα type

1 2 3 yr

Two phase type

Max

imum

Pitt

ing

Dep

th (µ

m)

304C (color developed)304HL

Total Research Institute for Marine Technology of Suruga Bay

Figure 3.4.6-3: Relation between the Pitting Index of Various Stainless Steels Dipped in Actual Seawater and the

Maximum Pitting Depth

However, it is necessary to be careful when seawater is polluted. Figure 3.4.6-4 (a) and (b) show these examples. Figure 3.4.6-4 (a) shows that the pitting potential (the critical value of the electric potential to cause pitting) is reduced by some hundreds of mV and makes the material more liable to be corroded when concentration of hydrogen sulfide in seawater increases. In the result, the average corrosion rate under this environment is found to rapidly increase with the increase the concentration of hydrogen sulfide.

Seawater 50°C pH: 5.5 Flow speed: 40 m/s

Aver

age

Cor

rosi

on R

ate

(mm

/y)

Con

cent

ratio

n of

Hyd

roge

n S

ulfid

e (p

pm)

SUS316 Seawater Dissolved oxygen: Open to atmosphere pH: 4.8

Pitting Electric Potential (mV vs SCE) Concentration of Hydrogen Sulfide (ppm)

(a) Relation between Concentration of Hydrogen Sulfide and Fluid Temperature regarding Pitting Electric Potential

(b) Relation between Concentration of Hydrogen Sulfide and Average Corrosion Speed

Figure 3.4.6-4: Corrosion Behavior of Stainless Steel in Polluted Seawater Containing Hydrogen Sulfide

Although natural seawater containing such highly concentrated hydrogen sulfide does not actually exist, it implies that such severe corrosion damage can occur in the event that formation of a local polluting environment due to attachment and decomposition of marine creatures or a micro-polluting environment due to dissolution of non-metallic impurities such as MnS in the steel occur.

For parts with thin wall thickness, it is very necessary to use the high-quality steel indicated above. For the condenser pipes, the two-phase stainless steel SUS329J1 or the recently developed super stainless steel SUS329J4L is used (12)(13).

319

1.2 Crevice Corrosion Damage and its Countermeasures Crevice corrosion occurs not by chemical components in the seawater but also crevices that play an important

role in the starting point of the corrosion. We have experienced crevice corrosion damage from which seawater hydraulic machines have suffered at the

places shown in Figure 3.4.6-5 (14).

Crevice corrosion

Gra

vity

The impeller

Crevice corrosion

Sediment A bolt

Crevice corrosion

A shaft

Marker paint

Pitting

A shaft Zone being washed by the tide

Figure 3.4.6-5: The places of a Structure Made of Stainless Steel Used for a Seawater Pump Where Pitting

Corrosion Occurred

1) For the impellers of pumps with a lot of downtime, dust is accumulated at the lower part by the force of gravity and becomes the starting point of corrosion.

2) For bolts in hollow parts, mats made from marine creatures have accumulated and become the starting point of corrosion as (1) above.

3) When a part number is written on the shaft by a marker, the ink film becomes a crevice and occurs crevice corrosion.

4) For pump shafts of pumps with a lot of downtime, corroded pone occurs at the point of a lot of floating dust floats in the zone of the tide.

In these cases, the degree of corrosion damage is so small as to seldom cause a functional problem even in long-time operation.

Thus, the occurrence of crevice corrosion strongly depends on the structure of the crevice and existence of attached substances.

The next matter to be discussed is the influence of environmental conditions on the crevice corrosion shown in Figure 3.4.6-6. The results indicate that the observed maximum corrosion depth increases according to the increase in the free surface area as well as the DO concentration, and this increase in the amount of cathode action around a crevice decides the speed of anode action.

Max

imum

Cre

vice

Cor

rosi

on D

epth

(m

m)

Concentration of Dissolved Oxygen

Outside Free Surface Area (cm2)

Figure 3.4.6-6: Relation between the Outside Free Surface Area and the Concentration of Dissolved Oxygen Influencing the Maximum Depth of Crevice with crevice corrosion of SUS304 Dipped in Seawater

320

This phenomenon explains that structures where the area of the crevice working as the anode is small but the area of the free surface working as the cathode is large are liable to suffer severe corrosion damage.

The next exhibit is Figure 3.4.6-7 (15) that shows the condition of metal materials related to crevice corrosion. In the same manner as that of Figure 3.4.6-3, it indicates that an alloy element (here, the content of Mo) considerably increases the critical creive corrosion temperature. And the fact that the critical crevice corrosion temperature is lower than that causing pitting by about 40˚C shall be observed.

Crevice corrosion Pitting No crevice corrosion occurred. No pitting occurred.

Pitting occurred.

Crevice corrosion occurred.

Figure 3.4.6-7: Relation between Critical Temperature of pitting and Crevice Corrosion and Mo Content in Various Stainless Steels

Temperature (˚C)

Next, it explains measures to prevent crevice corrosion. As described above, since crevice corrosion damage

is basically relatively light two-dimensional damage and change over time in depth is slow, it is not necessary to be overly concerned. However, in the event that a thin material is used or in the case of a part whose air (water) tightness is critical, it is recommended to see that the structure to prevent dust from adhering, that the device of the structure is designed so as to prevent corrosion damage from concentrating in a limited area, and that the appropriate operation is carried out.

Regarding the selection of materials, Figures 3.4.6-3 and 3.4.6-7 should be referenced and Table 3.4.6-1 can be referenced when there is much freedom of material selection. In other words, the crevice corrosion resistance of stainless steel is ranked as the lowest, thus utilization of niresist cast iron, or copper alloy is effective when full corrosivity can be sacrificed to a certain extent.

Table 3.4.6-1: Crevice corrosivity of various materials to use for seawater structures

Crevice corrosion resistance Major materials No crevice corrosion occurred. Hastelloy C Ti* Inconel 625

Crevice corrosion resistance: Strong 90Cu-10Ni70Cu-30Ni

Bronze Brass

Crevice corrosion resistance: Medium Niresist cast iron Cast iron Carbon steel Crevice corrosion resistance: Weak Inconel 823 Carpenter 20 Monel Copper

Severe crevice corrosion occurred. SUS304 13Cr S/S steel SUS316 Ni-Cr alloy

A paste containing metal zinc grain to prevent crevice corrosion can be available in the market(18). Application of the paste to crevices in structures can prevent crevice solutions from becoming acidic and can then suppress crevice corrosion. Also, welding alloy known as prevention of crevice corrosion ressistance.(19).

321

1.3 Selective Corrosion along the Metal Structure and its Countermeasures First, Photo 3.4.6-1 shows corrosion damage of the threads of a SUS 304 steel bolt and screw used for

seawater hydraulic devices along the non-metallic inclusion. Corrosion dissolution occurs in the shape of many tunnels along the longitudinal direction of the bolt and the edge is dissolved along Mns non-metallic inclusion.

The longitudinal direction of the bolt

non-metallic inclusion

(a) Cross-sectional View of a Stainless Steel Bolt Where Corrosion Occurred

(b) An Enlarged View of Portion (a)

Photo 3.4.6-1: Corrosion Damage on the screw of a SUS 304 Steel Bolt along Non-metallic inclusion

Based on this knowledge, in countermeasures against this kind of damage, it is decided not to use bolt materials produced through too much rolling and to obsreve the S content where possible. As a result, the occurrence frequency of damage has been reduced.

Next, we explain the relationship between cutting conditions and corrosion resistance. The difference in corrosion resistance quality between the best stainless steel and the poorest is a 3-digit

number at maximum even if all of them conform to the JIS standards because the conditions of the formation of machining-induced martensite phase change greatly depending on the alloy composition and the cutting conditions.

Figure 3.4.6-8 shows the results of tests where a part of a specimen is dipped in actual seawater for 16 months in total with the content of Cr as well as the content of Ni changed by a few steps within the JIS standard and also with the cutting conditions variously changed.

Cum

ulat

ive

Occ

urre

nce

(%)

Corroded Area Ratio

EvaluationCutting ConditionsMaterialSymbol

Fair

Excellent

Good

Depth of Pitting (mm)

Figure 3.4.6-8: Advancement of Pitting in SUS 316 Dipped in Seawater with Varying Cutting Conditions

Based on the above-mentioned knowledge and information, it is understood that the corrosion resistance of stainless steel dipped in seawater cannot always be grasped only by the average chemical composition but is strongly influenced by minute impurities, metallographic structure, and attached scale (25).

322

1.4 Grain Boundary Corrosion Damage and its Prevention Photo 3.4.6-2 shows an example of corrosion damage to a welded part of stainless steel. (a) is for cases

where grain boundary corrosion of an area near to the deposited metal and affected by heat resulted in a hollow being made, and (b) is for cases where priority dissolution of the deposited metal occurred.

Deposited metal

(a) A Hole Made due to Grain Boundary Corrosion of an Area Influenced by Welding Heat and the Surrounding Area

(b) A Hole Made due to Priority Dissolution of the Deposited Metal

Photo:3.4.6-2 Example of corrosion damages of welded part of stainless steel

It is basically important to reduce the influence of heat caused by welding to prevent this damage from occurring, and using a metal of high quality is not always effective.

Next, information on selective dissolution is presented in Figure 3.4.6-9(29)(30). This figure shows the result of a test where deposited metal and the base metal of SUS 304 stainless steel were tested in a corrosive environment. It also shows the fact that the deposited metal is inferior to the base metal over wide conditions of heat treatment in terms of corrosion resistance. As measures to prevent this damage, cancellation of alloy segregation by re-solution treatment at 900˚C or higher, utilization of hyper-low C steel to prevent sensitization, or prevention of alloy segregation by using 9% Mo- or N-added steel are considered effective (29)(30).

Wei

ght R

educ

tion

due

to C

orro

sion

(g

/m2 /h

)

: Deposited metal: Base metal

Two-hour hold One-hour holdAs welded As received

Figure 3.4.6-9: Influence of Heat Treatment on Pitting Corrosion Resistance of the Deposited Metal of SUS 304 Steel

Temperature of Heat Treatment (˚C)

323

1.5 Stress Corrosion Cracking and Hydrogen Embrittlement and Their Prevention Generally, in our experience, chloride stress corrosion cracking of austenite-group steel does not occur in

seawater at normal temperature. This material easily cracks due to the rich concentration of magnesium chloride in special seawater

environments where the temperature becomes 60˚C or higher(33) to (35) or in the event that it is exposed to an oceanic climate in a dry atmosphere whose relative humidity is as low as 30%(36) to (38).

Next, Figure 3.4.6-10(39) and Figure 3.4.6-11(40) show the hydrogen embrittlement of 17-4 PH steel that is used for the sliding parts of a seawater pump. According to the information obtained from these figures, cracks occur to materials whose hardness is more than 320 Hv, and the apparent advancing speed of a crack increases more for steels of higher hardness.

App

aren

t Spe

ed o

f Adv

ance

men

t of

Cra

ckin

g (m

m/h

) : Solution treatment at 850˚C : Solution treatment at 1050˚C

Numeral: Cracked specimens/Total specimens

Material hardness (Hv)

Figure 3.4.6-10: Behavior of the Sliding Part Material Made of 17-4 PH Steel Dipped in 3% Saltwater at Room

Temperature Related to Hydrogen Embrittlement

Low

est C

ritic

al S

tress

Exp

ansi

on

Con

stan

t K

I SC

C (k

g⋅f/m

m3/

2 )

Isolated dipping

When contacted with Zn

Yield strength σY (kg⋅f/mm2)

Figure 3.4.6-11: Behavior of 17-4 PH steel Dipped in the 3% Saltwater at Room Temperature Related to

Hydrogen Embrittlement

Figure 3.4.6-11 shows that a material with higher strength has more crack sensitivity. And crack sensitivity is higher in the case where the material makes contact with Zn (electrolyted protection) than in the case of isolated dipping.

Next, Figure 3.4.6-12 shows the influence of environmental conditions on hydrogen embrittlement cracking. For behavior in this type of cracking, seawater and freshwater are considered to have the same effect. In other words, pitting at the starting point occurs at first, and then a crack may occur when the dynamic conditions are satisfied. This point represents a large difference from chloride stress corrosion cracking

Cra

ck S

ensi

tivity

Inde

x I S

CC

Temperature (˚C)

Figure 3.4.6-12: Influence of environmental conditions on hydrogen embrittlement cracking of high-hardness

324

13Cr stainless steel (Hv=495)

Considering these findings, it is necessary to recognize the risk that hydrogen embrittlement crack can occur to very hard stainless steel even under mild environmental conditions if it is hardened too much by attaching to much importance to smooth movement of the part.

1.6 Cavitation Erosion Damage and its Preventive Measures

In some cases, cavitation erosion-corrosion damage may occur to the seawater pump, the valve, and the piping(42) to (47). This type of damage is caused by cavitation (a cavity filled with vapor of the liquid) appearing around a minute bubble or particle of dust in the liquid as a nucleus when the static pressure of the fluid boundary in a high-speed flow drops to a lower value than the saturated vapor pressure. The damage is caused by a high-pressure shock wave or a micro-jet formed in a situation where the bubbles are rapidly broken when the flow speed is reduced or when the flow reaching a high-pressure area crashes strongly into the surface of the material.

The special feature of this damage is the phenomenon where, at the initial stage, countless minute holes are formed on the surface at a specific place just after the point where the flow direction is suddenly changed (for example, a place just after the leading edge of a vane), then, a deep part of the material is gradually and selectively corroded by the so-called “inlet effect” as time passes. As a result, this corrosion often develops to cause such severe damage that the surface becomes so rough that it can cut a finger. And it is not unusual that this corrosion leads to fatigue rupture resulting in serious damage for a long time because a load is repeatedly added even if the impact force is weak.

The parameters of the fluid dynamics influencing this damage are flow speed (strictly speaking, specific speed), pressure, and properties of the fluid. Since the intensity of cavitation is in proportion to approximately a fifth to sixth power of the flow speed, great prudence should be exercised when using a higher fluid speed in a fluid machine (49) to (52).

Thus, since this kind of damage is complex where both mechanical and chemical actions are involved, profound knowledge and experience are necessary to ascertain the situation of damage and to decide future measures to be taken. A careful attitude towards measures to be taken where you record conditions of the facility and equipment and listen to specialists is required.

Next, we explain countermeasures against cavitation damage. What method you use is a matter of choice between utilization of fluid dynamics and selection of the material to use. Basically, the former is more effective (45)(53).

On the other hand, the life is extended by a 1-digit number at the most even if a superior material in terms of durability is used. So, this choice is not very positive. However, since the life can be extended more or less by remaking a part with a higher-class material (partly plastering a part with such a material) (repeated amendment can further extend the life), this method is often used. A material with superior cavitation erosion resistance is also corrosion resistant and strong (54) to (56).

Well-known materials with the highest cavitation erosion resistance include Co base padding alloy (e.g., stelite), Ti alloy, the 300 type (e.g., SUS304), precipitation hardening-type stainless steel (e.g., 17-4 PH steel), and Ni base alloy (e.g., hastelloy 625).

However, since even two types of steel belonging to the same group can be very different in their cavitation corrosion resistance, the actual choice of material should be conducted very carefully.

Figure 3.4.6-13 shows the cavitation erosion resistance of the materials to be used for the structural components of hydraulic machines such as the seawater pump (58).

Austenite steel as a deposited metal

Martensite steel Cast iron

cast iron

Red

uctio

n in

wei

ght d

ue to

ca

vita

tion

eros

ion

(mg/

2 h)

Nickel equivalent

Figure 3.4.6-13: Relation between the Amount of Damage due to Cavitation Erosion and Hirayama’s Nickel Equivalent

325

Figure 3.4.6-13 shows that each curve representing a hardness value of 2~7Cr steel or 13Cr steel shows a downhill tendency against the Ni equivalent, while each curve representing a hardness value of 18~30Cr-7~20Ni steel shows a uphill tendency.

This fact makes us recognize that there is a big difference in cavitation erosion resistance among some types of steel even belonging to the same standard due to even slight differences in their alloy composition.

Therefore, when you engage in the design of an actual hydraulic machine or selection of a material to be used for repair of damage, you should calculate the nickel equivalent of the material to be used and check the place to be repaired in the drawing. In some cases, extension of damage life can be expected by a slight change to the metal composition.

Figure 3.4.6-14 and Figure 3.4.6.15 indicate that it is preferable to choose a material whose strain energy is as large as possible among types of steel whose corrosion resistance is almost the same.

Am

ount

of d

amag

e du

e to

cav

itatio

n er

osio

n (m

g/2

h)

Hirayama’s Ni equivalent

C zoneB zoneA zone

Figure 3.4.6-14: Relation between Amount of Damage due to Cavitation Erosion of18Cr-6Co Stainless Steel and Hirayama’s Ni Equivalent Using the C Content as a Parameter

Mat

eria

l har

dnes

s jus

t und

er th

e su

rfac

e su

bjec

ted

to a

cav

itatio

n er

osio

n te

st (H

v)

After the cavitation erosion test

Before the test

C content (weight %)

Figure 3.4.6-15: Relation between Material Hardness Just under the Surface Subjected to a Cavitation Erosion

Test of 18 Cr-8Ni and 18Cr-6Co Stainless Steel 1.7 Corrosion Fatigue

Material characteristics in terms of fatigue in the atmospheric environment do not have a significant meaning in the prevention of corrosion fatigue damage or in assessment including characteristics of the material related to progression of pitting. However, we have not been able to find a common rule through various kinds of materials. It is evident that tensile strength and fatigue characteristics in the atmospheric environment do not have a significant meaning in terms of corrosion fatigue strength, and there seems to be no simple and logical damage prevention.

326

2 Copper Alloy Photo 3.4.6-3 shows dezincification corrosion of a part made of brass.

Dez

inci

ficat

ion

corr

osio

n

Photo 3.4.6-3: Cross-sectional View of Dezincification Corrosion of Brass Part

Table 3.4.6-2 lists the maximum allowable flow speed in a seawater heat exchanger (66). Since the mechanical strength of the copper alloy surface is weak, it is liable to suffer from erosion damage or impingement attack. Therefore, the max. service flow speed is approximately 1 m/s or lower for a pure copper pipe and 2 m/s for a steel alloy pipe.

Table 3.4.6-2: Maximum Flow Speed that can be Used in the Seawater Heat Exchanger Pipe

Material Max. Flow Speed (m/s)

Copper 0.9 Silicone bronze 0.9 Admiral brass 1.5 Aluminum brass 2.4 90/10 cupro nickel 3.1 90/30 cupro nickel 3.6 Monel 400 No limitation to the max. flow speed specified 316 steel No limitation to the max. flow speed specified Incoloy 825/Carpenter20Cb No limitation to the max. flow speed specified Inconel 625/Hastelloy C No limitation to the flow speed specified Titanium No limitation to the flow speed specified

On the other hand, in the case of a stainless steel pipe, since a slow flow speed (1 to 2 m/s or slower) allows marine creatures to easily accumulate causing local corrosion, a faster flow speed is preferable if pressure loss can be ignored.

Pitting of a copper alloy pipe used for a seawater heat exchanger can occur when the seawater is polluted (67). Injection of ferric sulfate (Fe2+) into seawater whose concentration is brought to 0.01 ppm is considered

effective. The mechanism according to which this works is explained as follows. FeOOH that is produced by oxidization of Fe2+ is attached to the inner surface of the pipe, forming a protective film there (68)(69).

However, the industry has recently been restrained from using this method in some cases from an environmental protection viewpoint because the seawater is slightly colored when this method is used. And this method is not very effective when the seawater is polluted (70)(71).

327

3 Heavy Corrosion Protection Coating and the Lining Parts 3.1 Steel Plates to Which Heavy Corrosion Protection Coating is Applied

Table 3.4.6-3 shows examples of causes of corrosion that occurred to parts to which heavy corrosion protection coating was applied. This table reveals that the damage was caused in various production processes such as assembly operation, painting operation, design, and maintenance operation. Table 3.4.6-3: Causes of Corrosion Damage to Parts in a Structure Exposed to a Flowing Seawater Environment to

Which Heavy Corrosion Protection Coating Was Applied and the Their Classification

No. Cause of Damage Classification A-1 Separation of a paint film caused by tightening of bolts or nuts

-2 Separation of a paint film caused by contact and/or impact by a wire rope Assembling operation

B-1 Faulty pretreatment (remaining rust, oil, and/or chalk) -2 Faulty pretreatment (without Zn-rich primer) -3 Faulty paint film (too much hardener or solvent) -4 Improper application interval/inappropriate cure

Painting operation

C-1 Improper macroscopic shape of the base part -2 Improper detailed shape of the base part (corner of the end portion)

Design work

D-1 Deterioration of a paint film (too a period of irradiation of sunlight) E-1 Painting of a stainless steel (crevice corrosion)

Maintenance operation

It is required to choose such a coating material that will not cause problems in each process (for example, a material with mechanical flexibility) as shown in the Table 3.4.6-3 in the case where the heavy protection coating under discussion provides a film thickness of hundreds of µm. Coal tar epoxy paint that is widely used is considered to be reasonable from a practical viewpoint (70)(71). 3.2 The Lining Parts

Carbon steel is preferable for the base material. Sufficient care shall be taken to prevent a faulty operation from being conducted when installation of the

rubber lining is manually conducted. 4. Methods of Corrosion Protection 4.1 Suppression ofGalvanic Corrosion

The most noticeable feature of seawater corrosion is that the seawater environment tends to cause galvanic corrosion.

This chemical action causing the galvanic corrosion in question not only occurs between different metals but also between a sensitized part such as a welded part and a sound part, and between a defective part of a paint film and a sound part of the film.

From this viewpoint, corrosion prevention technology is a technology to minimize galvanic action. Table 3.4.6-4 is a summary of these methods. In this table, the sections corresponding to A-1 to A3 describe

methods of reducing electrochemical action between the materials, and those corresponding to B-1 to B-3 explain methods of suppressing corrosion damage even in presence of electrochemical action between the materials. Detailed explanation is omitted.

328

Table 3.4.6-4: Examples of Methods of Reducing Galvanic Corrosion Damage in a Structure Used in a Seawater Environment

No. Method Description Actual Case

A-1

Isolation of the materials from each other (by means of a spacer, paint, etc.)

b) a)Coating

A spacer

A-2 Use of materials whose potential difference is small (use of the same metal)

The combination should be stainless steel/stainless steel, copper alloy/copper alloy, carbon steel/carbon steel, etc. (The same grade is preferable.)

A-3

Methods of relieving interaction between the materials

Provision of a longer distance between materials (to lessen the potential gradient)

B-1

Provision of a larger anode/cathode area ratio (A2/A1)

B-2

Application of paint to the part to be the cathode

B-3

Methods of relieving corrosion damage even if interaction exists

Making the anode material replaceable (Treating it as an expendable part)

Using austenite cast iron as a washer

A copper alloy bolt Carbon steel

b) A2/A1 = large (correct)

a)A2/A1 = small (wrong)

Stainless steel Carbon steel

a) b) Paint film

Even corrosion occurs (correct)

Local corrosion occurs (wrong) Paint film

Using carbon steel for bolts that are easy to replace (large-sized ones) and replacing them from time to time

You should understand that the word “corrosion potential series” that often appears when discussing galvanic action just means an order and has no relation to the speed of galvanic action. For example, it is a well-known fact that if the speeds of consumption of a zinc plate are compared between two cases, one where a zinc plate is made to contact a stainless plate whose area amount is the same as that of the zinc plate (the potential difference is 1.0 V or less) and the other where the zinc plate is made to contact a carbon steel whose area amount is the same as that of the zinc plate (the potential difference is 0.5 V or less), the speed of consumption of the zinc plate in the latter case is faster by as much as 30 times at maximum than the former case. 4.2 The Electrochemical Protection Method

The cathode corrosion prevention method where the surface electric potential is made to be the base potential is the most popular among electrochemical protection methods in the seawater environment. The anode corrosion protection method where the surface potential is made to be the passive state potential is not practically used because this method sometimes dangerously accelerates anode dissolution in crevices in a complex structures.

There is the impressed current method and the galvanic anode protection method (the sacrificial anode method), but the former is relatively rarely used. For the impressed current method, it is necessary to install an insoluble electrode (graphite or magnetic iron oxide electrode) at the fluid surface and to connect the lead wires extended from the outside power source with each part. For this method, it is necessary to devise a special fixing measure so that the fixed parts do not interfere with the fluid flow. And since corrosion damage to parts that are made of stainless steel sometimes occurs during suspension of operation (installation, periodic inspection, etc.) when the flow in the equipment also stops, in some cases, unexpected damage occurs because electricity is not supplied during these periods. And an inappropriate design may cause overprotection in certain cases, resulting in separation of the paint film or the lining due to the large amount of hydrogen gas generated at the cathode. Also, in certain cases, severe corrosion damage may occur due to chlorine gas with strong oxidizing power in equipment whose valve is left completely closed.

On the other hand, this kind of bad effect seldom occurs in the case of the galvanic anode protection method because the current amount is relatively small. Aluminum (current efficiency: 80%) or zinc (current efficiency: 95%) rather than iron or steel should be used for the anode to be attached because they are cheaper and their self-electrode potential is lower than iron or steel (71). Mg, when attached, provides too low an open circuit potential and lower current efficiency (fast consumption), causing overprotection in certain cases. It is necessary to periodically replace the anode because the anode is consumed when the galvanic anode protection method under discussion is used. When a replacement period longer than that of the periodic inspection is desired, it is important to reduce the value of the [protected area/area of the anode]. However, it becomes important to pay as much attention as possible to the reliability of the heavy corrosion protection coating when it is difficult to attach many anodes. You should expect very fast consumption of the anode when a carbon steel part with a large area is exposed to fast-flowing seawater.

329

3.5 Power Generators 3.5.1 Preventive maintenance of electric and control facilities and remaining life assessment technology 3.5.1.1 Turbine generator

Securing the reliability of the turbine generator is a very important issue from the viewpoint of stable supply of electricity.

The major reasons are that the percentage of each machine in the system has been increased and the rate of shutdown of large-capacity generators due to accidents is high in the world. In addition to the increase in the aged thermal power plants, thermal power plant units can be managed more effectively than before. Middle management of existing units increased, and strict management of power facilities is obliged, resulting in acceleration of deterioration.

In order to prevent severe accidents that require long-term shutdown and high cost for repair, improvement based on various knowledge in the field and careful inspections are required. In addition, even after starting the operation, it is important to detect accidents at the initial stage by various failure diagnosis methods and take proper measures for them.

Preventive maintenance and remaining life assessment technology of turbine generators are introduced below. 1. Maintenance of turbine generators

The repair of turbine generators is generally classified into two: by daily repair, the failure is repaired during operation or by stopping power generation, and by periodical inspection, power generation is stopped periodically for a long period of time for inspection and repair. One example of maintenance at periodical inspection is shown in Table 3.5.1-1.

Table 3.5.1-1: Maintenance of power generator at periodical inspection Power generator Power generator accessories Test related to power generator

(1) Stator • Coil, supporter, iron core, precision

inspection (2) Rotor • Coil end section, fan attachment

section, wedge, slip ring, brush holder inspection

(3) Bearing • Inspection and maintenance of white

metal • Inspection of insulation plate • Gap measurement (4) Airway • Cleaning of air filter (5) Hydrogen gas cooler • Cleaning of tube, water pressure test

(1) Stator cooling water device • Replacement of ion-exchange resin • Overhaul of cooling water pump (2) Hydrogen gas-sealing oil device • Cleaning of tank inside • Overhaul of sealing oil pump and

vacuum pump • Inspection of drying agent of gas drier

• Measurement of insulation resistance • Hydrogen-sealing oil and rotor-cooling

water device sequence test • Measurement of derivative loss angle • Measurement of bearing insulation

resistance • Gas-sealing test

2. Aged deterioration and remaining life management

The causes of aged deterioration of the structural parts of turbine generators are generally classified as follows: (1) Low-cycle fatigue and wear due to increase in the number of starts and stops High-cycle fatigue and

wear due to vibration, etc. (2) High cycle fatigue and wear due to vibration, etc. (3) Fatigue and wear due to heat cycle (4) Deterioration and functional decrease due to long-term operation and changes in the environment (5) Combination of (1) to (4)

(1) above is caused mainly by elongation due to centrifugal force, and has a large influence on rotor parts. (2) is the high-cycle fatigue caused by electromagnetic vibration of the iron core and coil. It causes loosening

of the rotor parts and stator parts. When it is combined with other causes, it can cause aged deterioration. (3) is the deterioration of the coils, insulation materials, or the like that is influenced by the heat cycle due to

changes in load, etc. In addition to this, the causes of deterioration are thermal deterioration due to local overheating by abnormal operation and changes due to expansion/contraction, following the temperature.

(4) is exactly the aged deterioration. In addition to wear, erosion, and thermal deterioration, it includes ground sinking of the mount and basis, lowering of functions due to the atmosphere and the environment, etc.

The mechanism of structural and functional aged deterioration of structural parts has almost been clarified. Periodical inspection and failure diagnosis device are applied to each section. Various strength improvement measures are taken, and remaining life is estimated from the operation history to improve the reliability.

330

Concerning dielectric breakdown that is peculiar to electric power generators in particular, active studies are made including the sampling investigation of actual coils. 3. Preventive maintenance and life control of major parts 3.1 Rotor 3.1.1 Prevention of destruction of rotor shaft material

In the rotor shafts that were manufactured before modern advanced production technology and inspection technology were established, defects in materials such as non-metal or sand most frequently tended to be detected in the area near the center hole with the highest stress in the shaft.

Recently, internal defects of the rotor shaft can be more reliably detected than before by the new methods including ultrasonic flaw detection from the center hole and high-sensitive ultrasonic flaw detection from the surface. The method of judging the possibility of progress of non-metal substances by fracture mechanics method has advanced a great deal, and deterioration caused by internal defects of the rotor shaft can be prevented without fail.

Power generators with a large number of starts and stops and power generators that are operated for a long time are inspected first. Even if it is not started or stopped frequently, if it is operated for 15 years or more or the total operation hours is 100,000 hours or more, it is desirable to perform non-destruction inspection. 3.1.2 Rotor wedge

There are some types of rotor wedge: high-strength aluminum, magnetic iron, non-magnetic iron, and copper. They are used depending on their character.

During unbalanced load operation, a dual-frequency current flows on the rotor surface due to the reverse phase magnetic field on the rotor surface, and the rotor body end might be damaged. If excessive current flows on the rotor surface, the arc might be stroked when excessive current is transited between the teeth section and wedge section depending on their contact condition. A trace of electric corrosion remains on the wedge and teeth section.

If contact between the teeth section and wedge section is uneven due to centrifugal force, circulation current is concentrated or the teeth are rusted. Even slight rust is significantly increased, resulting in damage to the wedge. Thermal damage to on the rotor surface due to multi-frequency excessive current is protected by the reverse phase relay. It must be noted in particular that electric corrosion can occur even in the protection range of the reverse phase relay because the reverse phase relay protects the rotor from the thermal viewpoint.

The area with a trace of electric corrosion is in the quenched condition, and the hardness is increased and the extension toughness lowers. Therefore, initial cracks tend to occur at this area, and there is a fear of progress of cracks starting from here. The electric corrosion trace cannot be detected in non-destructive inspection. The only detection method is a direct check by sampling of wedges. The wedge damage and inspection concept are shown in Fig. 3.5.1-1.

331

During production and precision inspection

During operation

Periodical nondestructive inspection

Serious accident

Progress of cracks

Superimposed start/stop and

high-temperature creep

Cracks Centrifugal

stress during operation

Accumulation of starts and

stops

Electric corrosion trace

Deterioration of creep strength

Deterioration of low-cycle fatigue

strength

Reverse phase current

High-frequency current

High-temperature

tank

Asynchronous operation Unbalanced operation High-frequency operation Overexcitation operation Excessive speed operation Short circuit accident Re-close operation

Abnormal operation

Inspection Check of properness

Reverse phase magnetic fieldRotor-rotating direction

Polar surface Cross slot

ProbeTeeth (iron core) Wedge

Ultrasonic ShaftWedge

Damper ring Field coil

DefectWedge

Coil

Wedge ultrasonic flaw detector

Retaining ring Coil slot

Fig. 3.5.1-1: Damage to wedge and concept of inspection

In general, it is desirable to perform wedge diagnosis by ultrasonic flaw detection from the rotor surface in the first year after starting operation and every 4 years. It is also desirable to perform visual inspection by 100% sampling and precision inspection by dye penetration flaw detection every 8 years. The interval of “every 8 years” was decided with proper allowance based on the occurrence of cracks in the superimposed test of creep and fatigue of high-strength aluminum at high temperature and the progress speed in Fig. 3.5.1-2. This is preventive maintenance with the intention of inspection of the field magnetic coil by pulling the retaining ring along with wedge sampling. This creep crack progress speed should be controlled yearly regardless of the operation mode because static creep, depending on the operation hours, is dominant in the high-temperature field, and dependence on the number of starts and stops is low. In the low-temperature field, the rate of starts and stops is dominant, and closer inspection of DSS machines is desirable.

(DSS operation)

(Continuous operation)

Fig. 3.5.1-2: Crack progress speed of high-strength aluminum alloy wedge

Knowledge on the progress of creep cracks is valuable data obtained in the wake of total-loss accidents of turbines and power generators due to damage and splash of the wedge of the industrial power generators in the

332

1970s. In the remaining life evaluation, the presence of thermal influence is checked by confirming wedge hardness.

Defects are detected by ultrasonic flaw detection or magnetic powder flaw detection. The crack size is evaluated by the superimposed creep and fatigue from the actual stress. 3.1.3 Retaining ring

The retaining ring holds the end section of the field magnetic coil from centrifugal force. For electromagnetic reasons, the ring must be nonmagnetic and highly durable. Austenitic steel is normally selected. It is attached to the rotor shaft end by shrink fit. Stress functions even in the static state, and centrifugal stress is added during operation. That is, it must be noted that tensile stress is always produced even while operation is stopped. Many ring fracture accidents have been reported including one accident in Germany, one in Sweden, and one in Denmark. The cause is estimated to be SCC (stress corrosion cracks).

If the retaining ring is not handled properly during installation or periodical inspection and dewing occurs or leak water is accumulated in the ring, it is dangerous for this material, and special attention is required.

Recently, extremely severe SCC cracks were detected at the periodical inspection of industrial machines in Japan as shown in Fig. 3.5.1-3. The progress of cracks by approx. 10 mm was confirmed over 3 years.

Periodical inspection Periodical inspection

Retaining ring External UT External PTNo defects

Retaining ring Eternal SCC found Internal SCC defect depth about 10mm

SCC

cra

ck d

epth

(mm

)

3 yearsSCC progress speed 2.5~3.3 mm/year

Inspection history and SCC progress condition

After removing 6 mm

Fig. 3.5.1-3: Example of SCC of domestic industrial machine

SCC defects during deletion

As preventive maintenance, in the same way as the wedge, ultrasonic flaw detection from the external surface

is used at periodical inspection, and the entire surface is dyed for flaw detection when the ring is pulled at precision inspection.

SCC in the retaining ring occurred because it was exposed to a wet atmosphere. It is important to keep the gas temperature in the machine above the dew point while the machine is stopped for a long period of time, and to take moisture-proof measures of the retaining ring during open inspection. Dewing can be prevented by keeping the retaining ring temperature higher than the ambient temperature by several degrees. The hot air-type heater, heater, or lamp irradiation is effective.

Exchanging the retaining ring of the hydrogen-cooling device with a rotor of a large diameter is promoted. It is also necessary to exchange the air-cooling-type power generator or exciter of small capacity promptly.

Because the retaining ring has the maximum stress among turbine power generator rotors, it is necessary to confirm that there are no cracks by the above-mentioned periodical precision inspection even if the material is 18Mn18Cr that has overcome the problems of SCC. 3.1.4 Cutting of polar connection conductor

As shown in Fig. 3.5.1-4~6, the polar connection conductor is integrated in the retaining ring that is displaced in the radius direction due to centrifugal force. It repeats extension and contraction, following the swelling of the retaining ring for starts and stops, and fatigue cracks might occur.

It is necessary to confirm the progress of occurrence and deformation of fatigue cracks by a fiber scope and mirror at periodical inspection and to take proper measures.

333

Insulation cylinder Polar connection

conductor section

Retaining ring

Lead conductor

Rotating span Rotor shaft

Stopped span

Polar connection conductor

Retaining ringInsulation cylinder

Fig. 3.5.1-4: Example of structure of polar connection conductor and behavior

#7 coil

#7 coil #6 coil

Retaining ring (Insulation cylinder side)

#6 coil

(The coil quantity is indicated.) Crack Rotor shaft side

(Note) 1. The odd-number coil (#5, 7 coil) passes on the retaining side.

2. The even-number coil (#6, 8) passes on the rotor shaft side.

Polar connection conductor

[Example of #7 coil]

Fig. 3.5.1-5 Example of damage to polar connection conductor

Insulation Retaining ring

Centering ring Rotor coilPolarity crossing

Polarity crossing

Flexible section

Shape of polarity crossing

Fig. 3.5.1-6: Example of polarity connection conductor structure 3.1.5 Rotor coil (1) Copper powder

In the power generator that directly cools the rotor coil by cooling gas (hydrogen), the contact surface between two coil conductors in the same turn becomes rough like pear skin.

As shown in Fig. 3.5.1-7, there is a gap between the slot and the coil in the width direction and vertical direction due to coil assembly. The coil moves laterally and vertically in the slot at low-speed turning. In this case, copper powder is generated due to relative sliding between the same turns of multiple coils.

334

Surface roughnessWear between copper bands

Copper powder

Movement of coil in slot due to own weight

Hydrogen gas atmosphere (No oxygen)

No centrifugal forceLow-speed turning

Gap at slot assembly launching

Connection between coppersDouble-coil copper band

Coil of parallel section Slot Coil of taper section

Movement of coil in slot during

low-speed turning

Fig. 3.5.1-7: Mechanism of generation of copper powder

If copper powder accumulates in the slot, there is a possibility of short circuit or earth fault as shown in Fig. 3.5.1-8. Some accidents have been reported in the U.S. Because the copper powder increases along with the increase in the turning time, it is necessary to clean it at an early stage. Countermeasures include the method of fixing layers by dot brazing by eliminating relative sliding in the same turn consisting of multiple coils and the method of inserting an insulation sheet between conductors.

Rotor wedge

Clippage block

Short circuit between coils (if failure also occurs in other slots) Rotor grounding

Rotor coil (Same turn)Layer short circuit

Slot insulation

Turn insulation

Fig. 3.5.1-8: Failure of rotor due to copper powder (2) End turn

It is necessary to check the rotor coil under the retaining ring periodically using a fiber scope or mirror to see that there is no thermal elongation during operation or deformation due to starts and stops. It is also necessary to check clogging of the vent hole of the cooling gas of the direct cooling machine as well as the clogging of the cooling hole of the straight section. 3.1.6 Insulation of rotor coil (1) Other insulators

Other insulators include the following: layer insulation between coil conductors, lining plate for reinforcing the slot insulation under the slot insulation, and insulating block between the rotor coils under the retaining ring for preventing coil deformation due to thermal elongation during operation and start/stop.

At periodical inspection, it is necessary to inspect the dropping of blocks and the positions of the layer insulation and lining plate, using the fiber scope and mirror under the retaining ring. 3.1.7 Field magnetic coil lead

The rotor coil has a lead conductor for connection from #1 coil, and is fixed by the wedge in the same way as the field magnetic coil.

The lead conductor is connected to the terminal stud and connected to the collector ring via the center hole conductor. There are two types of lead conductors: one is the solid type consisting of an integrated conductor, and the other is the flexible type consisting of thin layers of copper plates.

It is necessary to inspect the lead conductor rise section periodically because it repeats the movement, following swelling of the retaining ring when the unit starts and stops. (1) Flexible type

Because of the structure, the flexible type in Fig. 3.5.1-9 is inspected from below the retaining ring for damage

335

to the insulator and deformation of the conductor without disassembling the lead wedge.

Coil Retaining ring

Centering ring

Inspection

Lead conductor (Flexible)

Lead wedgeInsulation plate

Fig. 3.5.1-9: Flexible-type field magnetic coil lead

The soundness of the lead wedge is confirmed by the ultrasonic flaw detection test in which the lead wedge is inserted to the rotor shaft and, if possible, by magnetic powder flaw detection inspection after the lead wedge is pulled out.

Recently, the flexible type has been replaced with and improved to the solid type for improving reliability and for facilitating and securing inspection. (2) Solid type

In the case of the solid type in Fig. 3.5.1-10, the lead wedge is disassembled, and the lead conductor rise and bending section (R) are inspected. Soundness is confirmed by visual inspection using a fiber scope and liquid penetration flaw detection test. The soundness of the pulled lead wedge is confirmed by the liquid penetration flaw detection test or magnetic powder flaw detection test.

#1 coil Retaining ring Centering ring

Fan boss

Inspection

Fiber scopeLead conductor (Solid) Terminal studLead conductor

rise section

Fig. 3.5.1-10: Solid-type field magnetic coil lead 3.2 Stator 3.2.1 Transition in insulation method

The capacity of the turbine power generator has been increased and the size has been reduced mainly by improvement of the cooling method and development of the insulation system. Stator coil insulation of the rotating electric machines including the turbine generator mainly consists of mica with superior corona resistance and impregnated resin for holding the mica.

As shown in Fig. 3.5.1-11, the impregnated resin of natural resin (shellac resin, asphalt compound, etc.) was used. Along with development of synthetic resin with superior heat resistance, polyester resin and epoxy resin came to be used. Polyester resin has been used from the middle of the 1950s, and epoxy resin has been used from the late 1950s.

No. Mica base Impregnated resin Natural resin Flake mica

Polyester resin

Flake mica and laminated mica

Laminated mica Epoxy resin

Year

Fig. 3.5.1-11: Transition in insulator

Because the characteristics of polyester resin are lower than those of epoxy resin, epoxy resin is mainly used recently.

Flake mica was used as mica base at first. Laminated mica (crushed flake mica) has been used from the early

336

1960s. At present, flake mica and laminated mica are used depending on the voltage level, coil dimension, etc. Because it has become difficult to obtain flake mica of good quality recently, the use of insulation systems with laminated mica is increasing.

It is necessary to eliminate the void in the insulation layers of the stator coil to the extent for the prevention of corona. For this purpose, impregnated resin is used. The resin impregnation method is generally classified into two: the vacuum pressure method (VPI method), and the resin-rich method. The vacuum pressure method of the coil was used for synthetic resin mica insulation at first. In the middle of the 1960s, the resin-rich method came into practical use. In this method, semi-cured mica tape is used, and an insulation layer is formed by pressurization and heating after taping. At present, both the vacuum pressure method and resin-rich method are used depending on the voltage level, power generator dimension, or the production facilities of the maker. 3.2.2 Mechanism of dielectric breakdown

The causes of deterioration of the stator coil insulation are generally classified as follows: (i) heat (ii) electricity (iii) machine (iv) environment. Figure 3.5.1-12 shows the deterioration process of the stator coil insulation(7). Because the impregnated resin of natural resin has plastic elasticity, heat deterioration such as softening, flow-out, and sublimation progress, resulting in voiding and peeling. It generally leads to dielectric breakdown due to partial discharge.

Dielectric breakdown

Tracking due to increase in leak current

Surge voltage Regular voltage

Deterioration of insulation resistance due to moisture absorption and pollution

Environmental deterioration

Electric deterioration

Mechanical deterioration

Void and peeling due to softening, flow-out, sublimation, etc. of impregnated resin caused by temperature increase

Regular voltage

Erosion and void increase due to partial discharge

Peeling and cracks due to electromagnetic vibration

Void and peeling due to thermal/mechanical stress by heat cycle including start/stop and load change

Thermal deterioration

Fig. 3.5.1-12: Deterioration process of stator coil insulation

On the other hand, the synthetic resin insulation that has been used recently has high heat resistance, and the influence of heat deterioration is small. However, as shown in Fig. 3.5.1-13, voiding and peeling occur mainly in the insulation layer due to the compound effect of the mechanical stress in the heat cycle along with the change in load, starts and stops. In the operation after that, voiding and peeling grow and progress and partial discharge occurs, resulting in dielectric breakdown(7).

Wire

Main insulation layer

Tape layer VoidPeeling

Fig. 3.5.1-13: Deterioration of insulation layer Wire insulation

In addition to the above-mentioned dielectric breakdown in the main insulation layer, in the case of the

synthetic resin insulation, damage (vibration spark) to the coil surface is reported. The mechanism of occurrence of vibration sparks is as follows.

Because the synthetic resin insulation has the characteristics of heat contraction over time, the coil is loosened and the coil is vibrated by the magnetic force. On the other hand, a low-resistance shield is applied on the coil surface to prevent surface corona. Almost the same voltage as the coil occurs in this low-resistance shield. Along with coil vibration, the contact point on the coil surface and iron core moves and partial discharge (vibration spark) occurs on the coil surface, resulting in damage to the coil surface.

To prevent these, a liner with a spring function is inserted for absorbing the heat contraction of the coil and

337

338

wedge. Inspection of wedge looseness at periodical inspection and re-insertion of the wedge, if necessary, are also effective measures to prevent these phenomena.

edge. Inspection of wedge looseness at periodical inspection and re-insertion of the wedge, if necessary, are also effective measures to prevent these phenomena. 3.2.3 Dielectric breakdown diagnosis technology 3.2.3 Dielectric breakdown diagnosis technology

The diagnosis method of dielectric breakdown is generally classified into the visual inspection and electric insulation characteristics test. Visual inspection is very important in general. Damage, cracks, peeling, discoloration, and wear on the insulation are checked. The electric insulation characteristics test(8) in Table 3.5.1-2 is currently executed.

The diagnosis method of dielectric breakdown is generally classified into the visual inspection and electric insulation characteristics test. Visual inspection is very important in general. Damage, cracks, peeling, discoloration, and wear on the insulation are checked. The electric insulation characteristics test

Table 3.5.1-2: Nondestructive test items (Typical example) Table 3.5.1-2: Nondestructive test items (Typical example)

No. No. Item Item Contents Contents

(8) in Table 3.5.1-2 is currently executed.

1 Measurement of insulation resistance Megger 2 Direct absorption test PI: Megger (1 min., 10 min.) 3 Alternative current test

Applied voltage (kV)

4 Dielectric tangent test

Applied voltage (kV)

5 Partial discharge test

Qmax = Max. discharge charge amount

(Note) 2I: Average deterioration Qmax: Local deterioration (1) Measurement of insulation resistance

Insulation resistance is measured using a simplified insulation resistance gauge called “megger.” Degradation of insulation resistance due to absorption and pollution on the insulation layer surface is checked to estimate the tendency of dielectric breakdown. 1000 V mega is usually used. (2) Direct current absorption test

Leak current on the insulation layer is measured 1 minute and 10 minutes after DC high voltage is applied. Insulation performance, absorption, and pollution on the insulation layer, in particular, are judged. (3) AC test

The current-voltage characteristics when AC voltage is applied change depending on absorption, pollution, dielectric breakdown, and partial discharge. Insulation properties are estimated from these characteristics. (4) Dielectric tangent test

This is called the “tanδ test” in general. If void discharge occurs in the insulation layer, the discharge current is added and the tanδ value becomes larger. In general, the increase (∆tanδ) from the tanδ value at 2 kV (expressed as tanδ0 in general) indicates the relative void discharge amount. (5) Partial discharge test

By measuring partial discharge pulses directly, generation and progress of voids in the insulation layers are checked and deterioration progress is estimated. Other deterioration diagnosis methods that are put into practical use and tried are as follows. In partial discharge diagnosis, slot discharge between the coil surface and slot wall during operation is detected and the insulation layer condition is judged. By analysis of the chemical substances in the machine, abnormal heating in the power generator during operation is judged. 3.2.4 Dielectric breakdown judgment standard

For judgment of dielectric breakdown, the deterioration characteristics values including ∆tanδ and Qmax obtained in the above-mentioned nondestructive test are summed up and judged. The list of standards that are proposed or practically used as the deterioration judgment standard is shown in Table 3.5.1-3(9). If it is estimated that the dielectric breakdown voltage is (2E+1) kV or (2E+3) kV or more according to each nondestructive characteristics value, it is regarded that the machine has the insulation resistance required for operation.

AC

Partial disch rge superimpositiona

Applied voltage waveform

339

Table 3.5.1-3: List of dielectric breakdown judgment standards of wires on high-pressure rotating machine

Company name A B C D E F G H

Diagnosed rotating machine

6.6~11 kV Power generator

≥11 kV Power generator




11 kV Power generator



Insulation resistance R1 (MΩ) <12 NG <12 NG

Polarization index (PI=R10/R1)

≥1.5 OK ≥1.5 OK ≥1.5 OK ≥1.5 OK ≥2.0 OK ≥2.0 OK ≥1.5 OK

tanδ0 (%) at 2 kV (6.6 kV: at 1 kV) ≤10 OK

≤2.0

≤1.0

Caution needed

NG 0.3~3 OK ≤10 OK

at 1.25E/ 3 ≥2.5 NG ≥2.5 NG ≤2.5 OK ≥0.8

≥2.5

Caution needed

NG ∆tanδ (%)

at E ≥1.5

≥6.5

Caution needed

NG ≥1.5 Caution

needed ≥6.5 NG >6.5 NG

∆tanδ+∆C/C0 (%) at E <12 OK

at 1.25E/ 3 ≥5 NG >5 NG <5 OK∆1 (%)

at E ≥12 NG >12 NG

at E/ 3 <10,000 OK≥10,000

≥30,000

Caution needed

NG

≥10,000

≥30,000

Caution needed

NG

≥10,000

≥30,000

Caution needed

NG <10,000 OK

at 1.25E/ 3 >10,000 NG≥10,000

≥30,000

Caution needed

NG

qmax (pC)

at E NG ≥22,000

Nq qmax

at 1.25E/ 3 ≥2.0

≥30,000

Surface discharge is dominant.

Insulation resistance required for operation 2E+1 kV 2E+3 kV 2E+1 kV 2E+1 kV 2E+1 kV 2E+1 kV 2E+1 kV 2E+3 kV

3.2.5 Remaining life evaluation method There are varieties of models and ratings of power generators. The operation conditions including

management methods, start/stop frequency, and operation hours are also diverse. There are various nondestructive remaining life estimation methods considering these, and examples of applications are shown in Table 3.5.1-4(9).

Table 3.5.1-4: Example of application of estimation of remaining life of power generator stator coil [Nondestructive method]

Estimation method Method Outline NY map Operation condition

Operation history method

Equivalent operation time

Obtained from operation hours and number of start/stop times

D map Obtained from discharge parameter ∆ and max. discharge charge amount

Max. discharge charge amount Obtained from max. discharge charge amount

Insulation diagnosis method

Nondestructive electric test Obtained from each test amount of AC, dielectric tangent, and partial discharge

(1) Method for estimation from operation history [1] Remaining life estimation method by NY map

Among the causes of dielectric breakdown, the influences of electric deterioration, heat deterioration, and heat cycle deterioration are particularly large. In actual machines, these causes of deterioration are combined. That is, the dielectric breakdown voltage rate is expressed as a product of each deterioration cause, and can be estimated from the number of start/stop times (N) and operation years (Y). Thus, it is called the “NY map method.” The NY map average is shown in Fig. 3.5.1-14. The minimum value is obtained from (x-3σ). The vertical axis is expressed as the number of equivalent starts/stops. It indicates the influence of changes in load or ineffective electricity on heat cycle fatigue converted to the equivalent number of starts/stops by using the minor rule. The number of equivalent starts/stops NE can be obtained by the following formula:

NE/N10 = (N1/N10) ⋅ (N2/N20) ⋅ (N3/N30)Here, NE: Number of equivalent starts/stops N1: Number of starts/stops N2: Number of load changes N3: Number of ineffective electricity changes N10: Amount of life by starts/stops N20: Amount of life by load change N30: Amount of life by ineffective electricity change

Average of remaining breakdown voltage (%)

Equ

ival

ent s

tart/

stop

tim

es

NB

EB (t

imes

)

Operation hours Y (h) (NY map average)

Fig. 3.5.1-14: Average epoxy insulation life curve [2] Remaining life estimation method by operation condition

The stator coil is sampled from the power generator with the polyester insulation coil that is operated for a long period of time. Based on the investigation result of the dielectric breakdown voltage, the breakdown voltage is obtained from the number of starts/stops and operation hours in the multiple regression formula. In this case, the degree of influence of dielectric breakdown causes are not classified. However, the formula is applied as the life evaluation formula of the stator coil because this is the sampling result of actual coils, and the multiple correlation coefficient is high and the inspection result is significant.

The life curve of the 99.9% reliability lower limit that is obtained from this life evaluation formula is shown in Fig. 3.5.1-15.

340

Remaining withstand voltage T(99.9% reliability lower-limit value)T

Num

ber o

f sta

rts/s

tops

X1

(×10

P3P

tim

es)

Operation hours (×10P4P h) X2 Fig. 3.5.1-15: Relation between remaining withstand voltage and operation conditions

[3] Remaining life estimation method from equivalent operation hours

Sample coils of epoxy resin insulation are collected from machines, and the destructive voltage test is performed. From the result, the following formula to estimate the remaining breakdown voltage (% indication, taking BDV at production to be 100%) using the equivalent operation hours is obtained. The equivalent operation hours are the operation hours calculated equivalently, regarding that one start/stop is equivalent to 20 operation hours.

BDVav = (-4 × 10-5)・YE + 100 BDV3σ = (-6 × 10-10)・YE

2 + (-6 × 10-5)・YE + 77.276 Here, YE: Equivalent operation hours (Equivalent operation hours) = (Operation hours) + 20 × (Number of starts/stops) The relation between the equivalent operation hours and remaining breakdown voltage is shown in Fig. 3.5.1-

16.

Average

Average - 3σ

Average Data rangeAverage - 3σ

Operating hours (h)

Fig.3.5.1-16: Correlation between equivalent operating hours and remaining BDV (2) Estimation method from insulation diagnosis result [1] Estimation by D map

The relation between the discharge parameter ∆ (=∆2+∆I) and the max. discharge charge amount Qmax is obtained from the following experimental formula based on the test result of the model coil and field sampling coil. Here, VR is the remaining breakdown voltage of the epoxy resin. By the AC current test method that detects VR and the average void amount of the insulation layer, current increase rate ∆I is obtained. ∆2 is obtained by the dielectric tangent test.

VR: Remaining rate of breakdown voltage VR = 100 – 1.8 (∆ – 0.8) – 27.4 log (Qmax/1500) This relation is shown in Fig. 3.5.1-17. This is called “D map” (discharge map). It is confirmed that most

data of the relation between the estimated value and measured value are in the 95% reliable area. Comparison between the estimated value VR and measured value Vr by the D map method is shown in Fig. 3.5.1-18. The D map method for polyester insulation is reported in the same way.

341

Remaining breakdown voltage

Dis

char

ge p

aram

eter

∆

(=∆

B2B

+∆B

1B) (

%)

Max. discharge charge amount QBmaxB (pC) : Sampling coil +: Compound deterioration coil

Fig. 3.5.1-17: D map of epoxy insulation

val

ue o

f rem

aini

ng b

reak

dovo

ltage

VB

rB (%

) w

n

95% reliable section

Average by least square

Mea

sure

d

Insulation life control curve

Estimated value of remaining breakdown voltage Vp (%)

: Single polyester insulation coil : Coil of polyester insulation power generator

: Single epoxy insulation coil

Fig. 3.5.1-18 Relation between estimated value Vp and measured value VR by D map method [2] Estimation from max. discharge charge amount

The investigation result of the relation between Q max and breakdown voltage VR is reported. Here, Q max is of the coils sampled from 10 power generators of polyester insulation at the rated voltage. The result is shown in Fig. 3.5.1-19. Both have a good correlation. “Deterioration judgment standard (40% of initial destructive value) = 22000 pC” at the 99% reliability lower limit is proposed.

Measured quantity: n=110

Rem

aini

ng b

reak

dow

n vo

ltage

VB

RB

(%)

99% reliability upper limit

99% reliability lower limit

QBmaxB (Coulomb)

Fig.3.5.1-19: Relation between Qmax and remaining breakdown voltage

The investigation result of the relation between Qmax and VR/E from various power generator coils and sampling coils is reported. This is shown in Fig. 3.5.1-20. A certain correlation can be recognized between Qmax and VR/E. “1×104 pC: Caution needed” and “3×104 pC: NG” are proposed as judgment standards.

342

Max

. dis

char

ge in

tens

ity q

B max

B (p

C)

Withstand insulation required for operation

Withstand insulation (Destructive voltage VBRB/Rated voltage E)

Wire

Coil

Power generator

Model

H: General hydraulic power P: Water pumping T: Turbine P: Polyester series E: Epoxy series

Fig. 3.5.1-20: Relation between Qmax and VR/E

[3] Estimation by the multiple regression method from nondestructive electric test

In the following formula, the statistical correlation between the destructive voltage collected from the sampling coil of the epoxy resin insulation and the insulation diagnosis data is calculated, and the remaining breakdown voltage (% indication, taking BDV at production as 100%) is estimated.

BDV (%) = 91.1 – 0.767 ⋅ (∆I12) – 0.151 ⋅ (∆tanδ12) – 1.78×10-6 ⋅ (qmax12) BDVav (%) = 42 – 29.2 ⋅ ln((BDV – 99.4)/(-56.1)) BDV3σ (%) = BDVav – ((95.3 – 0.395 ⋅ Y) ⋅ (57.9 + 0.474Y – 0.0405 ⋅ Y2)) Here, BDVav (%): Average remaining withstand voltage BDV3σ (%): Remaining breakdown voltage of variance 3σ Y: Operating years The relation between the estimated destructive voltage and measured value is shown in Fig. 3.5.1-21. The

correlation coefficient is as high as 0.61. An example of the relation between the operating years estimated by this estimation method and the breakdown voltage is shown in Fig. 3.5.1-22. This estimation method uses the actual insulation diagnosis data and operating years. The remaining life reflecting the actual dielectric breakdown condition can be estimated.

Est

imat

ed B

DV

by m

ultip

le

regr

essi

on e

quat

ion

(%)

Measured BDV (%)

BDV: Break Down Voltage

Fig. 3.5.1-21: Correlation between measured BDV and estimated BDV

343

Batch average Batch lower limit 3σ

BD

V(%

)

Withstand insulation required for operation: 2E+1 kV

Operation years (years)

Fig. 3.5.1-22: Example of estimation of remaining life 3.2.6 Preventive maintenance to leakage of water-cooling stator coil(10)

The water cooling stator coil has many advantages. Penetration of the cooling media to the insulation layer must be considered for maintenance.

The cooling medium leaks from the brazing section of the pipe for supplying and draining the cooling water and the connecting section between the conductor (wire) and the box (clip) for supplying and draining the cooling water. The structure around the clip is shown in Fig. 3.5.1-23. The former case occurs due to the remaining voids and corrosion during brazing. The latter case occurs when different types of metal (copper, brazing filler metal) make contact with each other in the solvent (cooling water). The metal (copper) with low corrosion potential is corroded (galvanic corrosion). According to the observation and experiment result, it was confirmed that cooling water accumulates in the small voids in the brazing section and water quality in the void worsens. Corrosion progresses over time and a water path is formed, resulting in penetration of the cooling water into the insulation layer.

WireStator core

Absorption

Penetration

Leak

B s

Insulation layerClip

Connection piece

Fig. 3.5.1-23 Struct

If the cooling water penetrates inhygrothermal aging. It is necessarycooling water into the insulation laye

Presence of a leak can be confirmetc. To judge the absorption degreinsulation layer, which focuses on theput into practical use, and is effective

A large differenpermittivity betw

insulation laye(1:20) is

Principle

Capacitance be(copper) and t

layer is measureabsorption

insulation

Fig. 3.5.1-24

razingection

Insulation layer

Wire

UA-A cross-section view Groun

Degradation of ins

Hygrother

Tempe+

ure of clip section and water penetration to insul

to the insulation layer, the insulation characteri to confirm formation of the water path (leak

r. ed by the coil pressure storage test, vacuum stoe of the insulation layer, the method of meas difference in the relative permittivity between t. A schematic diagram of measurement is show

Coilce in relative een the wire r and water used.

Upper coil

Pressing force

tween the coil he insulation d to judge the

degree to layer.

3-phase batc

Lower coilElectrode

Capacitance meter

Measurement method: 100% measurement of turbicollector side, and upper/low

: Principle and method of capacitance measurem

344

d fault

ulation resistance

mal aging

rature

ation layer

stics drastically lower due to path) and penetration of the

rage test, and tracer gas test, uring the capacitance of the he water and the insulator, is n in Fig. 3.5.1-24.

h short circuit

ne side, er coils

ent

There is another method of estimating the remaining life by obtaining the dielectric breakdown speed by hygrothermal aging from the capacitance and destructive voltage of the absorbed insulation layer. As shown in Fig. 3.5.1-25, it is used for examination of the maintenance program. It is important to execute the leak test and capacitor measurement test periodically. It is important to execute the leak test at each periodical inspection and to consider the number of years from the initial water supply in the capacitance measurement test.

Aged deterioration by operation

Insu

latio

n re

sist

ance

leve

l (PU

)

Absorption start point

Capacitance measurement point Insulation resistance lowering rate = αPU/year

Dielectric breakdown by hygrothermal aging

Insulation resistance level required for safe operation

Insulation resistance lowering rate = (Several ~ 10 times αPU/year)

Operation years (year)

Fig. 3.5.1-25: Operation years and estimated insulation resistance level 3.2.7 Wedge for stator coil

Conductive varnish is applied on the straight section to be inserted into the stator core to lower the surface potential. The stator coil is securely fixed in the slot. If the coil is vibrated by the magnetic force in the slot, mechanical deterioration occurs and the corona is discharged between the slot wall and the coil, resulting in insulator deterioration and damage(12).

In order to prevent coil vibration in the slot and fix the coil, a ripple spring is inserted into the coil side as shown in Fig. 3.5.1-26, or a ripple spring is inserted under the wedge as shown in Fig. 3.5.1-27.

Stator wedgeSpacer under wedge

Slide

Ripple spring

Upper coil

Lower coil

Ripple spring

Fig. 3.5.1-26: Side ripple method

Stator wedgeRipple spring

Filler

Upper coil

Lower coil

Fig. 3.5.1-27: Top ripple method

When the tapered bottoms are overlapped and inserted in the axial direction, stress is applied to the slot bottom and the wedge is securely fixed in the slot. There is another structure to fix the wedge by adjusting the filler thickness.

All parts for fixing the stator coil including the wedge are made of insulators. The stator coil is loosened due to vibration or temperature after a long period of operation. If operation is continued in this condition, discharge

345

between the coil surface and the slot wall is accelerated, and the insulation layer is damaged. There is a possibility of a ground fault accident.

Consequently, when the rotor is pulled out, it is necessary to check adhesion of the insulator powder of the wedge to the iron core visually. It is also necessary to check the hitting sound by the test hammer and the deformation of the ripple spring under the wedge. If the wedge does not satisfy the judgment standard, proper measures must be taken.

Power generators are frequently started and stopped recently. The insulators tend to become loose, compared to the base load machines. It is necessary to inspect and control the machines based on the operation condition. 3.3 Diagnosis device

Various diagnosis devices are introduced in order to prevent accidents and large-scale accidents by monitoring and diagnosing the power generator conditions at all times. Details of the diagnosis and monitoring technology are given in the technical report (II) No. 294(5) and others of the Institute of Electrical Engineers of Japan.

An outline is shown below. 3.3.1 Power generator conditioning monitor (GCM)

Particles of 0.1 ~ 0.001 µm that are generated before disassembly due to overheating of insulation materials used for the power generator are converted to electric signals and detected, and local overheating in the power generator is detected.

There are two types of conversion to the electric signals. By one method, hydrogen molecules are converted to plasma ions by α-radiation, and introduced to the electrode. The current is monitored. Particles that are generated by overheating absorb ionized hydrogen molecules. A decrease in current is detected to detect overheating.

In another method, the particles themselves are charged by corona discharge and introduced to the electrode. A minute electric current is detected to know overheating.

The principle of the former method is shown in Fig. 3.5.1-28.

Power generator

Flow of hydrogen gas

Indication, alarm

α radiation source

Ionization room Ion collection room

Fig. 3.5.1-28: Principle of operation of power generator condition monitor (GCM) 3.3.2 Partial discharge monitor

Instead of the current insulation diagnosis, the partial discharge charge amount during operation is always monitored, and deterioration of the stator coil is judged from the measured values and phase characteristics. There are some detection methods. The line is detected by the coupling capacitor, or the radiation wave is detected by the antenna. For the detection of radiation waves, there are two methods. The special pickup coil is used in one method, and the lead wire of the temperature measurement resistor is used instead in the other method. 3.3.3 Stator coil end vibration monitor

If coil end rigidity lowers and comes close to the magnetic vibration frequency (100 Hz, 120 Hz), coil end vibration increases, resulting in fatigue/wear of the coil conductor and damage to the insulator. Consequently, a monitor to check coil end behavior at all times is developed.

Power generator voltage occurs at the coil end. The optical fiber with superior insulation properties is used for measurement wires connected to the vibration pickup. 3.3.4 Rotor wire layer short monitor

If layer short circuit occurs to the rotor wire, the temperature distribution becomes uneven due to unbalanced heat, and the rotor vibrates abnormally. Consequently, it is necessary to detect layer short circuit promptly.

If layer short circuit occurs to the rotor wire, the number of effective turns of the coil decreases. Changes in the gap magnetic flux corresponding to it are detected. The gap magnetic flux is measured using the search coil near the rotor, and the presence of layer short circuit and coils with short circuit are judged. One example of the attachment method of the search coil is shown in Fig. 3.5.1-29. The search coil output waveform is shown in Fig.

346

3.5.1-30. The search coil output voltage corresponding to the coil with the layer short circuit is lower than usual. By comparing waveforms, the presence of layer short circuit and coils with short circuit are judged.

Stator core Stator frame external plate

Rotor

Search coil lead wire

Probe

Search coil Probe details

Lead wire Stainless pipe

Fig. 3.5.1-29: Attachment of search coil

N pole

Layer short circuitN pole

Pulse indicating magnetic pole position

(a) Normal (b) Layer short circuit Fig. 3.5.1-30: Search coil output waveform

In another method, layer short circuit is detected from the tendency of the changes in the rotor wire impedance

due to the turning speed. If layer short circuit occurs, the inductance of the coil decreases. An small AC current is supplied to the rotor wire by a constant current generation device. The rotor voltage is measured, and the rotor impedance at each rpm is measured to judge the presence of layer short circuit. The rotor impedance characteristics are shown in Fig. 3.5.1-31. The impedance characteristics of the coil with layer short circuit change drastically. By comparing characteristics, the presence of layer short circuit is judged(18).

Impe

danc

e (Ω

)

Normal impedance characteristics

Location with drastic change layer short circuit

Rpm

Fig. 3.5.1-31: Rotor impedance characteristics 3.3.5 Axial torsion monitor

The axial torsion vibration in the power generator and turbine shafts is always monitored, and the total fatigue life is calculated and evaluated to prevent deterioration of the shaft and shaft accessories. Gears and magnetic pickup for detecting angular speed are attached to some positions on the shaft. Changes in the obtained angular speed and distortion angle and the modal damping obtained from the detailed model of the shaft are used to calculate the shaft stress waveform and fatigue life consumption. One example of the axial torsion vibration monitoring device is shown in Fig. 3.5.1-32.

347

Average stress Detection of phase difference

Waveform smoothness

Detection of strain amplitude

Peak count method

Damage calculation (Fatigue life

consumption) Static vibration ingredient

Rot

atio

n ch

ange

un

it

Display request 1st

Display device

Monitor TV Oscilloscope Digital printer

5th

Modal conversion constant

multiplication

Del

ay u

nit

Data recorder

Stress waveform

Abnormal stress detection

Power generator current Alarm Additive

synthesis Power generator voltage

Fig. 3.5.1-32: Example of shaft twist vibration monitoring device Monitored pointRotation gap

4. Concept of precision inspection Parts of the power generator consist of various materials such as copper, insulator, aluminum, and iron. The

combination of them is one feature of the parts. The stator becomes loose and wear powder is generated when the coil wedge and coil end insulators are aged

or loose. Inspection and proper correction are required. As shown in the bathtub curve in Fig. 3.5.1-33, the looseness of the stator wedge is classified into the initial, stable, and wear periods. Inspection must be executed according to the period.

Worn periodInitial Stable period

Num

ber o

f loo

sene

d w

edge

s: f

(t)

[Bathtub curve]

Fig. 3.5.1-33: Looseness of stator wedge Operation hours (t)

High centrifugal force functions to the rotor. In addition, there is a problem of wedge corrosion due to overcurrent applied to the rotor surface along with the unbalanced load, etc. The inspection interval must be decided, considering the crack progress speed of the aluminum alloy wedge.

By judging from these phenomena comprehensively, initial and quarterly inspections by pulling out the rotor and inspection by pulling out the retaining ring every 8 years are required. 5. Concept of future maintenance

Operation of the power generator has been changed from the base operation to the middle, WSS, and DSS. Measures to improve durability have been taken for frequent starts/stops. Inspection and countermeasures based on the results of actual operations are required. That is, it is desirable to confirm the soundness of the machine that frequently starts and stops by precision inspection mainly of the rotor.

On the other hand, turning waiting and long-term stop have been applied to oil-burning thermal power machines recently. Power generators have been designed and manufactured mainly for operation. Each unprecedented operation mode needs to be examined sufficiently.

For example, the hydrogen gas in the power generator is not removed by the gas drier while the unit is stopped. The moisture in the machine increases, resulting in dewing and rusting. If the machine stops for a long period of time, rust enters the oil of the lubrication oil system. Because the gap between the seal ring and shaft is very small while the machine is stopped, the potential to catch foreign materials increases, resulting in damage to the seal ring and shaft.

As mentioned above, it is necessary to take proper measures against the failure potential peculiar to each operation mode. Some examples are the gas drier that can remove moisture even while the machine is stopped and the high-performance filter for the sealing oil systems.

348

3.4.5.2 High-pressure motor In thermal power plants, high-pressure motors are used for various devices for driving accessories. High-

pressure motors are sometimes one of the important devices in a plant. To secure the reliability of the systems of thermal power stations, preventive maintenance including periodical maintenance, exchange of parts, and estimation of remaining life by deterioration judgment are performed for the motors of accessories.

Here, the high-pressure basket-type inductive motor is taken as an example, and an outline of preventive maintenance and remaining life diagnosis is given. 1. Structure and deterioration form of each section 1.1 Structure

A cross-section view of the basket-type motor is shown in Fig. 3.5.1-34. The motor consists of three sections: the fixed section, rotation section, and bearing section.

Bearing bracketBearing Heat exchanger

Fig. 3.5.1-34: Cross-section view of basket-type motor

Rotor wire Shaft Internal fan External fan

Stator frame Stator core Rotor core Stator wire

1.2 Failure conditions

The fixed section consists of the stator wire and iron core. The rotation section includes the basket-type rotor conductor. The structural members include the bearing. According to the failure investigations(1) in the past, 50% of failures occurred in the fixed section, 12% occurred in the rotation section, and 38% occurred in the bearing and others.

Concerning the relation between the operation years and the number of failures, failure mostly occurs after 10~15 years of operation. The total failure rate is almost constant between 5 and 20 years of operation, but the failure rate increases after 21 years.

Each cause of the total failure rate is shown in Fig. 3.5.1-35. Failure of the bearing section is often detected before 20 years of operation. After 20 years, dielectric breakdown is often detected.

Cumulative failure rate of bearing (Number of failures: 34) Cumulative failure rate caused by dielectric breakdown (Number of failures: 29) Failures that are judged to be aged deterioration among failures caused by dielectric breakdown (Number of failures: 17)

Failu

re ra

te (%

)

Fig. 3.5.1-35: Transition in cumulative failure rate of each failure of 3 kV basket-type motor

Note) The failure is located at the stator and the coil end.

1.3 Control items, deterioration form, and diagnosis method of major sections

Deterioration of major sections of the motor is classified as follows: [1] Thermal cause [2] Electric cause [3] Physical cause [4] Mechanical cause [5] Chemical cause Deterioration is accelerated by a combination of these causes, and failure occurs. The causes of deterioration

are shown in Fig. 3.5.1-36. Examples of control items, deterioration form, and diagnosis method are shown in Table 3.5.1-5.

349

Fig. 3.5.1-36: Deterioration cause analysis

Electric cause Thermal cause

Surge Partial discharge

Heat cycle Overload Cooling failure Thermal deterioration

Swelling, contraction, deformation,

distortion

AbsorptionDewing Dust

Degradation of lubrication oil

Surface insulation

failure

Start, stop, vibration, shock, overcurrent

Fatigue, wear,

deformation, distortion

Mechanical cause Chemical cause

Corrosion by harmful substanceChemicals Oil Degradation

of insulation

Deterioration

Physical cause

Table 3.5.1-5: Control items, deterioration form, and diagnosis method of major sections

No. Control locations

Control items Deterioration form Diagnosis method

Pollution Dust accumulation, corrosion, damage • Visual Discoloration Paint discoloration, peeling, etc. • Visual 1 Appearance

Loosening of each part Screw loosening, backlash, etc. • Visual, hitting Sound Looseness, backlash, bearing failure • Sound and frequency analysis Vibration Mechanical unbalance, layer, wire breakage • Vibration measurement, frequency analysisOdor Degradation of insulation characteristics,

burning of bearing, lubrication failure • Odor judgment

Smoke Degradation of insulation characteristics, lubrication failure, overheating of bearing

• Visual

Voltage Layer, wire breakage • Voltage monitoring Current Pulsation of current by cutting of basket-type

rotor conductor • Current monitoring

2 Operation condition

Temperature Bearing lubrication failure, dust accumulation on cooling passage, degradation of cooler characteristics

• Thermometer

Lubrication oil Oxidation, lowering of viscosity, mixing of water or foreign materials

• X-ray fluorescent analysis

Oil ring Deformation • Visual Oil scraper Wear • Visual

Slide bearing

White metal Peeling, cracks, wear, abnormal touching • Penetration flaw detection, Dimension measurement

Vibration Damage to inner ring, outer ring, or ball • Bearing diagnosis device Abnormal noise Damage to inner ring, outer ring, or ball • Hearing Rim of outer ring Flaw and discoloration • Visual

3

Rel

ated

to b

earin

g

Roll bearing

Grease Separation, discoloration • Visual Duct piece Loosening • Visual Iron core Loosening • Visual, touching, hitting Brazing section Crack • Bar breakage diagnosis device, penetration

flaw detection Short-circuit ring Deformation • Visual

4 Rotor

Rotor conductor Loosening • Visual, hitting Duct piece Loosening • Visual Iron core Loosening • Visual, touching, hitting

Insulation section Discoloration, breakage, corrosion, degradation of electrical characteristics

• Nondestructive electric insulation diagnosis, physical/chemical diagnosis

5 Stator

Wedge Loosening, discoloration, breakage • Visual, touching, hitting

It is understood that daily inspection is very important for early detection and countermeasures against

deterioration. Examples of the standard inspection intervals are shown in Table 3.5.1-6.

Table 3.5.1-6: Standard inspection interval of motor Initial inspection 1st continuance inspection Continuance inspection

350

2,000 ~ 3,000 operation hours Within 1 year max.

Approx. 8,000 operation hours/starting 500 times Within 2 years max.

Approx. 8,000 operation hours/starting 500 times Within 2 years max.

The rotor is not pulled out if there is no failure when the shield is (partially) removed.

Pulling out of rotor

Inspection of coil end • Loosening of spacer • Loosening of string

Loosening of tightening bolts • Rotor frame • Bearing • Terminal

Investigation of dust and stains

Inspection of lubrication oil Starting time, vibration

In addition to initial inspection items; Iron core inspection • Loosening of iron core • Loosening of duct piece or finger

Inspection of stator wire • Loosening of wedge

Inspection of rotor wire • Bar or short-circuit ring • Brazing section

Slip bearing • Peeling, cracks, wear, or hitting of white

metal • Oil ring

The degree of inspection depends on the presence of problems at a previous inspection and during operation. After approx. 8 years, pull out the rotor again for detailed inspection.

2. Stator wire deterioration diagnosis technology and life estimation 2.1 Transition in stator wire insulation system

High-pressure motor wires have been changed from asphalt compound insulation used before the middle of 1960s to resin-rich insulation using polyester resin and epoxy resin and so-called resin insulation of total impregnation (vacuum pressure impregnation) insulation. 2.2 Deterioration mechanism of stator wire insulation

The stator wire insulation receives various kinds of deterioration stress during operation. Deterioration stress can be classified as shown in Fig. 3.5.1-36. These causes do not function individually, but are mixed and accelerate deterioration.

The deterioration phenomena by these factors cause dielectric breakdown and lowering of insulation performance, resulting in dielectric breakdown. However, deterioration stress depends on the motor type (full-closed type, open type) and voltage level, and the deterioration mechanism is often complicated. 2.3 Dielectric breakdown diagnosis method of stator wire insulation

The dielectric breakdown of high-pressure motor wires has been diagnosed from long ago. Electric diagnosis is performed in general.

The physical/chemical method of diagnosing the thermal deterioration degree of the insulation materials used for the motor wire was developed and put into practical use. 2.3.1 Electric dielectric breakdown diagnosis method and judgment standard (1) Electric dielectric breakdown diagnosis method(3)

The insulation condition of the motor wire is tested to prevent wire accidents and to make a preventive maintenance plan in the future. [1] DC absorption test

In the DC absorption test, the absorption current phenomena when DC high voltage is applied to the wire are checked to see the insulation property. The insulation resistance (1-minute value), polar index (PI), and kick current are investigated. [2] Dielectric tangent (tanδ) test

In the tan δ test, the tanδ value and tanδ- voltage characteristics when AC voltage is applied to the wire insulation are checked to see the wire insulation property and obtain the tanδ0 and ∆tanδ characteristics value. [3] AC test

In the AC test, the relation between the current (I) and the voltage (V) when AC voltage is applied to the insulator, that is, the I-V characteristics, are checked to see the insulation property. The characteristics values of ∆I (current increase rate) and Pi (current rapid increase voltage) are obtained.

351

[4] Partial discharge test If any voids exist in the insulator, partial discharge occurs at the void when AC voltage is applied. This

discharge is called “partial discharge.” In this partial discharge test, insulation properties including wire insulation voids and partial insulation defects are investigated, and Qmax (maximum partial discharge charge amount) is obtained. (2) Dielectric breakdown judgment standard

The wire insulation characteristics differ depending on the insulation method. It is necessary to evaluate the insulation characteristics obtained by the insulation diagnosis test by this insulation method.

Asphalt compound insulation adopts the deterioration judgment standard of the Central Research Institute of the Electric Power Industry. As mentioned above, the materials and production method of the resin insulation depend on the motor maker. It is difficult to unify the deterioration judgment standard at the present time. Each maker sets its original judgment standard. An example is shown in Table 3.5.1-7(3) (4).

Table 3.5.1-7: Example of dielectric breakdown judgment standard of high-pressure rotor wire Column No. I II III IV V

Insulation method Compound Compound Varnish Resin Resin

Application range Rated

voltage 3.3 kV 6.6 kV 3.3 kV 3.3~4.4 kV 6.6~11 kV 3.3~11 kV

Judgment OK OK OK Caution needed NG OK OK OK R [MΩ] ~100~10~ >E+1

RC [Ω⋅F] >10 log (Rd/Rw) <3 P.I >1.5 >1.5 ~1.5~1.0~ >2.0 tanδ0 [%] ~20~30~ <10

at 1.25E/ 3 <0.9 ∆tanδ [%]

at E <0.7 <6.5 ~0.7~1.1~ ∆tanδ+∆C/C0 [%] at E <12

P11 >E pi [kV]

P12 >3.3 >6.6 ~4.6~3.0~

at 1.25E/ 3 <2.5 ∆ [%]

at E <4.0 <8.5 ~4.0~5.9~ q>500 pC ~1.9~1.3~

Vi [kV] q>1000 pC >E

N [Pcs/half cycle] at E, q>500 pC

~50~100~

at E/ 3 <10000

at 4.5 kV <10000 qm [pC]

at E <5000 ~1000~1400~ 2.3.2 Physical/chemical thermal deterioration diagnosis method

Electric insulation diagnosis of the high-pressure motor wire is mainly applied to the main insulation (slot insulation) of the stator wire. The stator wire consists of wire-reinforcing materials such as the wedge that mechanically supports the wire, the spacer between coils, and the string that fixes them. If these reinforcing materials are loosened or their mechanical strength is lowered, the upper and lower coils vibrate in the slot and the coil end touches the spacer and wearing occurs. It is estimated that the insulation layer is mechanically damaged. In fact, this damage once led to dielectric breakdown. Deterioration of the reinforcing materials cannot be detected by electric diagnosis.

The physical/chemical thermal deterioration diagnosis method was developed to diagnose deterioration of reinforcing materials. If it is combined with electric insulation diagnosis, the degree of deterioration of the entire wire can be comprehensively evaluated.

(1) Principle of physical/chemical heat deterioration diagnosis method(6)

Accelerated thermal deterioration is executed for the insulation-reinforcing materials used for stator wire preliminarily to measure physical characteristics such as lowering of mechanical strength. Next, the same

352

sample is used to obtain some heat analysis values using the heat analysis device. The correlation between the physical characteristic values and heat analysis value is examined to obtain the

best relation. This is called the master curve. It is obtained for each insulation material. The master curve is the principle of this test method. An example of a master curve is shown in Fig. 3.5.1-37.

Synthetic resin glass lamination plate

TG 3

/1 re

duct

ion

rate

Chemical characteristics of sampled material

Estimated physical characteristics

Fig. 3.5.1-37: Example of master curve

Bending strength holding rate (%)

The heat analysis value is obtained from the materials that are sampled from the machine by the heat analysis

test. It is applied to the master curve of the material to estimate the physical characteristic value. (See Fig. 3.5.1-37.) (2) Judgment of heat analysis

The characteristic value obtained by heat analysis is applied to the master curve to obtain the estimated physical mechanical strength. The deterioration judgment in this case is based on the deterioration judgment standard in Table 3.5.1-8. The deterioration judgment standard is based on IEC216-2.

Table 3.5.1-8: Deterioration judgment standard Mechanical

strength Loss of weight by heating Deterioration

degree Lamination plate, etc. Resin Varnish

Judgment

Small Over 75% Under 2.5% Under 10% Good: Deterioration progress is small. Operation is possible with no difficulty.

Medium 50~75% 2.5~5.0% 10~20% Acceptable: Deterioration progresses, but operation is possible with no difficulty at this moment.

Large Under 50% Over 5.0% Over 20% Unacceptable: Deterioration progresses. Updating is required.

Note 1. The mechanical strength is the strength-holding rate. The initial condition is regarded to be 100%. 2. Judgment is made based on IEC216-2. 2.4 Life estimation method

Insulation resistance of 2 E+1 kV that is required to continue stable operation in the future is judged from the characteristic value by insulation diagnosis.

The method of estimating the remaining destruction voltage and remaining life from the insulation diagnosis is partially used, but its reliability is not sufficient.

However, if the data of insulation diagnosis characteristics values and destructive voltage by abundant samples and sampled coils can be accumulated, it is expected that the remaining destructive voltage and remaining life can be estimated with a high degree of accuracy. 3. Diagnosis method of other sections 3.1 Deterioration diagnosis of basket-type rotor

A large current is applied to the conductor and short-circuit ring of the basket-type rotor when the operation is started. The combined stress of thermal stress by heating, stress by conductor vibration due to magnetic force, and the centrifugal force of rotation functions.

Due to its structure, the stress becomes largest near the silver brazing section of the connection section of the discontinuous conductor and short-circuit ring.

If a plant starts and stops frequently for operation of the plant, in particular, repeated stress is applied and there is a possibility of leading to the fatigue breakdown. Because high stress occurs in high-inertial-load machines and high-speed machines, adequate attention must be paid to them.

In order to detect failure of conductors such as the silver brazing section promptly, a rotor bar breakage monitor or the like is used.

A bar breakage monitor captures the pulsation of the load current from the motor whose bar is broken and

353

detects errors from the frequency. A bar looseness measurement device detects looseness by hitting. The measured material is hit by a hammer

and the looseness is detected from the rate of the vibration response to the hitting strength. The vibration response of a loose bar is larger than that of a fixed bar, and the rate is larger. Looseness can be judged objectively. 3.2 Bearing 3.2.1 Roll bearing diagnosis machine

The bearing is an important part of the rotating machine. As mentioned before, the failure rate is rather high. It is very important to detect the failure before fatal errors occur.

The life of the roll bearing is particularly short among structural parts of the motor. The life is defined with 99% reliability. The life is mainly judged by the occurrence of flaking. Flaking is fatigue breakdown on the surface caused by repeated stress on the rolling contact section. Due to this fatigue, the surface peels off in flakes.

To detect this kind of abnormal phenomena, bearing vibration is detected by acceleration and enveloped. By this method, according to the enveloped vibration acceleration data, the calculation circuit that is weighed according to the failure type is passed, and the presence of failure is numerically converted to facilitate judgment. 3.2.2 Diagnosis of slip bearing The life of the slip bearing is longer compared to the roll bearing. However, due to various causes,

temperature increase, wear, white metal fatigue, peeling, and cracks occur, leading to the life of the bearing diminishing.

By the thermometer element that is buried in the bearing, the bearing temperature during operation is monitored throughout the year. The measurement result is used for tendency control. If any temperature out of the tendency is detected, it must be noted because the lubrication surface might have some errors.

Investigation of the lubrication oil is one of the methods to detect bearing failure. By sampling a small amount of lubrication oil from the bearing for fluorescent X-ray analysis, the amount of metal materials that form the white metal ingredients in the lubrication oil can be clarified. If this investigation is performed periodically, the tendency is controlled to judge the presence of failure. Afterword

An outline of the preventive maintenance and remaining life diagnosis of the high-pressure motor is introduced here.

It is desirable to continue a close relationship between makers and users for future development and improvement of these technologies.

354

3.5.1.3 Electric facilities in the plant Many thermal power plants in Japan were built between the 1950s and 1970s. More than 70% of units have

been operated for 15 years or more. The number of plants used for 15 years or more will increase in the future. In addition, for stable supply of electricity and economical merits of extension of life, recent themes of study are to diagnose the aged facilities properly and effectively execute life control and preventive maintenance based on the diagnosis result.

From this point of view, as a typical device of the electric facilities of thermal power plants, an outline of remaining life diagnosis and preventive maintenance of the static devices (including the main transformer, transformer in the plant) and high-/low-pressure switch gear (including the metal clad, power center, and control center) is given here. 1. Transformer

Transformers in thermal power plants include the main transformer for increasing the power generator voltage, starting transformer for supplying the plant-starting power source, and the transformer in the plant for supplying the power source during normal operation. Based on the concept in Table 3.5.1-9, daily inspection and periodical analysis investigation of the insulation oil-dissolved gas are performed for these transformers. In addition, internal detailed inspection and overhaul are performed at the time of exchange of parts with relatively short life and at periodical inspection of the plant to secure reliability(1)-(7).

Table 3.5.1-9: Maintenance and overhaul of transformer Inspection cycle and updating schedule Remarks

Oil-sealing section All gaskets (including bushings) must be replaced approx. every 15 years.

Mechanical protection relay

Overhaul must be performed once every few years. Major protection relays (oil temperature gauge of the main body, shock oil pressure Ry, etc.) need to be replaced after approx. 15 years.

Gauges Overhaul is performed approx. every 15 years, and gauges are updated depending on the result.

The life is shorter than the main body. It is necessary to update the units at least once before the end of the life of the main body.

Oil-cooling device All devices are replaced if the bearing issues abnormal noise after 5~10 years.

Periodical cleaning of cooling pipes is required depending on the environmental conditions.

Tap-switching device for load The device needs inspection every 50,000~100,000 times of switching.

Electric life: 200,000 times Mechanical life: 800,000 times

Paint The cycle is decided depending on the rust condition in the past. Rust (life) depends on the environmental condition.

Inside of main body The main body is inspected by gas analysis of the oil once every 6 months or 1 year. It is decided whether the internal inspection is required depending on the result.

The expected life is 30 years in general.

Along with the increase in the number of devices that have been used for 20~30 years, it is important to

execute deterioration diagnosis of oil-immersed transformers and estimate the remaining life from the view points of effective use of devices and securing reliability. The oil-immersed gas analysis is widely used to diagnose device failure. The method is widely known as the “Electric Technology Research Association method(1) (5).” Here, we focus on remaining life diagnosis mainly by deterioration. 1.1 Dominant reason for life of transformer

Materials whose characteristics lower due to aged deterioration in the transformer are the insulation oil and insulator. The general concept of the life of the transformer related to the deterioration of these materials is shown below. 1.1.1 Insulation oil

The most important characteristics of the insulation oil are the dielectric breakdown voltage. Lowering by deterioration is small in general. The dielectric breakdown voltage of insulation oil lowers when dissolved water in the oil increases for some reason in most of the cases. If the dielectric breakdown voltage or other characteristics of the insulation oil are degraded, it is specified to exchange the oil with new oil or take deaeration filtering measures following the control values in Table 3.5.1-10 defined in the maintenance control guideline of the insulation oil. By taking these measures, characteristics of the insulation oil do not cause any problems in the operation of the transformer. Consequently, it is regarded that degradation of characteristics of the insulation oil does not affect the life of the transformer.

355

Table 3.5.1-10: Maintenance control value of insulation oil after starting operation Voltage level 11~77 kV 110~275 kV ≥500 kV Tap switcher

Standard value Test frequency Standard value Test frequency Standard value Test frequency Standard value Test frequency

<40 Once/3 years <30 Once/3 years <20 Once/3 years ⎯

40~50 Once/year 30~50 Once/year 20~30 Once/year ⎯ Water (ppm)

>50 Countermeasures >40 Countermeasures >30 Countermeasures ⎯

>30 Once/3 years >40 Once/3 years >50 Once/3 years >20

⎯ ⎯ 30~40 Once/6 months 40~50 Once/6 months ⎯

Dielectric breakdown voltage (kV) <30 Countermeasures <30 Countermeasures <40 Countermeasures ≤20

>1×1012 Once/3 years >1×1012 Once/3 years >5×1012 Once/3 years ⎯

>1×1011~>1×1012 Once/year >1×1011~>1×1012 Once/year >1×1011~>5×1012 Once/year ⎯

Volume resistivity (Ωcm) (80°) <1×1011 Countermeasures <1×1011 Countermeasures <1×1011 Countermeasures ⎯

<0.2 Once/3 years <0.1 Once/3 years <0.1 Once/3 years ⎯

0.2~0.5 Once/year 0.1~0.5 Once/year 0.1~.5 Once/year ⎯ Total oxidation (mgKOH/g)

>0.5 Countermeasures >0.5 Countermeasures >0.5 Countermeasures ⎯

Once every 3 years or once every 50,000 times of operation if a hot-line washer is provided; once every 20,000 times of switching operation if it is not provided. Take measures if the voltage is 20 kV or less.

1.1.2 Insulator

The temperature of the insulating paper that is wound on the wire is the highest. The insulating paper tends to be affected by degradation of characteristics caused by deterioration. Oil-immersed dielectric breakdown strength, which is an important characteristic of the insulating paper, is not lowered by deterioration, in the same way as the insulation oil. No problems normally occur even after the long-term operation of the transformer. Tensile strength, which is another important characteristic of the insulating paper, is degraded due to deterioration, and problems might occur when operation is continued. For example, when a system ground fault accident (2-wire ground fault, etc.) occurs, external ground-fault or short-circuit current is applied. The tensile stress functions to the wire coating due to the magnetic mechanical strength that occurs on this occasion. When the strength of the insulating paper lowers below the strength, the insulating paper is torn or broken. This is the end of the life of the insulating paper. It is in fact impossible to exchange the wire insulating paper of the transformer. If the transformer is continuously used, the wire must be exchanged.

According to the above, the life of the transformer depends on the tensile strength of the insulating paper that is wound on the wire. 1.2 Deterioration diagnosis

The life of the transformer depends on the life of the insulating paper of the wire. Diagnosis of the life of the insulating paper is the deterioration diagnosis of the transformer. As the insulating paper used for the wire is wound on the conductor, it is folded. It might be difficult to measure the correct tensile strength. The average degree of polymerization that is closely related to the tensile strength is used as the basic scale of deterioration. Figure 3.5.1-38 indicates the changes in the remaining rate of average degree of polymerization of the insulating paper used for the transformer by the operation years. The average degree of polymerization lowers(8) along with the transformer operation years.

Rem

aini

ng ra

te o

f ave

rage

de

gree

of p

olym

eriz

atio

n (%

)

Operation years (year) Fig. 3.5.1-38: Relation between remaining rate of average degree of polymerization and transformer operation

years 1.2.1 Diagnosis standard

It has been general to measure the average degree of polymerization for measuring the deterioration degree of 356

the insulating paper(5) (7). The average degree of polymerization indicates the length of the molecule that constitutes the insulating paper. If the paper material is deteriorated, the average degree of polymerization becomes smaller. The initial average degree of polymerization of the insulating paper of the wire is around 1000.

In JEM1463-1993, the following evaluation standards of the average degree of polymerization of the insulating paper for transformers are defined:

Life level: 450 or less Dangerous level: 250 or less

1.2.2 Deterioration product material and deterioration index ingredient

Because the insulating paper in the operating transformer and characteristics cannot be measured, it is impossible to directly check the degree of deterioration of the insulating paper. The deterioration product material that is closely related to deterioration of the insulating material is found, and the aged deterioration of the insulation materials is obtained from the deterioration product material. That is, deterioration is externally diagnosed.

The generation amount of CO2+CO that is dissolved in the oil and furfural is analyzed as an effective item of the deterioration diagnosis index of large-capacity oil-immersed transformers.

Figure 3.5.1-39(10) indicates the relation between the CO2+CO generation amount and the average degree of polymerization. Figure 3.5.1-40(11) indicates the relation between the furfural generation amount and the average degree of polymerization. As shown in the figures, we can see that the CO2+CO generation amount and furfural generation amount are closely related to the average degree of polymerization of the insulating paper. Consequently, the tensile strength and average degree of polymerization can be directly known by measuring the CO2+CO generation amount and furfural generation amount in the insulating oil of the transformer. That is, the aged deterioration of the transformer can be externally diagnosed.

: In the case of oxygen addition: In the case of water addition

×: If oxygen or water is not added

Rem

aini

ng ra

te o

f ave

rage

deg

ree

of

poly

mer

izat

ion

(%)

COB2B + CO generation amount (ml/g)

Fig. 3.5.1-39: Relation between remaining rate of average degree of polymerization of insulating paper and CO2+CO generation amount

Paper weight (g)/Oil amount (ml) Condition

3% 10% No addition

Oxygen addition Water addition

Furf

ural

gen

erat

ion

amou

nt (m

g/g)

Remaining rate of average degree of polymerization (%)

Fig. 3.5.1-40: Relation between remaining rate of average degree of polymerization of insulating paper and

furfural generation amount 1.3 Life extension measures

Extension of the life of the cooling device, various measurement devices, and relays can be expected by periodical maintenance. Yet, the life is about 10~15 years. If they are used for a long period of time, it is

357

desirable to replace them with new ones. On the other hand, if the insulating paper comes to the end of its life, the wire needs to be replaced. In this

case, the major structural parts of the tank and iron core can be reused. In the case of important transformers with a high utilization ratio and without a spare device like a major transformer, it is necessary to change it at a proper timing in the plant life cycle in a well-planned manner. 2. Switchgear on switchboard

Switchgears on the switchboard (hereafter called “switchgears”) are classified into two: the metal-clad switchgear that has a 6 kV-class magnetic circuit breaker, SF6 gas breaker or vacuum breaker, and the power center that has a 600 V-class air circuit breaker.

The switchgear consists of structural parts and control accessories. Structural parts include the breaker, protection relay, measurement transformer, and bus bar. Control accessories include the lamp, fuse, auxiliary relay, and timer.

For extension of the life of equipment and evaluation of soundness, it is important to take proper measures for the major devices and insulators by remaining life diagnosis in the appropriate time. Concerning the control accessories, it is rational to update them in a well-planned manner, referring to the estimated usable years.

Remaining life diagnosis technology and preventive maintenance of major structural devices and parts of the switchgear are introduced below. 2.1 Remaining life diagnosis technology

Remaining life diagnosis technology (deterioration diagnosis and life evaluation method) of the switchgear can be classified into two: by the [Soundness evaluation method], it is judged whether the switchgear maintains the specified performance at each point in time, and by the [Remaining life evaluation method], the remaining life is evaluated by evaluating the field products quantitatively.

It is practical to use both methods to evaluate the switchgear life. An example of the remaining life evaluation method of the structural parts of the switchgear is shown in Table 3.5.1-11. Each item is explained below. 2.1.1 Soundness evaluation

Whether the field switchgear and its structural devices, parts, and accessories function normally is investigated and evaluated. If there is no failure as a result of diagnosis, it is judged that they can be used without problems. Lubrication items are added to this judgment method, and this is the periodical inspection that is widely performed. (1) Visual inspection (VI)

Parts and accessories are inspected visually to see visual defects (damage, pollution, and discoloration). Dimension measurement might be included. (2) Operation test

Structural devices are actually operated to check functions. The lubrication condition of the operation structural sections is also judged at the same time. A characteristics test is performed in some cases.

358

Table 3.5.1-11: Life evaluation method of major structural parts of switchgear Life evaluation method Switchgear

types Major structural

parts Failure mode Cause of deterioration

Soundness evaluation Nondestructive Destructive

Insulation bus wire


Temperature VI Insulation resistance measurement

Partial discharge characteristics

Withstand voltage limit test

Bus wire support Dielectric breakdown

Humidity, pollution

VI Insulation resistance measurement

Measurement of pollution degree

Material characteristics investigation

PT, CT Dielectric breakdown

Humidity, corrosion

VI Partial discharge characteristics

Material characteristics investigation

Elevating structure

Operating failure Wear VI Operation test

⎯ ⎯

Auxiliary switch Operating failure Temperature, humidity, pollution

VI Conductivity check

Contact resistance measurement

⎯

Control line Operating failure Temperature VI ⎯ Material characteristics investigation

MBB bushing


Temperature, humidity

VI Insulation resistance measurement


Withstand voltage limit test Material characteristics investigation

MBB arc shoot

Breaker malfunction

Humidity, wear VI Insulation resistance measurement

⎯ Actual interrupting test

MBB operating structure

Operating failure Wear, fatigue VI Operation test

⎯ Shock life characteristics

Met

al c

lad

Operating motor Operating failure Wear Insulation resistance, VI, operation test

Operating characteristics

Withstand voltage limit test

Main circuit bus wire

Damage Temperature VI ⎯ Material characteristics investigation

Insulation support

Damage Insulation failure

Temperature VI Insulation resistance measurement

⎯ Withstand voltage limit test Material characteristics investigation

Main circuit disconnecting section

Damage Superheating

Wear Heat deterioration

VI Conductivity check

⎯ Thermal cycle test

Auxiliary relay Operating failure Wear, corrosion VI Conductivity check/Insulation resistance measurement


Accelerated deterioration test

ACB Pull-out structure

Damage Operating failure

Wear VI Operation check

Operation test ⎯

ACB Insulation base

Damage Operating failure

Temperature, fatigue

VI Operation check

⎯ Shock life characteristics Material characteristics investigation

ACB Arc shoot

Breaker malfunction

Humidity, wear VI Insulation resistance measurement

⎯ Actual interrupting test

ACB Overcurrent trip device

Operating failure Wear VI Operation test

Operation characteristics test

⎯

Pow

er c

ente

r

ACB Loading trip coil

Operating failure Temperature, humidity, corrosion

VI Resistance measurement

Layer test Accelerated deterioration test

VI: Visual inspection

2.1.2 Remaining life evaluation(12)

Characteristics of devices, parts, and accessories are measured and analyzed or evaluated by the acceleration test, etc. to estimate the remaining life. The evaluation method is classified into two: the destructive method, and

359

the nondestructive method. Typical evaluation methods are explained below.

(1) Partial discharge test(13)

The long-term electric destruction of insulators occurs due to partial discharge deterioration or tracking deterioration. There are two forms of partial discharge deterioration. One is internal partial discharge deterioration that is caused by centralized electrolysis at peeled or cracked sections due to foreign materials or heat stress. The other is surface partial discharge deterioration caused near the contact surface between the metal electrode and insulators related to the ambient environment and pollution on the insulator surface. If these partial discharges continue, tracking gradually progresses, resulting in dielectric breakdown. If the partial discharge start voltage of the insulator is sufficiently higher than the normal application voltage, partial discharge does not occur and there is no problem. If the partial discharge start voltage lowers due to aged deterioration and discharge is started at the normal voltage, or if overvoltage is tentatively applied due to some causes even if the partial discharge start voltage is high, the partial discharge continues when the partial discharge extinction voltage is lower than the normal voltage. In this case, dielectric breakdown rapidly progresses.

The partial discharge diagnosis method by AE measurement is used to detect and diagnose the partial discharge. Figure 3.5.1-41 shows the structural diagram of the partial discharge test.

Preventive maintenance of power-generating facilities and remaining life diagnosis

Applied voltage

AE amplifierAE sensor D

igita

l osc

illos

cope

or P

Cm

easu

rem

ent s

yste

m

Cou

plin

gca

paci

tor

RF output

Electrode

MBB or VBC (Housing)

Partial discharge

measurement device

Fig. 3.5.1-41: Structural diagram of partial discharge test (2) Shock life characteristics (Insulation structural material)

Both electric insulation resistance and mechanical strength as structural material are required for the insulation structural materials such as the breaker bushing. These high-molecular insulation materials have deteriorated elements by which the shock life characteristics degrade due to the deterioration characteristics of aged materials.

Figure 3.5.1-42 shows an example of the deterioration characteristics. Here, the shock life characteristics of the sampled breaker bushing phenol resin insulation (laminated materials) and shock life characteristics of new parts are measured. The characteristics reduction conditions are checked to estimate the approximate life of the standard strength. The operating stress in Fig. 3.5.1-42 is the addition of stress applied when the breaker is opened or closed and internal stress on the insulators. According to the investigation result, this value tends to increase over time as shown in the graph. This is caused by the deterioration of the insulator. According to the comparison between the shock fatigue stress of the insulator and operating stress, the life is about 15~20 years.

20 y

ears

Bre

ak s

treng

th (k

g/m

mP2P

) New product 700 h

18 y

ears

15 y

ears

Shock fatigue strengthOperating stress (Stress at switching + Internal stress)

Elapsed time (g) Figure 3.5.1-42 The change of the strength of MBB phenol bushing over time

(3) Accelerated deterioration test The aged deterioration of insulators is accelerated by the heating conditions, and equivalent evaluation is

360

performed by the check in a short period of time. An example of the characteristics of tensile strength to the same insulator is shown in Fig. 3.5.1-43. The parameters are “T” (heating temperature) and “t” (heating time). If the used temperature can be seen from the characteristics, the life of the standard strength can be estimated.

→

Bre

ak s

treng

th

(kg/

mm

P2P

)

Standard strength

Tem

pera

ture

→ Time

Fig. 3.5.1-43: Accelerated deterioration characteristics of insulators (4) Material physical property investigation

As investigation of the insulator deterioration, the characteristics test of the static tensile strength, bending strength, and elongation rate are also effective. The static characteristic specifications of insulation materials are initially known in many cases. They can be used as a material for deterioration judgment by the characteristic investigation of samples. (5) Contact resistance measurement(14)

The switchgear uses various auxiliary switching devices. There are many contacts, and disassembly inspection, cleaning, or maintenance are not actually performed. Consequently, the contact resistance increases over time, and reliability problems occur.

Typical items of remaining life diagnosis technology are explained above. Deterioration diagnosis of insulators is executed to some of devices and parts of the switchgear. By establishment of this technology and the diagnosis method of accessories in the panel including the relays, it is regarded that the remaining life of the switchgear is quantitatively evaluated. 2.2 Preventive maintenance

Preventive maintenance methods of the switchgear are classified into two: one is the soundness evaluation by periodical inspection of housing devices and accessories including breakers, and the other is the replacement by updating the aged switchgear itself or major housing devices including breakers. 2.2.1 Parts exchange by periodical inspection

Soundness is evaluated by periodical inspection. Parts with relatively short life are exchanged. Parts are also exchanged by horizontal development of problems.

Periodical inspection of switchgears has a long track record both for makers and users. Parts to be exchanged at periodical exchange might be listed. It is important to update parts and reflect improvement of field failure.

It is important to estimate the life of major devices to take measures against the increase in aged deterioration and for extension of life. It is necessary to update applicable parts and devices in a well-planned manner by predicting structural parts and parts to be exchanged by the above-mentioned deterioration diagnosis, using the aged switchgears as samples. 3. Preventive maintenance of control center

The basis of the preventive maintenance measures of the control center is to exchange the devices and parts with failure or that have almost come to the end of their life. The range of exchange is classified into three as shown in Table 3.5.1-12.

Table 3.5.1-12: Comparison of control center durability improvement measures Durability improvement measures Contents Exchange of parts Only the parts of the magnetic contactors or magnetic relays with relatively short life are

updated. Exchange of units The deterioration is in progress generally with devices or wires in the units, for example. It

is necessary to check that bus wires and insulation wires in the panel are in good condition. Update of panel The bus wires and insulation wires in the panel are deteriorated, but are planned to be used for

more 5~10 years.

361

To select the range, future operating hours, economic efficiency, and time required for updating need to be comprehensively compared. In the case of parts exchange, if the control center type is old, it might be difficult to acquire parts because of model change by makers or it takes a long time for exchange. It might sometimes be necessary to remodel existing panels to use a new incompatible part. Unit exchange might be more economical than parts exchange. Afterword

Many electric facilities in plants have been used for 25 years or more. Reliability has been maintained by periodical inspection, exchange of parts with short life, or horizontal development of nonconformity. In order to use these aged facilities for a longer period in the future, it is strongly desirable to establish life diagnosis and evaluation technology of wires that are the major structural parts, like transformers.

For this purpose, cooperation between the maker and users is effective. We need the understanding and cooperation of related parties.

362

3.5.2 Corrosion of power generators and countermeasures Power generators can be classified by their cooling method: air cooling and hydrogen cooling (including stator

cooling). Corrosion problems due to the water environment are common to both methods. In this chapter, examples of the corrosion of the structural parts of power generators are shown. Causes and

countermeasures are introduced and explained. 3.5.2.1 Corrosion of power generators and countermeasures

Turbine power generators are classified to air cooling, hydrogen cooling, and water cooling according to the internal cooling media.

In the case of hydrogen cooling and water cooling power generators that are operated under a clean environment in a closed power generator casing as well as air cooling generators, the internal air temperature is above 40°C in general. The inside of the power generator is less subject to corrosion.

As shown in Fig. 3.5.2-1, turbine power generators have important and characteristic corrosion problems, and countermeasures against them need to be taken.

Corrosion of water cooling stator coil

Stress corrosion of holding ring

1 . Stator frame 2 . Stator iron core 3 . Stator coil 4 . Bushing 5 . Lead box 6 . Rotor 7 . Rotor coil 8 . Bearing bracket (Turbine side) 9 . Bearing bracket (Exciter side) 10. Blower 11. Hydrogen gas cooler

Corrosion of hydrogen (air) cooler and cooling pipe

Fig. 3.5.2-1: Typical corroded sections in a turbine power generator

If water leaks from the cooler (water-hydrogen-type and water-air-type heat exchanger) in the power generator or cooling water circuits of the cooling water coil due to damage caused by corrosion, serious and fatal accidents such as short-circuiting or ground fault may be caused because the power generator is an electric machine.

Because the turbine power generator is an electric, high-speed rotating machine, the structural parts of the rotor must have a structure resistant to strong centrifugal force. The retaining ring is an important part of the rotor, and some accidents have occurred due to stress corrosion cracks. In the following section, serious corrosion problems and countermeasures are explained. 3.5.2.1.1 Corrosion cases of coolers and countermeasures

Coolers used for the turbine power generator are classified as follows: The hydrogen gas cooler or air cooler cools the cooling media in the power generator. The cooling water cooler cools the water system of the stator coil. There is a cooler that cools the oil for sealing the hydrogen gas. The air cooler is used for exciters. Comprehensive reliability including of the water systems and oil systems is required.

In these coolers, corrosion problems of cooled heat media including hydrogen gas-water, air-water, purified water-water, and oil-water do not occur very often, but water leakage due to corrosion might lead to serious accidents such as hydrogen gas leakage and insulation failures. High reliability is required. Most failures of these coolers are caused by corrosion.

The corrosion tendency depends on the water quality (freshwater, seawater), cooling water speed (low speed, high speed), and materials. The cooling water speed is simply classified into two, but it must be noted that the speed range depends on the material. If the component value of the cooling water quality changes, the corrosion tendency of the cooler also changes. Consequently, in designing corrosion resistance of the cooling water system, seasonal changes in cooling water quality, cooling method, and ambient environment must be understood to evaluate the water quality and select the materials.

363

3.5.2.1.2 Corrosion cases of water cooling coil and countermeasures The purity of the water used for the water cooling power generator is high. The frequency of corrosion is

much lower compared to general devices. Cooling water does not leak from the stator coil due to corrosion. Corrosion progresses very slowly from past experience. Corrosion product materials are deposited only in the filters or pipes. Major ingredients of corrosion product materials are CuO and Cu2O. One of the parameters related to corrosion is the dissolved oxygen concentration in the cooling water. In a system that has been put to practical use, the primary cooling water tank is covered with air or hydrogen. The dissolved oxygen concentration in the cooling water is high in the former case and low in the latter case. Both are operated in good condition under a stable dissolved oxygen concentration condition. However, it is regarded that there is an area where corrosion tends to occur in the middle of both concentrations. It is important to control the dissolved oxygen concentration for various reasons such as the plant operation method. Figure 3.5.2-2 shows the relation between the dissolved oxygen concentration and the corrosion amount.

Rel

ativ

e va

lue

of c

orro

sion

pr

ogre

ss sp

eed

(p.u

.) Measurement point

Dissolving oxygen concentration (ppb)

Material: Copper

Fig. 3.5.2-2: Relation between dissolving oxygen concentration and corrosion progress speed

The experiment result of the air dissolved condition in Fig. 3.5.2-3 clearly indicates this. The corrosion amount increases linearly in the case of air of 5 m/s. It saturates in the case of hydrogen or nitrogen. Concerning the difference in the corrosion amount depending on the speed, the corrosion amount is small in general if the speed is low. There is no large difference at 1.6 ~ 0.16 m/s.

Dissolved gas: Air Temperature: 75°C Water specific resistance:1 MΩ or more

Cor

rosi

on a

mou

nt (µ

)

Period (year)

Fig. 3.5.2-3: Long-term corrosion test result (Copper corrosion)

Next, the corrosion due to the current that is supplied to/from water by the application of high voltage (hereafter called galvanic corrosion) is explained. This is peculiar to electric devices of direct cooling by water. The current has corrosive effects only when a current is supplied between metal and water. If a current is supplied to the hollow conductor by supplying water to the hollow conductor inside, a current is not supplied between the water and the conductor because the conductor resistance is smaller than the water resistance. Consequently, this way of supplying a current does not induce corrosion. The experiment result when a current is supplied between metal and water is shown in Fig. 3.5.2-4. According to the experiment, the influence of the current on the copper is large, and corrosion is accelerated even in the case of AC. Stainless steel (SUS304) is extremely strong against a current. It can be considered that corrosion does not occur practically in the case of AC in particular.

364

Copper DC⊕ 1.28 mA/cm2

Copper AC 2.85 mA/cm2



Cor

rosi

on re

duct

ion

amou

nt (m

g/cm

2 )

Copper DC⊕0.025 mA /cm2


SUS27 DC⊕ 1.28 mA/cm2


Copper 0 mA/cm2

SUS27 DC⊕ 0.32 mA/cm2

Temperature: 85°C Dissolved gas: Air • AC is 60 Hz. • DC⊕ indicates the

corrosion amount on the plus side by dcurrent. Almostcorrosion is found on the minus side.

no irect

SUS27 AC 2.86 mA/cm2

Fig. 3.5.2-4: Galvanic corrosion of copper and stainless steel

Time (Day)

In general, corrosion of corrosion-prone metal is accelerated if different kinds of metal make contact with one

another. However, along with the increase in the specific resistance of the corrosive media, the influence becomes smaller. Consequently, it is regarded that contact corrosion problems do not occur in pure water based on common sense. In the corrosion experiment of the low specific resistance (100 kΩ-cm) of the silver brazing section of the stainless steel (SUS304) and copper, no contact corrosion was recognized in the galvanic pair of the stainless steel and silver brazing.

365

3.6 Efficiency and operation improvement of thermal power plants 3.6.1 Technology for improving the bearing force of boiler equipment

Boiler equipment receives various types of damage depending on the environment of use, most of which are combinations of several damaging factors. With respect to such damage, various measures for improving the bearing force have been taken, as shown in Table 3.6.1-1. (1) Example of measures taken against the portions where thermal fatigue damage occurred

Most of the damage boiler equipment receives is caused by thermal fatigue. The measures taken against such damage vary depending on the structure of each individual member. Table 3.6.1-2 shows an example of measures taken in order to improve the bearing force.

Table 3.6.1-1 Classification of measures for improving the bearing force

Cause Countermeasures Portion subject to countermeasures

Phenomenon

Creep aged strength deterioration of the welded portion

Inspection by replica, ultrasonic testing, TOFD method

Superheater/reheater pipe header, main/longitudinal direction of high temperature reheating steam piping, surrounding welded portions, elbow/Y-piece welded portion

Wear

Restriction on elongation by heat

Add flexibility Pipe header stub, finish of sealing, expansion of casing

Thermal shock

Change of shape, improvement of material, improvement of the shapes of the seat and piping

Desuperheater spray, small diameter piping with main piping (drain pressure tank)

Dissimilar metal welding (SUS/Cr-Mo)

Inconel solvent Joint of different piping material, fixture of different material

Corrosion fatigue

Change of structure and shape, water quality control

Fixture welded to the furnace wall piping, bent portion of the economizer

High temperature fatigue, oxidation

Improvement of the bearing force of material, addition of extra welding

Superheater, reheater (STBA28) Furnace wall

Oxidation of steam (SUS piping)

Fine particle SUS materialInner face shot blast

Superheater, reheater

R-machining, chamfering, change of shape

Shape the stress concentrates

Corrosion

Fatigue (including creep fatigue)

Creep

Piping supporting fixture, back-stay prevention fixture, fin end portion, pipe header lid plate at the corner of the burner wall box, expansion for the smoke duct

Sliding

Furnace wall, superheater, reheater

Protector, pipe thermal spraying

Coal ash, soot blow

Corrosion fatigue damage on the inner face of the furnace wall piping occurs when the strength against fatigue

is reduced due to corrosion by the inner fluid. In view of the fact that corrosion fatigue damage also occurs at the portion where a fixture is mounted on the external side, it is assumed that thermal stress is the major cause. An effective countermeasure is to reduce the thermal stress by improving the structure of the attached fixture on the external side.

366

Table 3.6.1-2 Example of measures for improving the bearing force of the boiler equipment

No. Name of portion Conventional structure Example of measures for improving the bearing force

1. Furnace wall Wall boxes such as burners, OAPs, inside TVs, soot blowers, etc. The corner portion cracks when a

temperature difference occurs between the furnace wall and the wall box in the course of the temperature rise after starting the operation of the furnace. As the wall box has been welded to the pipe, the temperature on the pipe side rises rapidly, causing the temperature difference to become larger.

Provide a step to the corner portion and chamfer the sharp edges to reduce the thermal stress. Weld the wall box to the fin to reduce the temperature difference.

2. Side wall at the furnace outlet/portion welded to the side wall of the furnace

Steam typeWater cooling type

Current type

In welded portions of different paths temperature difference occurs in the process of water filling, cooling stop, etc, causing cracks to occur on the fin.

Steam type

Water cooling type

Separation in the center

Arrange the end of the panel fin in an arch shape and apply R-machining to each fin edge. The furnace outlet should have 3-part structure as shown on the left. The same path should be used at the connection. In case of a different path connection, connect the path in 2 steps to make the temperature difference smaller.

3. Fin edge at the interim wall, etc. of the rear thermal transmission portion

If the fin edge has a flat shape, the concentration of stress becomes larger. A temperature difference is likely to cause cracks.

R-machining and cutting off the fin stop edge for improving the shape

Apply R-machining to the fin edge to reduce the stress concentration.

Chamfering and R-machining of the corner portion

367


4. Pipe header of the high temperature portion and stub tube welded portion

Honeycomb Honeycomb

(Flat shape) (Half-oval shape)

Half-oval shape of honeycomb Attached

fixtures

<Thick welded portion> The internal and external temperature difference becomes larger at the start of operation of the furnace at T-piece, pipe table and lid plate. <Stub> If any temperature difference occurs between the upper and lower portions of the pipe header and on any pipe within the panel at the start of operation of the furnace, a displacement difference occurs with the stub pipe causing stress to occur at the portion welded to the pipe header (at the base). A pipe welded at the lower side of both edges of the pipe header is affected largely by displacement of the pipe header. The stress becomes larger because the drain likely flows in and flexibility is insufficient.

MT or PT R-machining R-machining of the corners on the inner wall of

the pipe

Measures to give flexibility to the stub pipe

<Thick welded portion> Apply R-machining to the portion where the stress concentrates. The shape of the lid plate must be half-oval. <Stub> Add further flexibility to the structure in order to secure the bearing force against more frequent starts/stops.

5. Dissimilar metal welded joint under high temperature

Stress caused by the difference in thermal expansion and the strength reduction caused by the decarbonization phenomenon on the Cr-Mo steel side is superimposed, causing damage to occur at the weld border of dissimilar metals.

Inconel solvent has been used to reduce the thermal expansion difference and to prevent carbon migration as well.

6. Portion a loop pipe is connected

Tie rod Sliding spacer

Hanging loop pipe

If any temperature difference occurs when starting the furnace, cracks may occur to the linked metal fitted portion due to the stress concentration.

(Single lag) (Oval lag)

Improvement of the structure of the connection fixture

A sliding spacer should be used at the high temperature transferring portion to prevent locking, should a temperature difference occur. The tie lag at the rear heat transferring portion should be an oval lag to soften the stress concentration.

Crown Crown

Ceiling piping Ceiling piping

Additional sleeve

7. Portion passing through the ceiling

Due to a temperature difference between the crown and pipe, stress concentration occurs, which may cause cracks. By using a sleeve for the SUS pipe where it passes through the ceiling, the thermal stress can be softened.

A sleeve is used for both SUS pip and 2.25 Cr-1Mo pipe.

368


8. Devices in the spray pipe of the superheater

Swing type nozzle

(single-hole type)

Perforated type nozzle

Swing type nozzle of steam cooling type (single-hole

type)

Damage may occur to the spray nozzle, venturi edge, or support fixture due to thermal impact when spray water is discharged.

Flow nozzleVenturi pipe

The welded portion should have a flat surface of the same thickness.

The material has been graded up from SUS 304 to Inconel 600 (NCF600).

To improve spraying characteristics and nozzle bearing force during low flow rate by changing the spray nozzle to the perforated type from the single-hole type. When further additional bearing force is required, a structure to cool down the nozzle by steam should be employed. In addition, Inconel material is to be used at the venturi end. At the same time, a structure that can absorb thermal deformation should be employed for the support of internal devices to increase the bearing force.

9. 3-faces joining corner of the panel

Due to stress concentration at the fin stop end, cracks may occur.

Furnace side wall Furnace side wall at the outlet

Rear wall pipe at the furnace outlet

Apply R-machining to the fin end and change the shape to an arch.

10. Inner-casing on the ceiling

Corner casing may crack and cause gas leakage when it cannot absorb the expansion of 3-faces.

Corrugated expansion

Pipe header on the forepart wall of the furnace

Pipe header on the furnace wall

Employ corrugated expansion at the corner to increase the flexibility of the structure.

or

369

Table 3.6.2-1 Preventive maintenance and technologies for improving the bearing force of steam turbine (outline)

Cause of damage

Portion subject to countermeasures

Dam

age

by c

reep

D

amag

e by

fatig

ue

Frag

ility

Cor

rosi

on

Eros

ion

Oth

ers

Preventive maintenance and technologies for improving the bearing force

Employment of low Si content rotor material

Employment of improved rotor material Expansion of stress softening

grooves/flattening of 1st step rotating blade grooves

High/medium pressure rotor

Expansion of corner R of dummy grooves Employment of super clean rotor material Employment of improved type blade

grooves

Low pressure rotor

Integrated rotor Employment of ISB (integral shroud blade)High/medium

pressure rotating blade

Employment of large sized blade grooves

Employment of the new type of long blade Employment of snapper blade

Low pressure rotating blade

Improvement of the structure of low pressure blade (against erosion)

Nozzle against erosion (Cr pack treatment) Employment of improved type nozzle blade Employment of boron treated nozzle Employment of thermal spraying nozzle

High/medium pressure nozzle

Operation by injecting a full arc when starting operation (Employment of electro-hydraulic governor)

Low pressure stationary blade

Improvement of the outer wall shape of the nozzle diaphragm

Employment of high toughness casing material

High/medium pressure internal/external casing

Employment of improved cast steel material

Forged valve Main valve casing Employment of high toughness casing

material Main valve rod Alteration of valve rod and bushing material

370

3.6.2 Preventive maintenance and technologies for improving the bearing force turbine equipment Due to continuing operation of aged thermal power generation plants under severe conditions, aged

deterioration of steam turbine equipment has accelerated. In view of the extension of regular inspection interval and enforcement of self control of the equipment under

the above-mentioned operation conditions, preventive maintenance and measures for improvement of bearing force have become more important.

Table 3.6.2-1 outlines the preventive maintenance and measures for improvement of bearing force of steam turbine that have been developed and employed. (1) Turbine rotor ① Measures against creep damage and fragility

As measures against damage caused by high temperature creep or fragility of high/medium pressure rotors, rotor material is employed that has higher strength against high temperature creep than conventional rotor material, with low speed fragility and that corresponds to operation changes such as DSS, etc.

• Employment of low Si content rotor material

• Employment of improved rotor material

② Measures against fatigue damage As measures against fatigue damage to the portion of high/medium pressure rotors with high level thermal

pressure, processes for removing fatigued and deteriorated layers and improving the shape are employed.

• Expansion of stress softening grooves

• Flattening of 1st step rotating blade grooves

• Expansion of corner R of dummy grooves

③ Measures against fragility As measures against the reduction of toughness and ductility due to aged fragility of low pressure rotors, rotor

material with minimized impure chemical element is employed. • Employment of super clean rotor material

④ Measures against corrosion To attain improvement by changing the shapes of portions subject to corrosion and corrosion fatigue

• Employment of improved type blade grooves

• Integrated rotor

(2) Rotating blade ① Measures against creep damage

As measures against creep damage occurring to high/medium pressure rotating blades, improvement is attained by eliminating the tenon crimped structure by integrating a blade and a shroud(ISB blade) and reduction of stress from blade base/blade grooves by employing large-sized blade grooves.

• Employment of ISB (integral shroud blade)

• Employment of large sized blade grooves

② Measures against corrosion and erosion As measures against failure caused by the corrosion of low pressure rotating blades, the new type of long blade

from which the tie wire has been eliminated is employed. In addition, as measures against the erosion of final step rotating blades, the drain discharge process, etc. will be improved.

• Employment of the new type of long blade

• Employment of snapper blade

• Improvement of the structure of low pressure blade (against erosion)

(3) Nozzle diaphragm ① Measures against erosion by solid particles

As suppression/prevention measures against erosion of high/medium pressure nozzles by solid particles, reduction of erosion and improvement of erosion resistance are attained by improving the nozzle blade type and method of steam inflow.

371

• Nozzle against erosion (Cr pack treatment) High pressure shaft Medium pressure

shaft

Fig. 3.6.2-1 Expansion of stress softening grooves/flattening of 1st step rotating blade grooves

Conventional blade (Tenon crimped type)

ISB blade (Shroud integrated type)

Fig. 3.6.2-2 Prevention of creep damage to rotating blade by employing ISB

• Employment of improved type nozzle blade

• Employment of boron treated nozzle

• Employment of thermal spraying nozzle

• Operation by injecting full arc when starting operation

② Measures against corrosion

As measures against corrosion damage to the diaphragm of low pressure stationary blade, the shape of the diaphragm is improved.

Fig. 3.6.2-3 Employment of the new type of long blade (Comparison between former/new final step blades)

Conventional blade (left side)/ new blade (right side)

Large angle of water drop collision

Small angle of water drop collision

Water drop Water drop

• Reasonable angle of attack

Reduction of the erosion damage ratio

J-type StelliteFlat plate Stellite

• Without cover

Face contact

• Low rigidity blade type

Tie wire

• 2 tie wires with silver soldering

• Subsonic speed blade type

• 12Cr Stainless steel

• Lightening cover Full arc 1 ring

Structure able to largely reduce vibration Reduction of resonance points

• High rigidity blade type

Reduction of vibration amplitude Prevention of Stellite from peeling off

• Without tie wire

Elimination of the cause of occurrence of SCC of silver soldering

• Transonic speed

blade type

• 12Cr-Nb Stainless steel

Strength increased to 1.1 times

With silver soldering

Erosion shield

Cover structure

Blade rigidity

Tie wire structure

Blade type

Material

372

• Improvement of the external wall shape of the nozzle diaphragm

(4) High/medium pressure casing ① Measures against creep damage and fragility

As measures against crack generation/deformation or fragility caused by high temperature creep in the high/medium pressure internal and external casing, casing material having strength against high temperature creep and excellent roughness properties against destruction is employed.

• Employment of high toughness casing material

• Employment of improved cast steel material

(5) Main valves ① Measures against creep damage

By eliminating any and all defects contained in the cast valve and employing casing material having excellent roughness properties against destruction, the main valve casing can be improved.

• Forged valve

• Employment of high toughness casing material

② Measures against erosion For the main valve rod, etc., material having excellent strength against creep rupture and material generating

less amount of oxidized scale (by surface treatment) are employed. • Improvement of material for valve rod

• Improvement of material for bushing and surface treatment

(6) Examples of preventive maintenance and technologies for improving the bearing force ① Measures against fatigue damage of high/medium pressure rotors

Examples of measures against aged fatigue damage of high/medium pressure turbine rotors are shown in Fig. 3.6.2-1. Its purpose is to improve the bearing force against fatigue by skin-cutting the fatigue deteriorated layer of stress softening grooves and 1st step rotating blade grooves and further expanding the R of stress softening grooves and flattening the 1st step rotating blade groove bottom. ② Measures against creep damage of high/medium pressure rotating blades

A tenon crimping structure was used for reaction step rotating blades in the past. However, it was found that creep damage occurred to the tenon portion where the shroud’s centrifugal force was applied by high/medium pressure rotating blades exposed to high temperature steam. As countermeasures, ISB (integral shroud blade) in which the blade and shroud are integrated together has been employed. ISB has other effects in improving vibration characteristics by full arc tracing structure and by improving the sealing structure.

373

Casing

Drain catcher Stationary blade ring

Stationary blade

Rotating blade

Trace of water drop

Rotor

Drain discharge by drain catcher

Drain discharging port

Drain

Drain

Stationary blade

Slit

Stationary blade

[Improvement of drain discharge]

Drain discharging port Drain

Fig. 3.6.2-4 Measures against erosion of low pressure blades ③ Measures against corrosion/erosion damage of low pressure rotating blades

Low pressure step long blades have defects such as blade crack of the tie wire or crack of the tie wire hole. The new type of long blade improves the bearing force against the weakness of conventional blades and at the same time, high efficiency is targeted.

In Fig. 3.6.2-3, a comparison between new and old types of final step rotating 26-inch blades for 60 Hz is shown. The characteristics of new type blade are as follows;

• Elimination of tie wire (improvement of reliability and efficiency)

• Integral shroud S type cover (high attenuation effect)

• Best angle for water drops to collide (reduction of erosion damage ratio)

• Best transonic blade type (improvement of efficiency)

374

In addition to the aged erosion of low pressure rotating blades, not only decreased efficiency but also crack damage may be caused by corrosion or erosion.

In order to prevent such damage, Stellite plates are bonded to the front edge of rotating blades and hardening treatment is applied. In addition, by employing a hollow stationary blade with a slit and drain catcher and by promoting drain discharge aggressively, the erosion of low pressure rotating blades can be largely suppressed.

Boron treated nozzle Improvement of nozzle blade shape Handling by plant operation method

Measures against erosion

Change to cassette nozzle Common spare nozzle

Improvement of ease of maintenance

Nozzle box of upside down type Common spare nozzle box

Improvement of ease of maintenance

Ceramics thermal spraying nozzle Expansion of distance in the shaft direction

Measures against erosion

Fig. 3.6.2-5 Measures against erosion of high/medium pressure 1st step nozzles

An example of improvement of drain discharge from low pressure blades is shown in Fig. 3.6.2-4. ④ Measures against erosion of high/medium pressure 1st step nozzles ⁽²⁰⁾⁽²¹⁾

Erosion phenomenon (SPE: Solid Particle Erosion) caused by oxidized scales flying from boiler, etc. can be detected on the 1st (initial) step nozzles of high/medium pressure turbines, which creates various issues with respect to performance, reliability, regular inspection interval, maintenance and control, etc.

Several measures against erosion in this respect are shown in Fig. 3.6.2-5.

In case of high pressure 1st step nozzles, erosion is generated at the outlet end of the nozzle by solid particles flowing into the steam path. The erosion can be suppressed by such diffusion penetration treatment as boron treatment (to have B (boron) make diffusion penetration on the metal surface and form a very hard and fine chemical compound (Fe2B) layer on the nozzle plate), which strongly adheres to the base metal and forms an ultra hard coat.

The medium pressure initial step nozzle can be damaged when solid particles passing through the nozzle rebound from the rotating blades. As measures against this, bearing force is improved by forming on the back of the nozzle a 250 – 300 µm thick plate hardened coat against high temperature and stable thermal spray metal of the carbon family mainly composed of chromium carbide by high velocity gas flame metal spraying method (HVOF).

375

chapter3_1

Documents

master curve

master curve

hydraulic

mt copying

life consumption

miniature

surrounding

pressure rotor