CONTROL VALVE SOURCEBOOK

PULP & PAPER

Copyright © 2011 Fisher Controls International LLC All Rights Reserved.

Fisher, ENVIRO-SEAL, Whisper Trim, Cavitrol, WhisperFlo, Vee‐Ball, Control‐Disk, NotchFlo, easy‐e and FIELDVUE are marksowned by Fisher Controls International LLC, a business of Emerson Process Management. The Emerson logo is a trademark andservice mark of Emerson Electric Co. All other marks are the property of their respective owners.

This publication may not be reproduced, stored in a retrieval system, or transmitted in whole or in part, in any form or by any means,electronic, mechanical, photocopying, recording or otherwise, without the written permission of Fisher Controls International LLC.

Printed in U.S.A., First Edition

iii

Table of Contents

Introduction v

Chapter 1 Control Valve Selection 1-1

Chapter 2 Actuator Selection 2-1

Chapter 3 Liquid Valve Sizing 3-1

Chapter 4 Cavitation & Flashing 4-1

Chapter 5 Gas Valve Sizing 5-1

Chapter 6 Control Valve Noise 6-1

Chapter 7 Steam Conditioning 7-1

Chapter 8 Process Overview 8-1

Chapter 9 Pulping 9-1

Chapter 10A Batch Digesters 10A-1

Chapter 10B Continuous Digesters 10B-1

Chapter 11 Black Liquor Evaporators/Concentrators 11-1

Chapter 12 Kraft Recovery Boiler 12-1

Chapter 13 Recausticizing & Lime Recovery 13-1

Chapter 14 Bleaching & Brightening 14-1

Chapter 15 Stock Preparation 15-1

Chapter 16 Wet End Chemistry 16-1

Chapter 17 Paper Machine 17-1

Chapter 18 Power & Recovery Boiler 18-1

iv

v

Pulp and Paper Control Valves

IntroductionThis sourcebook’s intent is to introduce a pulpand paper mill’s processes, as well as the use ofcontrol valves in many of the processes found inthe mill. It is intended to help you:

� Understand pulp and paper processes

� Learn where control valves are typicallylocated within each process

� Identify valves commonly used for specificapplications

� Identify troublesome/problem valves withinthe process

The information provided will follow a standardformat of:

� Description of the process

� Functional drawing of the process

� Fisher� valves to be considered in eachprocess and their associated function

� Impacts and/or considerations fortroublesome/problem valves

Valve SelectionThe information presented in this sourcebook isintended to assist in understanding the controlvalve requirements of general pulp and papermill’s processes.

Since every mill is different in technology andlayout, the control valve requirements andrecommendations presented by this sourcebookshould be considered as general guidelines.Under no circumstances should this informationalone be used to select a control valve withoutensuring the proper valve construction is identifiedfor the application and process conditions.

All valve considerations should be reviewed by thelocal business representative as part of any valveselection or specification activity.

Control ValvesValves described within a chapter are labeledand numbered corresponding to the identificationused in the process flow chart for that chapter.Their valve function is described, and aspecification section gives added information onprocess conditions, names of Fisher valves thatmay be considered, process impact of the valve,and any special considerations for the processand valve(s) of choice.

Process DrawingsThe process drawings within each chapter showmajor equipment items, their typical placementwithin the processing system, and process flowdirection. Utilities and pumps are not shownunless otherwise stated.

Many original equipment manufacturers (OEMs)provide equipment to the pulp and paperindustry, each with their own processes andproprietary information. Process drawings arebased on general equipment configurationsunless otherwise stated.

Problem ValvesOften there are references to valve-causedproblems or difficulties. The list of problemsinclude valve erosion from process media,stickiness caused by excessive friction (stiction),excessive play in valve to actuator linkages(typically found in rotary valves) that causesdeadband, excessive valve stem packingleakage, and valve materials that areincompatible with the flowing medium. Any one,or a combination of these difficulties, may affectprocess quality and throughput with a resultingnegative impact on mill profitability.

Many of these problems can be avoided orminimized through proper valve selection.Consideration should be given to valve style andsize, actuator capabilities, analog versus digitalinstrumentation, materials of construction, etc.Although not being all-inclusive, the informationfound in this sourcebook should facilitate thevalve selection process.

vi

www.Fisher.com

Chapter 1

Control Valve Selection

In the past, a customer simply requested a controlvalve and the manufacturer offered the productbest-suited for the job. The choices among themanufacturers were always dependent uponobvious matters such as cost, delivery, vendorrelationships, and user preference. However,accurate control valve selection can beconsiderably more complex, especially forengineers with limited experience or those whohave not kept up with changes in the control valveindustry.

An assortment of sliding-stem and rotary valvestyles are available for many applications. Someare touted as “universal” valves for almost anysize and service, while others are claimed to beoptimum solutions for narrowly defined needs.Even the most knowledgeable user may wonderwhether they are really getting the most for theirmoney in the control valves they have specified.

Like most decisions, selection of a control valveinvolves a great number of variables; the everydayselection process tends to overlook a number ofthese important variables. The followingdiscussion includes categorization of availablevalve types and a set of criteria to be considered inthe selection process.

What Is A Control Valve?Process plants consist of hundreds, or eventhousands, of control loops all networked togetherto produce a product to be offered for sale. Eachof these control loops is designed to control acritical process variable such as pressure, flow,level, temperature, etc., within a required operatingrange to ensure the quality of the end-product.

These loops receive, and internally create,disturbances that detrimentally affect the processvariable. Interaction from other loops in thenetwork provides disturbances that influence theprocess variable. To reduce the effect of theseload disturbances, sensors and transmitters collectinformation regarding the process variable and itsrelationship to a desired set point. A controller thenprocesses this information and decides what mustoccur in order to get the process variable back towhere it should be after a load disturbance occurs.When all measuring, comparing, and calculatingare complete, the strategy selected by thecontroller is implemented via some type of finalcontrol element. The most common final controlelement in the process control industries is thecontrol valve.

A control valve manipulates a flowing fluid such asgas, steam, water, or chemical compounds tocompensate for the load disturbance and keep theregulated process variable as close as possible tothe desired set point.

Many people who speak of “control valves” areactually referring to “control valve assemblies.”The control valve assembly typically consists ofthe valve body, the internal trim parts, an actuatorto provide the motive power to operate the valve,and a variety of additional valve accessories,which may include positioners, transducers, supplypressure regulators, manual operators, snubbers,or limit switches.

It is best to think of a control loop as aninstrumentation chain. Like any other chain, theentire chain is only as good as its weakest link. Itis important to ensure that the control valve is notthe weakest link.

1−2

Valve Types and CharacteristicsThe control valve regulates the rate of fluid flow asthe position of the valve plug or disk is changed byforce from the actuator. To do this, the valve must:

� Contain the fluid without external leakage.

� Have adequate capacity for the intendedservice.

� Be capable of withstanding the erosive,corrosive, and temperature influences of theprocess.

� Incorporate appropriate end connections tomate with adjacent pipelines and actuatorattachment means to permit transmission ofactuator thrust to the valve plug stem or rotaryshaft.

Many styles of control valve bodies have beendeveloped. Some can be used effectively in anumber of applications while others meet specificservice demands or conditions and are used lessfrequently. The subsequent text describes popularcontrol valve body styles utilized today.

Globe Valves

Single-Port Valve BodiesSingle-port is the most common valve body styleand is simple in construction. Single-port valvesare available in various forms, such as globe,angle, bar stock, forged, and split constructions.Generally, single-port valves are specified forapplications with stringent shutoff requirements.They use metal-to-metal seating surfaces orsoft-seating with PTFE or other compositionmaterials forming the seal.

Single-port valves can handle most servicerequirements. Because high pressure fluid isnormally loading the entire area of the port, theunbalance force created must be considered whenselecting actuators for single-port control valvebodies. Although most popular in the smallersizes, single-port valves can often be used in NPS4 to 8 with high thrust actuators.

Many modern single-seated valve bodies use cageor retainer-style construction to retain the seat ringcage, provide valve plug guiding, and provide ameans for establishing particular valve flow

Figure 1-1. Single-Ported Globe-Style ValveBody

W7027-1

characteristics. Retainer-style trim also offers easeof maintenance with flow characteristics altered bychanging the plug. Cage or retainer-stylesingle-seated valve bodies can also be easilymodified by a change of trim parts to providereduced-capacity flow, noise attenuation, orcavitation eliminating or reducing trim (see chapter 4).

Figure 1-1 shows one of the more popular styles ofsingle-ported or single-seated globe valve bodies.They are widely used in process controlapplications, particularly in sizes NPS 1 throughNPS 4. Normal flow direction is most often flow-upthrough the seat ring.

Angle valves are nearly always single ported, asshown in figure 1-2. This valve has cage-style trimconstruction. Others might have screwed-in seatrings, expanded outlet connections, restricted trim,and outlet liners for reduction of erosion damage.

Bar stock valve bodies are often specified forcorrosive applications in the chemical industry(figure 1-3), but may also be requested in otherlow flow corrosive applications. They can bemachined from any metallic bar-stock material andfrom some plastics. When exotic metal alloys arerequired for corrosion resistance, a bar-stock valvebody is normally less expensive than a valve bodyproduced from a casting.

High pressure single-ported globe valves are oftenfound in power plants due to high pressure steam(figure 1-4). Variations available include

1−3

Figure 1-2. Flanged Angle-StyleControl Valve Body

W0971

Figure 1-3. Bar Stock Valve Body

W9756

cage-guided trim, bolted body-to-bonnetconnection, and others. Flanged versions areavailable with ratings to Class 2500.

Balanced-Plug Cage-Style ValveBodiesThis popular valve body style, single-ported in thesense that only one seat ring is used, provides theadvantages of a balanced valve plug often

Figure 1-4. High Pressure Globe-StyleControl Valve Body

W0540

Figure 1-5. Valve Body with Cage-Style Trim,Balanced Valve Plug, and Soft Seat

W0992-4

associated only with double-ported valve bodies(figure 1-5). Cage-style trim provides valve plugguiding, seat ring retention, and flowcharacterization. In addition, a sliding pistonring-type seal between the upper portion of thevalve plug and the wall of the cage cylindervirtually eliminates leakage of the upstream highpressure fluid into the lower pressure downstreamsystem.

1−4

Figure 1-6. High Capacity Valve Body withCage-Style Noise Abatement Trim

W0997

Downstream pressure acts upon both the top andbottom sides of the valve plug, thereby nullifyingmost of the static unbalance force. Reducedunbalance permits operation of the valve withsmaller actuators than those necessary forconventional single-ported valve bodies.

Interchangeability of trim permits the choice ofseveral flow characteristics or of noise attenuationor anticavitation components. For most availabletrim designs, the standard direction of flow is inthrough the cage openings and down through theseat ring. These are available in various materialcombinations, sizes through NPS 20, and pressureratings to Class 2500.

High Capacity, Cage-Guided ValveBodiesThis adaptation of the cage-guided bodiesmentioned above was designed for noiseapplications, such as high pressure power plants,where sonic steam velocities are oftenencountered at the outlet of conventional valvebodies (figure 1-6).

The design incorporates oversized endconnections with a streamlined flow path and theease of trim maintenance inherent with cage-styleconstructions. Use of noise abatement trimreduces overall noise levels by as much as 35decibels. The design is also available in cagelessversions with a bolted seat ring, end connectionsizes through NPS 20, Class 600, and versions for

liquid service. The flow direction depends upon theintended service and trim selection, withunbalanced constructions normally flow-up andbalanced constructions normally flow-down.

Port-Guided Single-Port Valve Bodies� Usually limited to 150 psi (10 bar) maximum

pressure drop.

� Susceptible to velocity-induced vibration.

� Typically provided with screwed in seat ringswhich might be difficult to remove after use.

Three-Way Valve Bodies� Provide general converging (flow-mixing) or

diverging (flow-splitting) service.

� Best designs use cage-style trim for positivevalve plug guiding and ease of maintenance.

� Variations include trim materials selected forhigh temperature service. Standard endconnections (flanged, screwed, butt weld, etc.) canbe specified to mate with most any piping scheme.

� Actuator selection demands carefulconsideration, particularly for constructions withunbalanced valve plug.

A balanced valve plug style three-way valve bodyis shown with the cylindrical valve plug in the downposition (figure 1-7). This position opens thebottom common port to the right-hand port andshuts off the left-hand port. The construction canbe used for throttling mid-travel position control ofeither converging or diverging fluids.

Rotary Valves

Traditional Butterfly ValveStandard butterfly valves are available in sizesthrough NPS 72 for miscellaneous control valveapplications. Smaller sizes can use versions oftraditional diaphragm or piston pneumaticactuators, including the modern rotary actuatorstyles. Larger sizes might require high outputelectric or long-stroke pneumatic cylinderactuators.

Butterfly valves exhibit an approximately equalpercentage flow characteristic. They can be used

1−5

Figure 1-7. Three Way Valve withBalanced Valve Plug

W9045-1

Figure 1-8. High-Performance ButterflyControl Valve

W4641

for throttling service or for on-off control. Soft-seatconstructions can be obtained by utilizing a liner orby including an adjustable soft ring in the body oron the face of the disk.

� Require minimum space for installation(figure 1-8).

� Provide high capacity with low pressure lossthrough the valves.

Figure 1-9. Eccentric-Disk Rotary-ShaftControl Valve

W8380

� Offer an economic advantage, particularly inlarger sizes and in terms of flow capacity per dollarinvestment.

� Mate with standard raised-face pipelineflanges.

� Depending on size, might require high outputor oversized actuators due to valve size valves orlarge operating torques from large pressure drops.

� Standard liner can provide precise shutoffand quality corrosion protection with nitrile orPTFE liner.

Eccentric-Disk Control ValveEccentric disk rotary control valves are intendedfor general service applications not requiringprecision throttling control. They are frequentlyapplied in applications requiring large sizes andhigh temperatures due to their lower cost relativeto other styles of control valves. The control rangefor this style of valve is approximately one third aslarge as a ball or globe-style valves.Consequently, additional care is required in sizingand applying this style of valve to eliminate controlproblems associated with process load changes.They are well-suited for constant process loadapplications.

� Provide effective throttling control.

� Linear flow characteristic through 90 degreesof disk rotation (figure 1-9).

� Eccentric mounting of disk pulls it away fromthe seal after it begins to open, minimizing sealwear.

1−6

Figure 1-10. Fisher Control-Disk Valve with 2052 Actuator and FIELDVUE DVC6200 Digital Valve Controller

W9425 W9418

WAFER STYLE SINGLE FLANGE STYLE

� Bodies are available in sizes through NPS 24compatible with standard ASME flanges.

� Utilize standard pneumatic diaphragm orpiston rotary actuators.

� Standard flow direction is dependent uponseal design; reverse flow results in reducedcapacity.

Control-Disk ValveThe Control-Disk� valve (figure 1-10) offersexcellent throttling performance, while maintainingthe size (face-to-face) of a traditional butterflyvalve. The Control-Disk valve is first in class incontrollability, rangeability, and tight shutoff, and itis designed to meet worldwide standards.

� Utilizes a contoured edge and uniquepatented disk to provide an improved control rangeof 15 - 70% of valve travel. Traditional butterflyvalves are typically limited to 25% - 50% controlrange.

� Includes a tested valve sealing design,available in both metal and soft seats, to providean unmatched cycle life while still maintainingexcellent shutoff

� Spring loaded shaft positions disk against theinboard bearing nearest the actuator allowing forthe disk to close in the same position in the seal,and allows for either horizontal or verticalmounting.

� Complimenting actuator comes in three,compact sizes, has nested springs and a patented

Figure 1-11. Rotary-Shaft Control Valvewith V-Notch Ball

W8172-2

lever design to increase torque range within eachactuator size.

V-notch Ball Control ValveThis construction is similar to a conventional ballvalve, but with patented, contoured V-notch in theball (figure 1-11). The V-notch produces anequal-percentage flow characteristic. Thesecontrol valves provide precise rangeability, control,and tight shutoff.

� Straight-through flow design produces littlepressure drop.

� Bodies are suited to provide control oferosive or viscous fluids, paper stock, or otherslurries containing entrained solids or fibers.

1−7

Figure 1-12. Sectional of Eccentric-PlugControl Valve Body

W4170-4

� They utilize standard diaphragm or pistonrotary actuators.

� Ball remains in contact with seal duringrotation, which produces a shearing effect as theball closes and minimizes clogging.

� Bodies are available with either heavy-duty orPTFE-filled composition ball seal ring to provideexcellent rangeability in excess of 300:1.

� Bodies are available in flangeless orflanged-body end connections. Both flanged andflangeless valves mate with Class 150, 300, or 600flanges or DIN flanges.

� Valves are capable of energy absorbingspecial attenuating trim to provide improvedperformance for demanding applications.

Eccentric-Plug Control Valve� Valve assembly combats erosion. The

rugged body and trim design handle temperaturesto 800°F (427°C) and shutoff pressure drops to1500 psi (103 bar).

� Path of eccentric plug minimizes contact withthe seat ring when opening, thus reducing seatwear and friction, prolonging seat life, andimproving throttling performance (figure 1-12).

� Self-centering seat ring and rugged plugallow forward or reverse-flow with tight shutoff ineither direction. Plug, seat ring, and retainer areavailable in hardened materials, includingceramics, for selection of erosion resistance.

� Designs offering a segmented V-notch ball inplace of the plug for higher capacity requirementsare available.

This style of rotary control valve is well-suited forcontrol of erosive, coking, and otherhard-to-handle fluids, providing either throttling or

on-off operation. The flanged or flangeless valvesfeature streamlined flow passages and ruggedmetal-trim components for dependable service inslurry applications.

Control Valve End ConnectionsThe three common methods of installing controlvalves in pipelines are by means of:

� Screwed pipe threads

� Bolted gasketed flanges

� Welded end connections

Screwed Pipe ThreadsScrewed end connections, popular in small controlvalves, are typically more economical than flangedends. The threads usually specified are taperedfemale National Pipe Thread (NPT) on the valvebody. They form a metal-to-metal seal by wedgingover the mating male threads on the pipeline ends.This connection style, usually limited to valves notlarger than NPS 2, is not recommended forelevated temperature service. Valve maintenancemight be complicated by screwed end connectionsif it is necessary to take the body out of thepipeline. This is because the valve cannot beremoved without breaking a flanged joint or unionconnection to permit unscrewing the valve bodyfrom the pipeline.

Bolted Gasketed FlangesFlanged end valves are easily removed from thepiping and are suitable for use through the rangeof working pressures for which most control valvesare manufactured (figure 1-13). Flanged endconnections can be used in a temperature rangefrom absolute zero to approximately 1500°F(815°C). They are used on all valve sizes. Themost common flanged end connections includeflat-face, raised-face, and ring-type joint.

The flat face variety allows the matching flanges tobe in full-face contact with the gasket clampedbetween them. This construction is commonlyused in low pressure, cast iron, and brass valves,and minimizes flange stresses caused by initialbolting-up force.

The raised-face flange features a circularraised-face with the inside diameter the same asthe valve opening, and the outside diameter lessthan the bolt circle diameter. The raised-face is

1−8

Figure 1-13. Popular Varieties ofBolted Flange Connections

A7098

Figure 1-14. Common Welded End Connections

A7099

finished with concentric circular grooves forprecise sealing and resistance to gasket blowout.This kind of flange is used with a variety of gasketmaterials and flange materials for pressuresthrough the 6000 psig (414 bar) pressure rangeand for temperatures through 1500°F (815°C).This style of flanging is normally standard on Class250 cast iron bodies and all steel and alloy steelbodies.

The ring-type joint flange is similar in looks to theraised-face flange except that a U-shaped grooveis cut in the raised-face concentric with the valveopening. The gasket consists of a metal ring witheither an elliptical or octagonal cross-section.When the flange bolts are tightened, the gasket iswedged into the groove of the mating flange and atight seal is made. The gasket is generally soft iron

or Monel�, but is available in almost any metal.This makes an excellent joint at high pressuresand is used up to 15,000 psig (1034 bar),however, it is generally not used at hightemperatures. It is furnished only on steel andalloy valve bodies when specified.

Welding End ConnectionsWelding ends on control valves (figure 1-14) areleak-tight at all pressures and temperatures, andare economical in first cost. Welding end valvesare more difficult to take from the line and arelimited to weldable materials. Welding ends comein two styles:

� Socket welding

� Buttwelding

The socket welding ends are prepared by boring ina socket at each end of the valve with an insidediameter slightly larger than the pipe outsidediameter. The pipe slips into the socket where itbutts against a shoulder and then joins to the valvewith a fillet weld. Socket welding ends in a givensize are dimensionally the same regardless of pipeschedule. They are usually furnished in sizesthrough NPS 2.

The buttwelding ends are prepared by bevelingeach end of the valve to match a similar bevel onthe pipe. The two ends are then butted to thepipeline and joined with a full penetration weld.This type of joint is used on all valve styles and theend preparation must be different for eachschedule of pipe. These are generally furnished forcontrol valves in NPS 2-1/2 and larger. Care mustbe exercised when welding valve bodies in thepipeline to prevent excessive heat transmitted tovalve trim parts. Trims with low-temperaturecomposition materials must be removed beforewelding.

Valve Body BonnetsThe bonnet of a control valve is the part of thebody assembly through which the valve plug stemor rotary shaft moves. On globe or angle bodies, itis the pressure retaining component for one end ofthe valve body. The bonnet normally provides ameans of mounting the actuator to the body andhouses the packing box. Generally, rotary valvesdo not have bonnets. (On some rotary-shaftvalves, the packing is housed within an extensionof the valve body itself, or the packing box is aseparate component bolted between the valvebody and bonnet.)

1−9

Figure 1-15. Typical Bonnet, Flange,and Stud Bolts

W0989

On a typical globe-style control valve body, thebonnet is made of the same material as the valvebody or is an equivalent forged material because itis a pressure-containing member subject to thesame temperature and corrosion effects as thebody. Several styles of valve body-to-bonnetconnections are illustrated. The most common isthe bolted flange type shown in figure 1-15. Abonnet with an integral flange is also illustrated infigure 1-15. Figure 1-3 illustrates a bonnet with aseparable, slip-on flange held in place with a splitring. The bonnet used on the high pressure globevalve body illustrated in figure 1-4, is screwed intothe valve body. Figure 1-8 illustrates a rotary-shaftcontrol valve in which the packing is housed withinthe valve body and a bonnet is not used. Theactuator linkage housing is not a pressure-containing part and is intended to enclose thelinkage for safety and environmental protection.

On control valve bodies with cage- or retainer-styletrim, the bonnet furnishes loading force to preventleakage between the bonnet flange and the valvebody, and also between the seat ring and thevalve body. The tightening of the body-bonnetbolting compresses a flat sheet gasket to seal thebody-bonnet joint, compresses a spiral-woundgasket on top of the cage, and compresses anadditional flat sheet gasket below the seat ring toprovide the seat ring-body seal. The bonnet alsoprovides alignment for the cage, which, in turn,

guides the valve plug to ensure proper valve plugstem alignment with the packing.

As mentioned previously, the conventional bonneton a globe-type control valve houses the packing.The packing is most often retained by a packingfollower held in place by a flange on the yoke bossarea of the bonnet (figure 1-15). An alternatepacking retention means is where the packingfollower is held in place by a screwed gland (figure1-3). This alternate is compact, thus, it is oftenused on small control valves, however, the usercannot always be sure of thread engagement.Therefore, caution should be used if adjusting thepacking compression when the control valve is inservice.

Most bolted-flange bonnets have an area on theside of the packing box which can be drilled andtapped. This opening is closed with a standardpipe plug unless one of the following conditionsexists:

� It is necessary to purge the valve body andbonnet of process fluid, in which case the openingcan be used as a purge connection.

� The bonnet opening is being used to detectleakage from the first set of packing or from afailed bellows seal.

Extension BonnetsExtension bonnets are used for either high or lowtemperature service to protect valve stem packingfrom extreme process temperatures. StandardPTFE valve stem packing is useful for mostapplications up to 450°F (232°C). However, it issusceptible to damage at low processtemperatures if frost forms on the valve stem. Thefrost crystals can cut grooves in the PTFE, thus,forming leakage paths for process fluid along thestem. Extension bonnets remove the packing boxof the bonnet far enough from the extremetemperature of the process that the packingtemperature remains within the recommendedrange.

Extension bonnets are either cast (figure 1-16) orfabricated (figure 1-17). Cast extensions offerbetter high temperature service because of greaterheat emissivity, which provides better coolingeffect. Conversely, smooth surfaces that can befabricated from stainless steel tubing are preferredfor cold service because heat influx is usually themajor concern. In either case, extension wallthickness should be minimized to cut down heattransfer. Stainless steel is usually preferable to

1−10

Figure 1-16. Extension Bonnet

W0667-2

Figure 1-17. Valve Body withFabricated Extension Bonnet

W1416

carbon steel because of its lower coefficient ofthermal conductivity. On cold service applications,insulation can be added around the extension toprotect further against heat influx.

Figure 1-18. ENVIRO-SEAL� BellowsSeal Bonnet

W6434

Bellows Seal BonnetsBellows seal bonnets (figure 1-18) are used whenno leakage (less than 1x10−6 cc/sec of helium)along the stem can be tolerated. They are oftenused when the process fluid is toxic, volatile,radioactive, or highly expensive. This specialbonnet construction protects both the stem and thevalve packing from contact with the process fluid.Standard or environmental packing boxconstructions above the bellows seal unit willprevent catastrophic failure in case of rupture orfailure of the bellows.

As with other control valve pressure/ temperaturelimitations, these pressure ratings decrease withincreasing temperature. Selection of a bellowsseal design should be carefully considered, andparticular attention should be paid to properinspection and maintenance after installation. Thebellows material should be carefully considered toensure the maximum cycle life.

Two types of bellows seal designs are used forcontrol valves:

� Mechanically formed as shown in figure 1-19

� Welded leaf bellows as shown in figure 1-20

The welded-leaf design offers a shorter totalpackage height. Due to its method of manufactureand inherent design, service life may be limited.

1−11

Figure 1-21. Comprehensive Packing Material Arrangementsfor Globe-Style Valve Bodies

B2565

Figure 1-19. Mechanically Formed BellowsA5954

Figure 1-20. Welded Leaf BellowsA5955

The mechanically formed bellows is taller incomparison and is produced with a morerepeatable manufacturing process.

Control Valve PackingMost control valves use packing boxes with thepacking retained and adjusted by a flange andstud bolts (figure 1-27). Several packing materialscan be used depending upon the serviceconditions expected and whether the applicationrequires compliance to environmental regulations.Brief descriptions and service condition guidelinesfor several popular materials and typical packingmaterial arrangements are shown in figure 1-21.

PTFE V-Ring� Plastic material with inherent ability to

minimize friction.

� Molded in V-shaped rings that are springloaded and self-adjusting in the packing box.Packing lubrication not required.

� Resistant to most known chemicals exceptmolten alkali metals.

� Requires extremely smooth (2 to 4micro-inches RMS) stem finish to seal properly.Will leak if stem or packing surface is damaged.

� Recommended temperature limits: −40°F to+450°F (−40°C to +232°C)

� Not suitable for nuclear service becausePTFE is easily destroyed by radiation.

1−12

Figure 1-22. Measurement Frequency for ValvesControlling Volatile Organic Chemicals (VOC)

B2566

Laminated and Filament Graphite

� Suitable for high temperature nuclear serviceor where low chloride content is desirable (GradeGTN).

� Provides leak-free operation, high thermalconductivity, and long service life, but produceshigh stem friction and resultant hysteresis.

� Impervious to most hard-to-handle fluids andhigh radiation.

� Suitable temperature range: Cryogenictemperatures to 1200°F (649°C).

� Lubrication not required, but an extensionbonnet or steel yoke should be used when packingbox temperature exceeds 800°F (427°C).

USA Regulatory Requirements forFugitive Emissions

Fugitive emissions are non-point source volatileorganic emissions that result from processequipment leaks. Equipment leaks in the UnitedStates have been estimated at over 400 millionpounds per year. Strict government regulations,developed by the US, dictate Leak Detection andRepair (LDAR) programs. Valves and pumps havebeen identified as key sources of fugitiveemissions. In the case of valves, this is the

leakage to atmosphere due to packing seal orgasket failures.

The LDAR programs require industry to monitor allvalves (control and noncontrol) at an interval thatis determined by the percentage of valves found tobe leaking above a threshold level of 500 ppmv(some cities use a 100 ppmv criteria). Thisleakage level is so slight you cannot see or hear it.The use of sophisticated portable monitoringequipment is required for detection. Detectionoccurs by sniffing the valve packing area forleakage using an Environmental ProtectionAgency (EPA) protocol. This is a costly andburdensome process for industry.

The regulations do allow for the extension of themonitoring period for up to one year if the facilitycan demonstrate an extremely low ongoingpercentage of leaking valves (less than 0.5% ofthe total valve population). The opportunity toextend the measurement frequency is shown infigure 1-22.

Packing systems designed for extremely lowleakage requirements also extend packing seal lifeand performance to support an annual monitoringobjective. The ENVIRO-SEAL� packing system isone example. Its enhanced seals incorporate fourkey design principles including:

� Containment of the pliable seal materialthrough an anti-extrusion component.

1−13

� Proper alignment of the valve stem or shaftwithin the bonnet bore.

� Applying a constant packing stress throughBelleville springs.

� Minimizing the number of seal rings to reduceconsolidation, friction, and thermal expansion.

The traditional valve selection process meantchoosing a valve design based upon its pressureand temperature capabilities as well as its flowcharacteristics and material compatibility. Valvestem packing used in the valve was determinedprimarily by the operating temperature in thepacking box area. The available material choicesincluded PTFE for temperatures below 93°C(200°F) and graphite for higher temperatureapplications.

Today, choosing a valve packing system hasbecome much more complex due to the number ofconsiderations one must take into account. Forexample, emissions control requirements, such asthose imposed by the Clean Air Act within theUnited States and by other regulatory bodies,place tighter restrictions on sealing performance.Constant demands for improved process outputmean that the valve packing system must nothinder valve performance. Also, today’s trendtoward extended maintenance schedules dictatesthat valve packing systems provide the requiredsealing over longer periods.

In addition, end user specifications that havebecome de facto standards, as well as standardsorganizations specifications, are used bycustomers to place stringent fugitive emissionsleakage requirements and testing guidelines onprocess control equipment vendors. EmersonProcess Management and its observance oflimiting fugitive emissions is evident by its reliablevalve sealing (packing and gasket) technologies,global emissions testing procedures, andemissions compliance approvals.

Given the wide variety of valve applications andservice conditions within industry, these variables(sealing ability, operating friction levels, operatinglife) are difficult to quantify and compare. A properunderstanding requires a clarification of tradenames.

Figure 1-23. Single PTFE V-Ring Packing

A6161-1

Single PTFE V-Ring Packing (Fig.1-23)

The single PTFE V-ring arrangement uses a coilspring between the packing and packing follower.It meets the 100 ppmv criteria, assuming that thepressure does not exceed 20.7 bar (300 psi) andthe temperature is between −18°C and 93°C (0°Fand 200°F). It offers excellent sealing performancewith the lowest operating friction.

ENVIRO-SEAL PTFE Packing (Fig. 1-24)

The ENVIRO-SEAL PTFE packing system is anadvanced packing method that utilizes a compact,live-load spring design suited to environmentalapplications up to 51.7 bar and 232°C (750 psiand 450°F). While it most typically is thought of asan emission-reducing packing system,ENVIRO-SEAL PTFE packing is, also, well-suitedfor non-environmental applications involving hightemperatures and pressures, yielding the benefit oflonger, ongoing service life.

ENVIRO-SEAL Duplex Packing(Fig. 1-25)

This special packing system provides thecapabilities of both PTFE and graphitecomponents to yield a low friction, low emission,fire-tested solution (API Standard 589) forapplications with process temperatures up to232°C (450°F).

1−14

Figure 1-24. ENVIRO-SEAL PTFE Packing System

A6163

Figure 1-25. ENVIRO-SEAL Duplex (PTFE andGraphite) Packing System

KALREZ� Valve Stem Packing (KVSP)systems

The KVSP pressure and temperature limitsreferenced are for Fisher valve applications only.KVSP with PTFE is suited to environmental use upto 24.1 bar and 204°C (350 psi and 400°F) and, tosome non-environmental services up to 103 bar(1500 psi). KVSP with ZYMAXX�, which is a

Figure 1-26. ENVIRO-SEAL Graphite ULF Packing System

39B4612-A

carbon fiber reinforced TFE, is suited to 260°C(500°F) service.

ENVIRO-SEAL Graphite Ultra LowFriction (ULF) Packing (Fig. 1-26)This packing system is designed primarily forenvironmental applications at temperatures inexcess of 232°C (450°F). The patented ULFpacking system incorporates thin PTFE layersinside the packing rings and thin PTFE washers oneach side of the packing rings. This strategic

1−15

Figure 1-27. ENVIRO-SEAL GraphitePacking System for Rotary Valves

W6125-1

placement of PTFE minimizes control problems,reduces friction, promotes sealing, and extendsthe cycle life of the packing set.

HIGH-SEAL Graphite ULF Packing

Identical to the ENVIRO-SEAL graphite ULFpacking system below the packing follower, theHIGH-SEAL system utilizes heavy-duty, largediameter Belleville springs. These springs provideadditional follower travel and can be calibratedwith a load scale for a visual indication of packingload and wear.

ENVIRO-SEAL Graphite Packing forRotary Valves (Fig. 1-27)

ENVIRO-SEAL graphite packing is designed forenvironmental applications from −6°C to 316°C(20°F to 600°F) or for those applications where firesafety is a concern. It can be used with pressuresto 103 bar (1500 psi) and still satisfy the 500 ppmvEPA leakage criteria.

Graphite Ribbon Packing for RotaryValves

Graphite ribbon packing is designed fornon-environmental applications that span a widetemperature range from −198°C to 538°C (−325°Fto 1000°F).

The following table provides a comparison ofvarious sliding-stem packing selections and arelative ranking of seal performance, service life,and packing friction for environmental applications.

Braided graphite filament and double PTFE arenot acceptable environmental sealing solutions.

The following applies to rotary valves. In the caseof rotary valves, single PTFE and graphite ribbonpacking arrangements do not perform well asfugitive emission sealing solutions.

The control of valve fugitive emissions and areduction in industry’s cost of regulatorycompliance can be achieved through these stemsealing technologies.

While ENVIRO-SEAL packing systems have beendesigned specifically for fugitive emissionapplications, these technologies should also beconsidered for any application where sealperformance and seal life have been an ongoingconcern or maintenance cost issue.

Characterization of Cage-GuidedValve BodiesIn valve bodies with cage-guided trim, the shape ofthe flow openings or windows in the wall of thecylindrical cage determines flow characterization.As the valve plug is moved away from the seatring, the cage windows are opened to permit flowthrough the valve. Standard cages have beendesigned to produce linear, equal-percentage, andquick-opening inherent flow characteristics. Notethe differences in the shapes of the cage windowsshown in figure 1-28. The flow rate/travelrelationship provided by valves utilizing thesecages is equivalent to the linear, quick-opening,and equal-percentage curves shown for contouredvalve plugs (figure 1-29).

Cage-guided trim in a control valve provides adistinct advantage over conventional valve bodyassemblies in that maintenance and replacementof internal parts is simplified. The inherent flowcharacteristic of the valve can easily be changedby installing a different cage. Interchange of cagesto provide a different inherent flow characteristicdoes not require changing the valve plug or seatring. The standard cages shown can be used witheither balanced or unbalanced trim constructions.Soft seating, when required, is available as aretained insert in the seat ring and is independentof cage or valve plug selection.

Cage interchangeability can be extended tospecialized cage designs that provide noiseattenuation or combat cavitation. These cagesfurnish a modified linear inherent flowcharacteristic, but require flow to be in a specific

1−16

Figure 1-28. Characterized Cages for Globe-Style Valve Bodies

W0958 W0959 W0957

QUICK OPENING LINEAR EQUAL PERCENTAGE

Figure 1-29. Inherent Flow Characteristics Curves

direction through the cage openings. Therefore, itcould be necessary to reverse the valve body inthe pipeline to obtain proper flow direction.

Characterized Valve Plugs

The valve plug, the movable part of a globe-stylecontrol valve assembly, provides a variablerestriction to fluid flow. Valve plug styles are eachdesigned to:

� Provide a specific flow characteristic.

� Permit a specified manner of guiding oralignment with the seat ring.

� Have a particular shutoff ordamage-resistance capability.

Valve plugs are designed for either two-position orthrottling control. In two-position applications, thevalve plug is positioned by the actuator at either oftwo points within the travel range of the assembly.In throttling control, the valve plug can bepositioned at any point within the travel range asdictated by the process requirements.

The contour of the valve plug surface next to theseat ring is instrumental in determining theinherent flow characteristic of a conventionalglobe-style control valve. As the actuator movesthe valve plug through its travel range, theunobstructed flow area changes in size and shapedepending upon the contour of the valve plug.When a constant pressure differential ismaintained across the valve, the changingrelationship between percentage of maximum flowcapacity and percentage of total travel range canbe portrayed (figure 1-29), and is designated asthe inherent flow characteristic of the valve.

Commonly specified inherent flow characteristicsinclude:

Linear Flow

� A valve with an ideal linear inherent flowcharacteristic produces a flow rate directlyproportional to the amount of valve plug travelthroughout the travel range. For instance, at 50%of rated travel, flow rate is 50% of maximum flow;at 80% of rated travel, flow rate is 80% ofmaximum; etc. Change of flow rate is constantwith respect to valve plug travel. Valves with alinear characteristic are often specified for liquidlevel control and for flow control applicationsrequiring constant gain.

Equal-Percentage Flow

� Ideally, for equal increments of valve plugtravel, the change in flow rate regarding travel maybe expressed as a constant percent of the flow

1−17

Figure 1-30. Typical Construction to ProvideQuick-Opening Flow Characteristic

A7100

rate at the time of the change. The change in flowrate observed regarding travel will be relativelysmall when the valve plug is near its seat, andrelatively high when the valve plug is nearly wideopen. Therefore, a valve with an inherentequal-percentage flow characteristic providesprecise throttling control through the lower portionof the travel range and rapidly increasing capacityas the valve plug nears the wide-open position.Valves with equal-percentage flow characteristicsare used on pressure control applications, onapplications where a large percentage of thepressure drop is normally absorbed by the systemitself with only a relatively small percentageavailable at the control valve, and on applicationswhere highly varying pressure drop conditions canbe expected. In most physical systems, the inletpressure decreases as the rate of flow increases,and an equal percentage characteristic isappropriate. For this reason, equal percentageflow is the most common valve characteristic.

Quick-Opening Flow

� A valve with a quick opening flowcharacteristic provides a maximum change in flowrate at low travels. The curve is essentially linearthrough the first 40 percent of valve plug travel,then flattens out noticeably to indicate littleincrease in flow rate as travel approaches thewide-open position. Control valves withquick-opening flow characteristics are often usedfor on/off applications where significant flow ratemust be established quickly as the valve begins toopen. As a result, they are often utilized in reliefvalve applications. Quick-opening valves can alsobe selected for many of the same applications forwhich linear flow characteristics arerecommended. This is because the quick-openingcharacteristic is linear up to about 70 percent ofmaximum flow rate. Linearity decreasessignificantly after flow area generated by valveplug travel equals the flow area of the port. For atypical quick-opening valve (figure 1-30), thisoccurs when valve plug travel equals one-fourth ofport diameter.

Valve Plug GuidingAccurate guiding of the valve plug is necessary forproper alignment with the seat ring and efficientcontrol of the process fluid. The common methodsused are listed below.

� Cage Guiding: The outside diameter of thevalve plug is close to the inside wall surface of thecylindrical cage throughout the travel range. Sincethe bonnet, cage, and seat ring are self-aligningupon assembly, the correct valve plug and seatring alignment is assured when the valve closes(figure 1-15).

� Top Guiding: The valve plug is aligned by asingle guide bushing in the bonnet, valve body(figure 1-4), or by packing arrangement.

� Stem Guiding: The valve plug is aligned withthe seat ring by a guide bushing in the bonnet thatacts upon the valve plug stem (figure 1-3, leftview).

� Top-and-Bottom Guiding: The valve plug isaligned by guide bushings in the bonnet andbottom flange.

� Port Guiding: The valve plug is aligned by thevalve body port. This construction is typical forcontrol valves utilizing small-diameter valve plugswith fluted skirt projections to control low flow rates(figure 1-3, right view).

Restricted-Capacity Control ValveTrimMost control valve manufacturers can providevalves with reduced- or restricted- capacity trimparts. The reduced flow rate might be desirable forany of the following reasons:

� Restricted capacity trim may make it possibleto select a valve body large enough for increasedfuture flow requirements, but with trim capacityproperly sized for present needs.

� Valves can be selected for adequatestructural strength, yet retain reasonabletravel/capacity relationship.

� Large bodies with restricted capacity trim canbe used to reduce inlet and outlet fluid velocities.

� Purchase of expensive pipeline reducers canbe avoided.

� Over-sizing errors can be corrected by use ofrestricted capacity trim parts.

1−18

Conventional globe-style valve bodies can be fittedwith seat rings with smaller port size than normaland valve plugs sized to fit those smaller ports.Valves with cage-guided trim often achieve thereduced capacity effect by utilizing valve plug,cage, and seat ring parts from a smaller valve sizeof similar construction and adapter pieces abovethe cage and below the seat ring to mate thosesmaller parts with the valve body (figure 1-28).Because reduced capacity service is not unusual,leading manufacturers provide readily availabletrim part combinations to perform the requiredfunction. Many restricted capacity trimcombinations are designed to furnishapproximately 40% of full-size trim capacity.

General Selection CriteriaMost of the considerations that guide the selectionof valve type and brand are rather basic. However,there are some matters that may be overlooked byusers whose familiarity is mainly limited to just oneor a few valve types. Table 1-1 below provides achecklist of important criteria; each is discussed atlength following the table.

Table 1-1. Suggested General Criteria for Selecting Typeand Brand of Control Valve

Body pressure rating

High and low temperature limits

Material compatibility and durability

Inherent flow characteristic and rangeability

Maximum pressure drop (shutoff and flowing)

Noise and cavitation

End connections

Shutoff leakage

Capacity versus cost

Nature of flowing media

Dynamic performance

Pressure RatingsBody pressure ratings ordinarily are consideredaccording to ANSI pressure classes — the mostcommon ones for steel and stainless steel beingClasses 150, 300 and 600. (Source documentsare ASME/ANSI Standards B16.34, “SteelValves,” and ANSI B16.1, “Cast Iron PipeFlanges and Flanged Fittings.”) For a given bodymaterial, each NSI Class corresponds to aprescribed profile of maximum pressures thatdecrease with temperature according to thestrength of the material. Each material also has a

minimum and maximum service temperaturebased upon loss of ductility or loss of strength. Formost applications, the required pressure rating isdictated by the application. However, because allproducts are not available for all ANSI Classes, itis an important consideration for selection.

Temperature ConsiderationsRequired temperature capabilities are also aforegone conclusion, but one that is likely tonarrow valve selection possibilities. Theconsiderations include the strength or ductility ofthe body material, as well as relative thermalexpansion of various parts.

Temperature limits also may be imposed due todisintegration of soft parts at high temperatures orloss of resiliency at low temperatures. The softmaterials under consideration include variouselastomers, plastics, and PTFE. They may befound in parts such as seat rings, seal or pistonrings, packing, rotary shaft bearings and butterflyvalve liners. Typical upper temperature limits forelastomers are in the 200 - 350°F range, and thegeneral limit for PTFE is 450°F.

Temperature affects valve selection by excludingcertain valves that do not have high or lowtemperature options. It also may have some affecton the valve’s performance. For instance, goingfrom PTFE to metal seals for high temperaturesgenerally increases the shutoff leakage flow.Similarly, high temperature metal bearing sleevesin rotary valves impose more friction upon theshaft than do PTFE bearings, so that the shaftcannot withstand as high a pressure-drop load atshutoff. Selection of the valve packing is alsobased largely upon service temperature.

Material SelectionThe third criterion in table 1-1, “materialcompatibility and durability”, is a more complexconsideration. Variables may include corrosion bythe process fluid, erosion by abrasive material,flashing, cavitation or pressure and temperaturerequirements. The piping material usually indicatesthe body material. However, because the velocityis higher in valves, other factors must beconsidered. When these variables are included,often valve and piping materials will differ. The trimmaterials, in turn, are usually a function of thebody material, temperature range and qualities ofthe fluid. When a body material other than carbon,alloy, or stainless steel is required, use of analternate valve type, such as lined or bar stock,should be considered.

1−19

Flow CharacteristicThe next selection criterion, “inherent flowcharacteristic”, refers to the pattern in which theflow at constant pressure drop changes accordingto valve position. Typical characteristics arequick-opening, linear, and equal-percentage. Thechoice of characteristic may have a stronginfluence upon the stability or controllability of theprocess (see table 1-3), as it represents thechange of valve gain relative to travel.

Most control valves are carefully “characterized”by means of contours on a plug, cage, or ballelement. Some valves are available in a variety ofcharacteristics to suit the application, while othersoffer little or no choice. To quantitatively determinethe best flow characteristic for a given application,a dynamic analysis of the control loop can beperformed. In most cases, however, this isunnecessary; reference to established rules ofthumb will suffice.

The accompanying drawing illustrates typical flowcharacteristic curves (figure 1-29). The quickopening flow characteristic provides for maximumchange in flow rate at low valve travels with a fairlylinear relationship. Additional increases in valvetravel give sharply reduced changes in flow rate,and when the valve plug nears the wide openposition, the change in flow rate approaches zero.In a control valve, the quick opening valve plug isused primarily for on-off service; but it is alsosuitable for many applications where a linear valveplug would normally be specified.

RangeabilityAnother aspect of a valve’s flow characteristic is itsrangeability, which is the ratio of its maximum andminimum controllable flow rates. Exceptionallywide rangeability may be required for certainapplications to handle wide load swings or acombination of start-up, normal and maximumworking conditions. Generally speaking, rotaryvalves—especially partial ball valves—havegreater rangeability than sliding-stem varieties.

Use of PositionersA positioner is an instrument that helps improvecontrol by accurately positioning a control valveactuator in response to a control signal. They areuseful in many applications and are required withcertain actuator styles in order to match actuatorand instrument pressure signals, or to provide

operating stability. To a certain extent, a valve withone inherent flow characteristic can also be madeto perform as though it had a differentcharacteristic by utilizing a nonlinear (i.e.,characterized) positioner-actuator combination.The limitation of this approach lies in thepositioner’s frequency response and phase lagcompared to the characteristic frequency of theprocess. Although it is common practice to utilize apositioner on every valve application, eachapplication should be reviewed carefully. Thereare certain examples of high gain processeswhere a positioner can hinder valve performance.

Pressure DropThe maximum pressure drop a valve can tolerateat shutoff, or when partially or fully open, is animportant selection criteria. Sliding-stem valvesare generally superior in both regards because ofthe rugged nature of their moving parts. Manyrotary valves are limited to pressure drops wellbelow the body pressure rating, especially underflowing conditions, due to dynamic stresses thathigh velocity flow imposes on the disk or ballsegment.

Noise and CavitationNoise and cavitation are two considerations thatoften are grouped together because both resultfrom high pressure drops and large flow rates.They are treated by special modifications tostandard valves. Chapter four discusses thecavitation phenomenon and its impact andtreatment, while chapter six discusses noisegeneration and abatement.

End ConnectionsThe three common methods of installing controlvalves in pipelines are by means of screwed pipethreads, bolted flanges, and welded endconnections. At some point in the selectionprocess, the valve’s end connections must beconsidered with the question simply being whetherthe desired connection style is available in thevalve being considered.

In some situations, this matter can limit theselection rather narrowly. For instance, if a pipingspecification calls for welded connections only, thechoice usually is limited to sliding-stem valves.

Screwed end connections, popular in small controlvalves, offer more economy than flanged ends.

1−20

The threads usually specified are tapered femaleNPT on the valve body. They form ametal-to-metal seal by wedging over the matingmale threads on the pipeline ends. Thisconnection style is usually limited to valves notlarger than NPS 2, and is not recommended forelevated temperature service.

Valve maintenance might be complicated byscrewed end connections if it is necessary to takethe body out of the pipeline. Screwed connectionsrequire breaking a flanged joint or unionconnection to permit unscrewing the valve bodyfrom the pipeline.

Flanged end valves are easily removed from thepiping and are suitable for use through the rangeof working pressures that most control valves aremanufactured (figure 1-13).

Flanged end connections can be utilized in atemperature range from absolute zero (−273°F) toapproximately 1500°F (815°C). They are utilizedon all valve sizes. The most common flanged endconnections include flat face, raised face, and ringtype joint.

Welded ends on control valves are leak-tight at allpressures and temperatures and are economicalin initial cost (figure 1-14). Welded end valves aremore difficult to remove from the line and arelimited to weldable materials. Welded ends comein two styles, socket weld and buttweld.

Shutoff CapabilitySome consideration must be given to a valve’sshutoff capability, which is usually rated in terms ofclasses specified in ANSI/FCI70-2 (table 1-4). Inservice, shutoff leakage depends upon manyfactors, including but not limited to, pressure drop,temperature, and the condition of the sealingsurfaces. Because shutoff ratings are based uponstandard test conditions that can be different fromservice conditions, service leakage cannot bepredicted accurately. However, the shutoff classprovides a good basis for comparison amongvalves of similar configuration. It is not uncommonfor valve users to overestimate the shutoff classrequired.

Because tight shutoff valves generally cost moreboth in initial cost, as well as in later maintenanceexpense, serious consideration is warranted. Tightshutoff is particularly critical in high pressurevalves, considering that leakage in theseapplications can lead to the ultimate destruction of

the trim. Special precautions in seat materialselection, seat preparation and seat load arenecessary to ensure success.

Flow CapacityFinally, the criterion of capacity or size can be anoverriding constraint on selection. For extremelylarge lines, sliding-stem valves are moreexpensive than rotary types. On the other hand,for extremely small flows, a suitable rotary valvemay not be available. If future plans call forsignificantly larger flow, then a sliding-stem valvewith replaceable restricted trim may be theanswer. The trim can be changed to full size trimto accommodate higher flow rates at less cost thanreplacing the entire valve body assembly.

Rotary style products generally have much highermaximum capacity than sliding-stem valves for agiven body size. This fact makes rotary productsattractive in applications where the pressure dropavailable is rather small. However, it is of little orno advantage in high pressure drop applicationssuch as pressure regulation or letdown.

ConclusionFor most general applications, it makes senseboth economically, as well as technically, to usesliding-stem valves for lower flow ranges, ballvalves for intermediate capacities, and highperformance butterfly valves for the very largestrequired flows. However, there are numerousother factors in selecting control valves, andgeneral selection principles are not always thebest choice.

Selecting a control valve is more of and art than ascience. Process conditions, physical fluidphenomena, customer preference, customerexperience, supplier experience, among numerousother criteria must be considered in order to obtainthe best possible solution. Many applications arebeyond that of general service, and as chapter 4will present, there are of number of selectioncriteria that must be considered when dealing withthese sometimes severe flows.

Special considerations may require out-of-the-ordinary valve solutions; there are valve designsand special trims available to handle high noiseapplications, flashing, cavitation, high pressure,high temperature and combinations of theseconditions.

1−21

After going through all the criteria for a givenapplication, the selection process may point toseveral types of valves. From there on, selectionbecomes a matter of price versus capability,coupled with the inevitable personal and

institutional preferences. As no single control valvepackage is cost-effective over the full range ofapplications, it is important to keep an open mindto alternative choices.

Table 1-2. Major Categories and Subcategories of Control Valves with Typical General Characteristics

Valve Style MainCharacteristics

Typical SizeRange,inches

TypicalStandard Body

Materials

Typical StandardEnd Connection

TypicalPressureRatings

Relative FlowCapacity

RelativeShutoff

Capability

RegularSliding-stem

Heavy DutyVersatile

1 to 24Carbon Steel

Cast IronStainless

ANSI FlangedWeldedScrewed

To ANSI 2500 Moderate Excellent

Bar Stock Machined from BarStock

½ to 3 Variety of Alloys FlangelessScrewed

To ANSI 600 Low Excellent

EconomySliding-stem

Light DutyInexpensive

½ to 2Bronze

Cast IronCarbon Steel

Screwed To ANSI 125 Moderate Good

Thru-BoreBall

On-Of f Service 1 to 24 Carbon SteelStainless

Flangeless To ANSI 900 High Excellent

Partial Ball Characterized forThrottling

1 to 24 Carbon SteelStainless

FlangelessFlanged

To ANSI 600 High Excellent

Eccentric Plug Erosion Resistance 1 to 8 Carbon SteelStainless

Flanged To ANSI 600 Moderate Excellent

Swing-ThruButterfly

No Seal 2 to 96Carbon Steel

Cast IronStainless

FlangelessLuggedWelded

To ANSI 2500 High Poor

Lined Butterfly Elastomer orTFE Liner

2 to 96Carbon Steel

Cast IronStainless

FlangelessLugged

To ANSI 300 High Good

HighPerformance

Butterfly

Offset DiskGeneral Service

2 to 72 Carbon SteelStainless

FlangelessLugged

To ANSI 600 High Excellent

1−22

Table 1-3. Control Valve Characteristic Recommendations

Liquid Level Systems

Control Valve Pressure Drop Best InherentCharacteristic

Constant ΔP Linear

Decreasing ΔP with increasing load, ΔP at maximum load > 20% of minimum load ΔP Linear

Decreasing ΔP with increasing load, ΔP at maximum load < 20% of minimum load ΔP Equal-percentage

Increasing ΔP with increasing load, ΔP at maximum load < 200% of minimum load ΔP Linear

Increasing ΔP with increasing load, ΔP at maximum load > 200% of minimum load ΔP Quick Opening

Pressure Control Systems

Application Best InherentCharacteristic

Liquid Process Equal-Percentage

Gas Process, Large Volume (Process has a receiver, Distribution System or Transmission Line Exceeding 100 ft. ofNominal Pipe Volume), Decreasing ΔP with Increasing Load, ΔP at Maximum Load > 20% of Minimum Load ΔP Linear

Gas Process, Large Volume, Decreasing ΔP with Increasing Load, ΔP at Maximum Load < 20% of Minimum Load ΔP Equal-Percentage

Gas Process, Small Volume, Less than 10 ft. of Pipe between Control Valve and Load Valve Equal-Percentage

Flow Control Processes

Application Best Inherent Characteristic

Flow Measurement Signal toController

Location of Control Valve in Relationto Measuring Element

Wide Range of Flow Set PointSmall Range of Flow but

Large ΔP Change at Valvewith Increasing Load

Proportional to Flow In Series Linear Equal-Percentage

In Bypass* Linear Equal-Percentage

Proportional to Flow Squared In Series Linear Equal-Percentage

In Bypass* Equal-Percentage Equal-Percentage

*When control valve closes, flow rate increases in measuring element.

1−23

Table 1-4. Control Valve Leakage StandardsANSI

B16.104-1976Maximum Leakage Test Medium Pressure and Temperature

Class II 0.5% valve capacity at full travel Air Service ΔP or 50 psid (3.4 bar differential),whichever is lower, at 50� or 125�F (10� to 52�C)

Class III 0.1% valve capacity at full travel Air Service ΔP or 50 psid (3.4 bar differential),whichever is lower, at 50� or 125�F (10� to 52�C)

Class IV 0.01% valve capacity at full travel Air Service ΔP or 50 psid (3.4 bar differential),whichever is lower, at 50� or 125�F (10� to 52�C)

Class V 5 x 10- 4 mL/min/psid/inch port dia. (5x 10- 12 m3/sec/Δbar/mm port dia)

Water Service ΔP at 50� or 125�F (10� to 52�C)

Class VI Nominal PortDiameter

Bubbles perMinute

mL per Minute TestMedium

Pressure and Temperature

In1

1-1/22

2-1/23468

mm2538516476102152203

12346112745

0.150.300.450.600.901.704.006.75

AirService ΔP or 50 psid (3.4 bar

differential), whichever is lower, at 50�or 125�F (10� to 52�C)

Copyright 1976 Fluid Controls Institute, Inc. Reprinted with permission.

1−24

www.Fisher.com

Chapter 2

Actuator Selection

The actuator is the distinguishing element thatdifferentiates control valves from other types ofvalves. The first actuated valves were designed inthe late 19th century. Today, they would be betterdescribed as regulators since they operateddirectly from the process fluid. These “automaticvalves” were the mainstay of industry through theearly 1930s.

It was at this time that the first pneumaticcontrollers were used. Development of valvecontrollers and the adaptation of standardizedcontrol signals stimulated design of the first, true,control valve actuators.

The control valve industry has evolved to fill avariety of needs and desires. Actuators areavailable with an array of designs, power sourcesand capabilities. Proper selection involves processknowledge, valve knowledge, and actuatorknowledge.

A control valve can perform its function only aswell as the actuator can handle the static anddynamic loads placed on it by the valve.Therefore, proper selection and sizing are veryimportant. Since the actuator can represent asignificant portion of the total control valve price,careful selection of actuator and accessory optionscan lead to significant dollar savings.

The range of actuator types and sizes on themarket today is so great that it seems the selectionprocess might be highly complex. With a few rulesin mind and knowledge of fundamental needs, theselection process can be simple.

The following parameters are key as they quicklynarrow the actuator choices:

� Power source availability

� Fail-safe requirements

� Torque or thrust requirements

� Control functions

Power Source AvailabilityThe power source available at the location of avalve can often point directly to what type ofactuator to choose. Typically, valve actuators arepowered either by compressed air or by electricity.However, in some cases water pressure, hydraulicfluid, or even pipeline pressure can be used.

Since most plants have both electricity andcompressed air readily available, the selectiondepends upon the ease and cost of furnishingeither power source to the actuator location.Reliability and maintenance requirements of thepower system must also be considered.Consideration should also be given to providingbackup operating power to critical plant loops.

Fail-safe RequirementsThe overall reliability of power sources is quitehigh. However, many loops demand specific valveaction should the power source ever fail. Desiredaction upon a signal failure may be required forsafety reasons or for protection of equipment.

Fail-safe systems store energy, eithermechanically in springs, pneumatically in volumetanks, or in hydraulic accumulators. When powerfails, the fail-safe systems are triggered to drivethe valves to the required position and to thenmaintain this position until returned to normaloperation. In many cases, the process pressure isused to ensure or enhance this action.

2−2

Actuator designs are available with a choice offailure mode between failing open, failing closed,or holding in the last position. Many actuatorsystems incorporate failure modes at no extracost. For example, spring-and-diaphragmactuators are inherently fail open or closed, whileelectric operators typically hold their last position.

Torque or Thrust RequirementsAn actuator must have sufficient thrust or torquefor the prescribed application. In some cases thisrequirement can dictate actuator type as well aspower supply requirements.

For instance, large valves requiring a high thrustmay be limited to only electric or electro-hydraulicactuators due to a lack of pneumatic actuators withsufficient thrust capability. Conversely,electro-hydraulic actuators would be a poor choicefor valves with very low thrust requirements.

The matching of actuator capability with valvebody requirements is best left to the control valvemanufacturer as there are considerabledifferences in frictional and fluid forces from valveto valve.

Control FunctionsKnowledge of the required actuator functions willmost clearly define the options available forselection. These functions include the actuatorsignal (pneumatic, electric, etc.), signal range,ambient temperatures, vibration levels, operatingspeed, frequency, and quality of control that isrequired.

Signal types are typically grouped as such:

� Two-position (on-off)

� Analog (throttling)

� Digital

Two-position electric, electro-pneumatic, orpneumatic switches control on-off actuators. Thisis the simplest type of automatic control and theleast restrictive in terms of selection.

Throttling actuators have considerably higherdemands put on them from both a compatibilityand performance standpoint. A throttling actuatorreceives its input from an electronic or pneumaticinstrument that measures the controlled processvariable. The actuator must then move the finalcontrol element in response to the instrumentsignal in an accurate and timely fashion to ensureeffective control. The two primary additionalrequirements for throttling actuators include:

� Compatibility with instrument signal

� Better static and dynamic performance toensure loop stability

Compatibility with instrument signals is inherent inmany actuator types, or it can be obtained withadd-on equipment. But, the high-performancecharacteristics required of a good throttlingactuator cannot be bolted on; instead, lowhysteresis and minimal deadband must bedesigned into actuators.

Stroking speed, vibration, and temperatureresistance must also be considered if critical to theapplication. For example, on liquid loopsfast-stroking speeds can be detrimental due to thepossibility of water hammer.

Vibration or mounting position can be a potentialproblem. The actuator weight, combined with theweight of the valve, may necessitate bracing.

It is essential to determine the ambienttemperature and humidity that the actuator willexperience. Many actuators contain eitherelastomeric or electronic components that can besubject to degradation by high humidity ortemperature.

EconomicsEvaluation of economics in actuator selection is acombination of the following:

� Cost

� Maintenance

� Reliability

A simple actuator, such as aspring-and-diaphragm, has few moving parts andis easy to service. Its initial cost is low, andmaintenance personnel understand and arecomfortable working with them.

2−3

An actuator made specifically for a control valveeliminates the chance for a costly performancemismatch. An actuator manufactured by the valvevendor and shipped with the valve will eliminateseparate mounting charges and ensure easiercoordination of spare parts procurement.Interchangeable parts among varied actuators arealso important to minimize spare-parts inventory.

Actuator DesignsThere are many types of actuators on the market,most of which fall into five general categories:

� Spring-and-diaphragm

� Pneumatic piston

� Rack and Pinion

� Electric motor

� Electro-hydraulic

Each actuator design has weaknesses, strongpoints and optimum uses. Most actuator designsare available for either sliding stem or rotary valvebodies. They differ only by linkages or motiontranslators; the basic power sources are identical.

Most rotary actuators employ linkages, gears, orcrank arms to convert direct linear motion of adiaphragm or piston into the 90-degree outputrotation required by rotary valves. The mostimportant consideration for control valve actuatorsis the requirement for a design that limits theamount of lost motion between internal linkageand valve coupling.

Rotary actuators are now available that employtilting pistons or diaphragms. These designseliminate most linkage points (and resultant lostmotion) and provide a safe, accurate and enclosedpackage.

When considering an actuator design, it is alsonecessary to consider the method by which it iscoupled to the drive shaft of the control valve.Slotted connectors mated to milled shaft flats aregenerally not satisfactory if any degree ofperformance is required. Pinned connections, ifsolidly constructed, are suitable for nominal torqueapplications. A splined connector that mates to a

splined shaft end and then is rigidly clamped to theshaft eliminates lost motion, is easy todisassemble, and is capable of high torque.

Sliding stem actuators are rigidly fixed to valvestems by threaded and clamped connections.Because they don’t have any linkage points, andtheir connections are rigid, they exhibit no lostmotion and have excellent inherent controlcharacteristics.

Spring-and-Diaphragm ActuatorsThe most popular and widely used control valveactuator is the pneumatic spring-and-diaphragmstyle. These actuators are extremely simple andoffer low cost and high reliability. They normallyoperate over the standard signal ranges of 3 to 15psi or 6 to 30 psi, and therefore, are often suitablefor throttling service using instrument signalsdirectly.

Many spring-and-diaphragm designs offer eitheradjustable springs and/or wide spring selections toallow the actuator to be tailored to the particularapplication. Because they have few moving partsthat may contribute to failure, they are extremelyreliable. Should they ever fail, maintenance isextremely simple. Improved designs now includemechanisms to control the release of springcompression, eliminating possible personnel injuryduring actuator disassembly.

Use of a positioner or booster with aspring-and-diaphragm actuator can improvecontrol, but when improperly applied, can result inpoor control. Follow the simple guidelinesavailable for positioner applications and look for:

� Rugged, vibration-resistant construction

� Calibration ease

� Simple, positive feedback linkages

The overwhelming advantage of thespring-and-diaphragm actuator is the inherentprovision for fail-safe action. As air is loaded onthe actuator casing, the diaphragm moves thevalve and compresses the spring. The storedenergy in the spring acts to move the valve back toits original position as air is released from thecasing. Should there be a loss of signal pressureto the instrument or the actuator, the spring canmove the valve to its initial (fail-safe) position.

2−4

DIAPHRAGM

DIAPHRAGM CASING

DIAPHRAGMPLATE

ACTUATOR SPRING

ACTUATOR STEM

SPRING SEAT

SPRING ADJUSTOR

STEM CONNECTOR

YOKE

TRAVEL INDICATOR DISK

INDICATOR SCALE

W0363-1

LOWER DIAPHRAGM CASING

W0364-1

Figure 2-1. Spring-and-diaphragm actuators offer an excellent first choice for most control valves. They are inexpensive, simple and have built-in, fail-safe action. Pictured above are cutaways of the popular

Fisher 667 (left) and Fisher 657 (right) actuators.

Figure 2-2. Spring-and-diaphragm actuatorscan be supplied with a top-mounted handwheel.

The handwheel allows manual operation and alsoacts as a travel stop or means of emergency

operation.

W0368-2

Actuators are available for either fail-open orfail-closed action. The only drawback to thespring-and-diaphragm actuator is a relativelylimited output capability. Much of the thrustcreated by the diaphragm is taken up by the springand thus does not result in output to the valve.

Therefore, the spring-and-diaphragm actuator isused infrequently for high force requirements. It isnot economical to build and use very largespring-and-diaphragm actuators because the size,weight and cost grow exponentially with eachincrease in output force capability.

Piston ActuatorsPiston actuators are generally more compact andprovide higher torque or force outputs thanspring-and-diaphragm actuators. Fisher pistonstyles normally work with supply pressuresbetween 50 and 150 psi and can be equipped withspring returns (however, this construction haslimited application).

Piston actuators used for throttling service must befurnished with double-acting positioners thatsimultaneously load and unload opposite sides ofthe piston. The pressure differential created acrossthe piston causes travel toward the lower pressureside. The positioner senses the motion, and whenthe required position is reached, the positionerequalizes the pressure on both sides of the piston.

The pneumatic piston actuator is an excellentchoice when a compact unit is required to producehigh torque or force. It is also easily adapted to

2−5

Figure 2-3. The Fisher 2052 spring-and-diaphragm actuator has many features to provideprecise control. The splined actuator connectionfeatures a clamped lever and single-joint linkage

to help eliminate lost motion.

W9589-1 W9588-1

Figure 2-4. Double-acting piston actuators suchas the Fisher 1061 rotary actuator are a goodchoice when thrust requirements exceed thecapability of spring-and-diaphragm actuators.

Piston actuators require a higher supply pressure,but have benefits such as high stiffness and small

size. The 1061 actuator is typically used forthrottling service.

W3827−1

services where high ambient temperatures are aconcern.

The main disadvantages of piston actuators arethe high supply pressures required for positionerswhen used in throttling service and the lack offail-safe systems.

Figure 2-5. Spring fail-safe is present in thispiston design. The Fisher 585C is an example ofa spring-bias piston actuator. Process pressurecan aid fail-safe action, or the actuator can be

configured for full spring-fail closure.

W7447

Figure 2-6. Since the requirements for accuracyand minimal lost motion are unnecessary for

on-off service, cost savings can be achieved by simplifying the actuator design. The Fisher

1066SR incorporates spring-return capability.

W4102

There are two types of spring-return pistonactuators available. The variations are subtle, butsignificant. It is possible to add a spring to a piston

2−6

actuator and operate it much like a spring-and-diaphragm. These designs use a single-actingpositioner that loads the piston chamber to movethe actuator and compress the spring. As air isunloaded, the spring forces the piston back. Thesedesigns use large, high output springs that arecapable of overcoming the fluid forces in the valve.

The alternative design uses a much smaller springand relies on valve fluid forces to help provide thefail-safe action. In normal operation they act like adouble action piston. In a fail-safe situation thespring initiates movement and is helped byunbalance forces on the valve plug. Theseactuators can be sized and set up to provide fullspring closure action without process assistance.

An alternative to springs is a pneumatic tripsystem which often proves to be complex indesign, difficult to maintain and costly. While a tripsystem is completely safe, any fail-saferequirement consideration should be given first tospring-and-diaphragm operators if they arefeasible.

Special care should be given during the selectionof throttling piston actuators to specify a designthat has minimal hysteresis and deadband. As thenumber of linkage points in the actuator increases,so does the deadband. As the number of slidingparts increases, so does the hysteresis. Anactuator with high hysteresis and deadband canbe quite suitable for on-off service; however,caution is necessary when attempting to adapt thisactuator to throttling service by merely bolting on apositioner.

The cost of a spring-and-diaphragm actuator isgenerally less than a comparable piston actuator.Part of this cost saving is a result of the ability touse instrument output air directly, therebyeliminating the need for a positioner. The inherentprovision for fail-safe action in the spring-and-diaphragm actuator is also a consideration.

Rack and Pinion Actuators

Rack and pinion actuators may come in adouble-acting design, or spring return, and are acompact and economical solution for rotary shaftvalves. They provide high torque outputs and aretypically used for on-off applications with highcycle life. They may also be used in processeswhere higher variability is not a concern.

Figure 2-7. The FieldQ� actuator is a quarterturn pneumatic rack and pinion actuator. It comeswith an integrated module combining the solenoid

and switchbox into a low profile, compactpackage.

W9479

Electric ActuatorsElectric actuators can be applied successfully inmany situations. Most electric operators consist ofmotors and gear trains and are available in a widerange of torque outputs, travels, and capabilities.They are suited for remote mounting where noother power source is available, for use wherethere are specialized thrust or stiffnessrequirements, or when highly precise control isrequired.

Electric operators are economical versuspneumatic actuators for applications in small sizeranges only. Larger units operate slowly and weighconsiderably more than pneumatic equivalents.Available fail action is typically lock in last position.

One key consideration in choosing an electricactuator is its capability for continuous closed-loopcontrol. In applications where frequent changesare made in control-valve position, the electricactuator must have a suitable duty cycle.

High performance electric actuators usingcontinuous rated DC motors and ball screw outputdevices are capable of precise control and 100%duty cycles.

Compared to other actuator designs, the electricactuator generally provides the highest output

2−7

available within a given package size. Additionally,electric actuators are stiff, that is, resistant to valveforces. This makes them an excellent choice forgood throttling control of large, high-pressurevalves.

Actuator SizingThe last step in the selection process is todetermine the required actuator size.Fundamentally, the process of sizing is to matchas closely as possible the actuator capabilities tothe valve requirements.

In practice, the mating of actuator and valverequires the consideration of many factors. Valveforces must be evaluated at the critical positions ofvalve travel (usually open and closed) andcompared to actuator output. Valve forcecalculation varies considerably between valvestyles and manufacturers. In most cases it isnecessary to consider a complex summation offorces including:

� Static fluid forces

� Dynamic fluid forces and force gradients

� Friction of seals, bearings, and packing

� Seat loading

Although actuator sizing is not difficult, the greatvariety of designs on the market and the readyavailability of vendor expertise (normally at nocost) make detailed knowledge of the proceduresunnecessary.

Actuator Spring for Globe ValvesThe force required to operate a globe valveincludes:

A. Force to overcome static unbalance of thevalve plug

B. Force to provide a seat load

C. Force to overcome packing friction

D. Additional forces required for certain specificapplications or constructions

Total force required = A + B + C + D

A. Unbalance ForceThe unbalance force is that resulting from fluidpressure at shutoff, and in the most general sensecan be expressed as:

Unbalance force = net pressure differential X netunbalance area

Frequent practice is to take the maximumupstream gauge pressure as the net pressuredifferential unless the process design alwaysensures a back pressure at the maximum inletpressure. Net unbalance area is the port area on asingle seated flow up design. Unbalance area mayhave to take into account the stem area dependingon configuration. For balanced valves there is stilla small unbalance area. This data can be obtainedfrom the manufacturer. Typical port areas forbalanced valves flow up and unbalanced valves ina flow down configuration are listed in table 2-1.

Table 2-1. Typical Unbalance Areas of Control Valves

Port Diameter,Inches

Unbalance AreaSingle-SeatedUnbalancedValves, In2

Unbalance AreaBalanced Valves,

In2

1/4 0.049 – – –

3/8 0.110 – – –

1/2 0.196 – – –

3/4 0.441 – – –

1 0.785 – – –

1 5/16 1.35 0.04

1 7/8 2.76 0.062

2 5/16 4.20 0.27

3 7/16 9.28 0.118

4 3/8 15.03 0.154

7 38.48 0.81

8 50.24 0.86

B. Force to Provide Seat LoadSeat load, usually expressed in pounds per linealinch or port circumference, is determined byshutoff requirements. Use the guidelines in table2-2 to determine the seat load required to meetthe factory acceptance tests for ANSI/FCI 70-2and IEC 534-4 leak Classes II through VI.

Because of differences in the severity of serviceconditions, do not construe these leakclassifications and corresponding leakage rates asindicators of field performance. To prolong seat lifeand shutoff capabilities, use a higher thanrecommended seat load. If tight shutoff is not aprime consideration, use a lower leak class.

2−8

Table 2-2. Recommended Seat Load Per Leak Class forControl Valves

Class I As required by customerspecification, no factory leak testrequired

Class II 20 pounds per lineal inch of portcircumference

Class III 40 pounds per lineal inch of portcircumference

Class IV Standard (Lower) Seat only—40pounds per lineal inch of portcircumference (up through a4–3/8 inch diameter port)Standard (Lower) Seat only—80pounds per lineal inch of portcircumference (larger than 4–3/8inch diameter port)

Class V Metal Seat—determine poundsper lineal inch of portcircumference from figure 2-9

C. Packing FrictionPacking friction is determined by stem size,packing type, and the amount of compressive loadplaced on the packing by the process or thebolting. Packing friction is not 100% repeatable inits friction characteristics. Newer live loadedpacking designs can have significant friction forcesespecially if graphite packing is used. Table 2-3lists typical packing friction values.

D. Additional ForcesAdditional forces to consider may include bellowsstiffness, unusual frictional forces resulting fromseals or special seating forces for soft metal seals.The manufacturer should either supply thisinformation or take it into account when sizing anactuator.

Actuator Force CalculationsPneumatic spring-and-diaphragm actuatorsprovide a net force with the additional air pressureafter compressing the spring in air-to-close, or withthe net pre-compression of the spring inair-to-open. This may be calculated in pounds persquare inch of pressure differential.

For example, suppose 275 pound-force (lbf) isrequired to close the valve as calculated per theprocess described earlier. An air-to-open actuatorwith 100 square inches of diaphragm area and abench set of 6 to 15 psig is one available option.The expected operating range is 3 to 15 psig. The

Figure 2-8. Recommended Seat Load

A2222−4/IL

pre-compression can be calculated as thedifference between the lower end of the bench set(6 psig) and the beginning of the operating range(3 psig). This 3 psig is used to overcome thepre-compression so the net pre-compression forcemust be:

3 psig X 100 sq. in. = 300 lbf.

This exceeds the force required and is anadequate selection.

Piston actuators with springs are sized in the samemanner. The thrust from piston actuators withoutsprings can be calculated as:

Piston area X minimum supply pressure =minimum available thrust(maintain compatibility of units)

In some circumstances an actuator could supplytoo much force and cause the stem to buckle, tobend sufficiently to cause a leak, or to damagevalve internals.

The manufacturer normally takes responsibility foractuator sizing and should have methodsdocumented to check for maximum stem loads.Manufacturers also publish data on actuatorthrusts, effective diaphragm areas, and springdata.

2−9

Table 2-3. Typical Packing Friction Values (Lb)Stem Size(Inches)

ANSIClass

PTFE Packing GraphiteRibbon/FilamentSingle Double

5/16 All 20 30 – – –

3/8 125150250300

38 56 – – –125

– – –190

600900

1500

250320380

1/2 125150250300

50 75 – – –180

– – –230

600900

15002500

320410500590

5/8 125150250300600

63 95 – – –218

– – –290400

3/4 125150250300

75 112.5 – – –350

– – –440

600900

15002500

66088011001320

1 300600900

15002500

100 150 610850

106013001540

1–1/4 300600900

15002500

120 180 8001100140017002040

2 300600900

15002500

200 300 12251725225027503245

Values shown are frictional forces typically encountered when using standardpacking flange bolt-torquing procedures.

Actuator Sizing for Rotary ValvesIn selecting the most economical actuator for arotary valve, the determining factors are the torquerequired to open and close the valve and thetorque output of the actuator.

This method assumes the valve has been properlysized for the application and the application doesnot exceed pressure limitations for the valve.

Torque EquationsRotary valve torque equals the sum of a number oftorque components. To avoid confusion, a numberof these have been combined, and a number ofcalculations have been performed in advance.Thus, the torque required for each valve type canbe represented with two simple and practicalequations.

Breakout Torque

TB=A(�Pshutoff)+B

Dynamic Torque

TD=C(�Peff)

Specific A, B, and C factors, for example, rotaryvalve designs are included in tables 2-4 and 2-5.

Maximum RotationMaximum rotation is defined as the angle of valvedisk or ball in the fully open position.

Normally, maximum rotation is 90 degrees. Theball or disk rotates 90 degrees from the closedposition to the wide-open position.

Some of the pneumatic spring-return piston andpneumatic spring-and-diaphragm actuators arelimited to 60 or 75 degrees rotation.

For pneumatic spring-and-diaphragm actuators,limiting maximum rotation allows for higher initialspring compression, resulting in more actuatorbreakout torque. Additionally, the effective lengthof each actuator lever changes with valve rotation.Published torque values, particularly for pneumaticpiston actuators, reflect this changing lever length.

The Selection ProcessIn choosing an actuator type, the fundamentalrequirement is to know your application. Controlsignal, operating mode, power source available,thrust/torque required, and fail-safe position canmake many decisions for you. Keep in mindsimplicity, maintainability and lifetime costs.

Safety is another consideration that must never beoverlooked. Enclosed linkages and controlledcompression springs available in some designsare important for safety reasons. Table 2-6 liststhe pros and cons of the various actuator styles.

2−10

Table 2-4. Typical Rotary Shaft Valve Torque Factors V-Notch Ball Valve with Composition Seal

Fisher TCM� Plus Ball SealValve Size,

NPS

Valve ShaftDiameter,

Inches

AB

C Maximum TD,Lbf�In.Composition Bearings(1) 60 Degrees 70 Degrees

11-1/2

23468

1/25/85/83/43/41

1-1/4

0.070.120.190.100.101.801.80

50100175280380500750

0.381.101.300.151.101.103.80

0.481.102.403.8018.036.060.0

515122512252120212041409820

101214161620

1-1/41-1/21-3/4

22-1/82-1/2

1.804.0042606097

125030002400280028005200

3.8011.075105105190

125143413578578

1044

982012,00023,52523,52555,76255,762

1. PEEK/PTFE or metal/PTFE bearings.

Table 2-5. Typical High Performance Butterfly Torque Factors for Valve with Composition SealPEEK/PTFE Bearings with PTFE Seal

Valve Size Shaft DiameterA B

C Maximum Allowable Torque

ANGLE OF OPENING S17400 H1075 S20910

NPS Inch 60� (�) 90� (�) lbf�in lbf�in

2 1/2 0.30 100 1.05 2.45 515 515

3 5/8 0.56 150 3.59 10.8 1087 1028

4 3/4 0.99 232 7.65 21.2 1640 1551

6 1 2.30 438 17.5 46.7 4140 4140

8 1-1/4 4.80 705 33.4 223 7988 7552

10 1-1/4 8.10 1056 82.2 358 9792 9258

12 1-1/2 12.5 1470 106 626 12000 12000

Table 2-6. Actuator Feature ComparisonActuator Type Advantages Disadvantages

Spring-and-Diaphragm Lowest costAbility to throttle without positionerSimplicityInherent fail-safe actionLow supply pressure requirementAdjustable to varying conditionsEase of maintenance

Limited output capabilityLarger size and weight

Pneumatic Piston High thrust capabilityCompactLightweightAdaptable to high ambient temperaturesFast stroking speedRelatively high actuator stiffness

Higher costFail-safe requires accessories or addition of a springPositioner required for throttlingHigh supply pressure requirement

Electric Motor CompactnessVery high stiffnessHigh output capability

High costLack of fail-safe actionLimited duty cycleSlow stroking speed

Electro-Hydraulic High output capabilityHigh actuator stiffnessExcellent throttling abilityFast stroking speed

High costComplexity and maintenance difficultyLarge size and weightFail-safe action only with accessories

2−11

Actuator Selection Summary� Actuator selection must be based upon a

balance of process requirements, valverequirements and cost.

� Simple designs such as the spring-and-diaphragm are simpler, less expensive and easierto maintain. Consider them first in most situations.

� Piston actuators offer many of theadvantages of pneumatic actuators with higherthrust capability than spring-and-diaphragm styles.They are especially useful where compactness isdesired or long travel is required.

� Electric and electro-hydraulic actuatorsprovide excellent performance. They are, however,much more complex and difficult to maintain.

� Actuator sizing is not difficult, but the widevariety of actuators and valves make it difficult tomaster. Vendor expertise is widely available.

� Systems such as control valves are bestpurchased, assembled and tested by one source.

Figure 2-9. The FIELDVUE Digital Valve Controllerbrings increased control accuracy and flexibility.When utilized with AMS ValveLink� software,

FIELDVUE instruments provide valuable diagnosticdata that helps to avoid maintenance problems.

W9915

Use of actuators and accessories of the samemanufacturer will eliminate many problems.

2−12

www.Fisher.com*Some information in this introductory material has been extracted from ANSI/ISAS75.01 standard with the permission of the publisher, the ISA.�All terms are defined in the nomenclature section.

Chapter 3

Liquid Valve Sizing

Valves are selected and sized to perform aspecific function within a process system. Failureto perform that given function in controlling aprocess variable results in higher process costs.Thus, valve sizing becomes a critical step tosuccessful process operation. The followingsections focus on correctly sizing valves for liquidservice: the liquid sizing equation is examined, thenomenclature and procedures are explained, andsample problems are solved to illustrate theiruse.2-

Valve Sizing Background

Standardization activities for control valve sizingcan be traced back to the early 1960s when atrade association, the Fluids Control Institute,published sizing equations for use with bothcompressible and incompressible fluids. Therange of service conditions that could beaccommodated accurately by these equations wasquite narrow, and the standard did not achieve ahigh degree of acceptance.

In 1967, the International Society of America(ISA�) established a committee to develop andpublish standard equations. The efforts of thiscommittee culminated in a valve sizing procedurethat has achieved the status of American NationalStandards Institute (ANSI). Later, a committee ofthe International Electrotechnical Commission(IEC) used the ISA works as a basis to formulateinternational standards for sizing control valves.*Except for some slight differences in nomenclatureand procedures, the ISA and IEC standards havebeen harmonized. ANSI/ISA Standard S75.01 isharmonized with IEC Standards 534-2-1 and534-2-2 (IEC Publications 534-2, Sections Oneand Two for incompressible and compressiblefluids, respectively).

Liquid Sizing Equation BackgroundThis section presents the technical substance ofthe liquid sizing equations. The value of this lies innot only a better understanding of the sizingequations, but also in knowledge of their intrinsiclimitations and relationship to other flow equationsand conditions.

The flow equations used for sizing have their rootsin the fundamental equations, which describe thebehavior of fluid motion. The two principleequations include the:

� Energy equation

� Continuity equation

The energy equation is equivalent to amathematical statement of the first law ofthermodynamics. It accounts for the energytransfer and content of the fluid. For anincompressible fluid (e.g. a liquid) in steady flow,this equation can be written as:

�V2

2gc� P

� � gZ�� w� q� U � constant (1)

The three terms� in parenthesis are allmechanical, or available, energy terms and carry aspecial significance. These quantities are allcapable of directly doing work. Under certainconditions more thoroughly described later, thisquantity may also remain constant:

V2

2gc� P

� � gZ � constant (2)

This equation can be derived from purelykinematic methods (as opposed to thermodynamicmethods) and is known as “Bernoulli’s equation”.

The other fundamental equation, which plays avital role in the sizing equation, is the continuity

3−2

Figure 3-1. Liquid Critical Pressure Ratio Factor for Water

equation. This is the mathematical statement ofconservation of the fluid mass. For steady flowconditions (one-dimensional) this equation iswritten as follows:

�VA � constant (3)

Using these fundamental equations, we canexamine the flow through a simple, fixedrestriction such as that shown in figure 3-1. Wewill assume the following for the present:

1. The fluid is incompressible (a liquid)

2. The flow is steady

3. The flow is one-dimensional

4. The flow can be treated as inviscid (having noviscosity)

5. No change of fluid phase occurs

As seen in figure 3-1, the flow stream mustcontract to pass through the reduced flow area.The point along the flow stream of minimum crosssectional flow area is the vena contracta. The flowprocesses upstream of this point and downstreamof this point differ substantially, thus it isconvenient to consider them separately.

The process from a point several pipe diametersupstream of the restriction to the vena contracta isvery nearly ideal for practical intents and purposes(thermodynamically isentropic, thus havingconstant entropy). Under this constraint,Bernoulli’s equation applies and we see that nomechanical energy is lost — it merely changesfrom one form to the other. Furthermore, changesin elevation are negligible since the flow streamcenterline changes very little, if at all. Thus,energy contained in the fluid simply changes frompressure to kinetic. This is quantified whenconsidering the continuity equation. As theflowstream passes through the restriction, thevelocity must increase inversely proportional to thechange in area. For example, from equation 4below:

VVC �(constant)

AVC

(4)

Using upstream conditions as a reference, thisbecomes:

VVC � V1� A1

AVC

� (5)

Thus, as the fluid passes through the restriction,the velocity increases. Below, equation 2 has been

3−3

applied and elevation changes have beenneglected (again using upstream conditions as areference):

�V1�2

2gc� P1 �

�VVC�2

2gc� PVC��(6)

In the equation below, equation 5 has beeninserted and rearranged:

PVC � P1�

�V1�2

2gc�� A1

AVC

�2 � 1� (7)

Thus, at the point of minimum cross sectionalarea, we see that fluid velocity is at a maximum(from equation 5 above) and fluid pressure is at aminimum (from equation 6 above).

The process from the vena contracta point to apoint several diameters downstream is not ideal,and equation 2 no longer applies. By argumentssimilar to the above, it can be reasoned (from thecontinuity equation) that, as the original crosssectional area is restored, the original velocity isalso restored. Because of the non-idealities of thisprocess, however, the total mechanical energy isnot restored. A portion of it is converted into heatthat is either absorbed by the fluid itself, ordissipated to the environment.

Let us consider equation 1 applied from severaldiameters upstream of the restriction to severaldiameters downstream of the restriction:

��U1 �V

1�2

2gc�

P1

� �gZ

1

gC� q �

U2�

V2�2

2gc�

P2

� �gZ

2

gC� w

(8)

No work is done across the restriction, thus thework term drops out. The elevation changes arenegligible and as a result, the respective termscancel each other. We can combine the thermalterms into a single term, HI:

�V1�2

2gc� P1 �

�V2�2

2gc� P2 �HI��(9)

The velocity was restored to its original value sothat equation 9 reduces to:

P1� P

2�HI��(10)

Consequently, the pressure decreases across therestriction, and the thermal terms (internal energyand heat lost to the surroundings) increase.

Losses of this type are generally proportional tothe square of the velocity (references one andtwo), so it is convenient to represent them by thefollowing equation:

HI � KI��V2

2��(11)

In this equation, the constant of proportionality, KI,is called the available head loss coefficient, and isdetermined by experiment.

From equations 10 and 11, it can be seen that thevelocity (at location two) is proportional to thesquare root of the pressure drop. Volume flow ratecan be determined knowing the velocity andcorresponding area at any given point so that:

Q � V2A2�� � 2� (P

1�� P

2)

�KI

 A2��(12)

Now, letting:

� � G��W

and, defining:

CV � A2�� �2�

�W�KI

��(13)

Where G is the liquid specific gravity, equation 12may be rewritten as:

Q � CV�� � �P

1�� P

2

G ��(14)

Equation 14 constitutes the basic sizing equationused by the control valve industry, and provides ameasure of flow in gallons per minute (GPM)when pressure in pounds per square inch (PSI) isused. At times, it may be desirable to work withother units of flow or independent flow variables(pressure, density, etc). The equationfundamentals are the same for such cases, andonly constants are different.

Determination of Flow CoefficientsRather than experimentally measure KI andcalculate Cv, it is more straightforward to measureCv directly.

3−4

Figure 3-2. Liquid Critical Pressure Ratio Factor for Liquids Other Than Water

A2738-1

In order to assure uniformity and accuracy, theprocedures for both measuring flow parametersand use in sizing are addressed by industrialstandards. The currently accepted standards aresponsored by the ISA.

The basic test system configuration is shown infigure 3-2. Specifications, accuracies, andtolerances are given for all hardware installationand data measurements such that coefficients canbe calculated to an accuracy of approximately 5%.Fresh water at approximately 68°F is circulatedthrough the test valve at specified pressuredifferentials and inlet pressures. Flow rate, fluidtemperature, inlet and differential pressure, valvetravel, and barometric pressure are all measuredand recorded. This yields sufficient information tocalculate the following sizing parameters:

� Flow coefficient (Cv)

� Pressure recovery coefficient (FL)

� Piping correction factor (Fp)

� Reynolds number factor (FR)

In general, each of these parameters depends onthe valve style and size, so multiple tests must beperformed accordingly. These values are thenpublished by the valve manufacturer for use insizing.

Basic Sizing Procedure OverviewThe procedure by which valves are sized fornormal, incompressible flow is straightforward.Again, to ensure uniformity and consistency, astandard exists that delineates the equations andcorrection factors to be employed for a givenapplication.

The simplest case of liquid flow applicationinvolves the basic equation developed earlier.Rearranging equation thirteen so that all of thefluid and process related variables are on the rightside of the equation, we arrive at an expression forthe valve Cv required for the particular application:

Cv ��Q

� � �P1��P

2

G ��(15)

It is important to realize that valve size is only oneaspect of selecting a valve for a given application.Other considerations include valve style and trimcharacteristic. Discussion of these features can bereferenced in chapter 2, chapter 4, and otherthorough resources.

Once a valve has been selected and Cv is known,the flow rate for a given pressure drop, or thepressure drop for a given flow rate, can bepredicted by substituting the appropriate quantitiesinto equation 16.

* The ability to recognize which terms are appropriate for a specific sizingprocedure can only be acquired through experience with different valve sizing problems. If any of the above terms appears to be new or unfamiliar, refer to the Abbreviations and Terminology Table (table 3-1) for a complete definition. 3−5

Many applications fall outside the bounds of thebasic liquid flow applications just considered.Rather than develop special flow equations for allof the possible deviations, it is possible (andpreferred) to account for different behavior withthe use of simple correction factors. Thesefactors, when incorporated, change the form ofequation 14 to the following:

Q � (N1�FP�FR)�CV�� � �P

1�� P

2

G ��(16)

All of the additional factors in this equation areexplained in the following sections.

Sizing Valves for LiquidsFollowing is a step-by-step procedure for thesizing of control valves for liquid flow using theIEC procedure. Each of these steps is importantand must be considered during any valve sizingprocedure. Steps three and four concern thedetermination of certain sizing factors that may, ormay not, be required in the sizing equationdepending upon the service conditions of thesizing problem. If one, two, or all three of thesesizing factors are to be included in the equation fora particular sizing problem, please refer to theappropriate factor determination section(s) locatedin the text proceeding step six.

1. Specify the variables required to size the valveas follows:

� Desired design

� Process fluid (water, oil, etc.)

� Appropriate service conditions Q or w, P1, P2or ΔP, T1, Gf, Pv, Pc, and υ*

2. Determine the equation constant, N.

N is a numerical constant contained in each of theflow equations to provide a means for usingdifferent systems of units. Values for these variousconstants and their applicable units are given inthe Equation Constants Table (table 3-2).

Use N1 if sizing the valve for a flow rate involumetric units (gpm or m3/h).

Use N6 if sizing the valve for a flow rate in massunits (lb/h or kg/h).

3. Determine Fp, the piping geometry factor.

Fp is a correction factor that accounts for pressurelosses due to piping fittings such as reducers,elbows, or tees that might be attached directly tothe inlet and outlet connections of the controlvalve to be sized. If such fittings are attached tothe valve, the Fp factor must be considered in thesizing procedure. If, however, no fittings areattached to the valve, Fp has a value of 1.0 andsimply drops out of the sizing equation.

For rotary valves with reducers (swagedinstallations), and other valve designs and fittingstyles, determine the Fp factors by using theprocedure for determining Fp, the piping geometryfactor.

4. Determine qmax (the maximum flow rate atgiven upstream conditions) or ΔPmax (theallowable sizing pressure drop).

The maximum or limiting flow rate (qmax),commonly called choked flow, is manifested by noadditional increase in flow rate with increasingpressure differential with fixed upstreamconditions. In liquids, choking occurs as a result ofvaporization of the liquid when the static pressurewithin the valve drops below the vapor pressure ofthe liquid.

The IEC standard requires the calculation of anallowable sizing pressure drop (ΔPmax) to accountfor the possibility of choked flow conditions withinthe valve. The calculated ΔPmax value iscompared with the actual pressure drop specifiedin the service conditions, and the lesser of thesetwo values is used in the sizing equation. If it isdesired to use ΔPmax to account for the possibilityof choked flow conditions it can be calculatedusing the procedure for determining qmax, themaximum flow rate, or ΔPmax, the allowable sizingpressure drop. If it can be recognized that chokedflow conditions will not develop within the valveΔPmax need not be calculated.

3−6

Table 3-1. Abbreviations and TerminologySymbol Symbol

Cv Valve sizing coefficient P1 Upstream absolute static pressure

d Nominal valve size P2 Downstream absolute staticpressure

D Internal diameter of the piping Pc Absolute thermodynamic criticalpressure

Fd Valve style modifier,dimensionless

Pv Vapor pressure absolute of liquid atinlet temperature

FF Liquid critical pressure ratio factor,dimensionless

ΔP Pressure drop (P1-P2) across thevalve

Fk Ratio of specific heats factor,dimensionless

ΔPmax(L) Maximum allowable liquid sizingpressure drop

FL Rated liquid pressure recoveryfactor, dimensionless

ΔPmax(LP) Maximum allowable sizing pressuredrop with attached fittings

FLP Combined liquid pressure recoveryfactor and piping geometry factorof valve with attached fittings(when there are no attachedfittings, FLP equals FL),dimensionless

q Volume rate of flow

FP Piping geometry factor,dimensionless

qmax Maximum flow rate (choked flowconditions) at given upstreamconditions

Gf Liquid specific gravity (ratio ofdensity of liquid at flowingtemperature to density of water at60�F), dimensionless

T1 Absolute upstream temperature(degree K or degree R)

Gg Gas specific gravity (ratio ofdensity of flowing gas to density ofair with both at standardconditions(1), i.e., ratio ofmolecular weight of gas tomolecular weight of air),dimensionless

w Mass rate of flow

k Ratio of specific heats,dimensionless

x Ratio of pressure drop to upstreamabsolute static pressure (ΔP/P1),dimensionless

K Head loss coefficient of a device,dimensionless

xT Rated pressure drop ratio factor,dimensionless

M Molecular weight, dimensionless Y Expansion factor (ratio of flowcoefficient for a gas to that for aliquid at the same Reynoldsnumber), dimensionless

N Numerical constant Z Compressibility factor,dimensionless

γ1 Specific weight at inlet conditions

υ Kinematic viscosity, centistokes

1. Standard conditions are defined as 60�F (15.5�C) and 14.7 psia (101.3kPa).

3−7

Table 3-2. Equation Constants(1)

N w q p(2) � T d, D

N1

0.08650.8651.00

- - -- - -- - -

m3/hm3/hgpm

kPabarpsia

- - -- - -- - -

- - -- - -- - -

- - -- - -- - -

N20.00214

890- - -- - -

- - -- - -

- - -- - -

- - -- - -

- - -- - -

mminch

N50.00241

1000- - -- - -

- - -- - -

- - -- - -

- - -- - -

- - -- - -

mminch

N6

2.7327.363.3

kg/hkg/hlb/h

- - -- - -- - -

kPabarpsia

kg/m3

kg/m3

lb/ft3

- - -- - -- - -

- - -- - -- - -

N7(3)

Normal ConditionsTN = 0�C

3.94394

- - -- - -

m3/hm3/h

kPabar

- - -- - -

deg Kdeg K

- - -- - -

Standard ConditionsTs = 15.5�C

4.17417

- - -- - -

m3/hm3/h

kPabar

- - -- - -

deg Kdeg K

- - -- - -

Standard ConditionsTs = 60�F

1360 - - - scfh psia - - - deg R - - -

N8

0.94894.819.3

kg/hkg/hlb/h

- - -- - -- - -

kPabarpsia

- - -- - -- - -

deg Kdeg Kdeg R

- - -- - -- - -

N9(3)

Normal ConditionsTN = 0�C

21.22120

- - -- - -

m3/hm3/h

kPabar

- - -- - -

deg Kdeg K

- - -- - -

Standard ConditionsTs = 15.5�C

22.42240

- - -- - -

m3/hm3/h

kPabar

- - -- - -

deg Kdeg K

- - -- - -

Standard ConditionsTS = 60�F

7320 - - - scfh psia - - - deg R - - -

1. Many of the equations used in these sizing procedures contain a numerical constant, N, along with a numericalsubscript. These numerical constants provide a means for using different units in the equations. Values for thevarious constants and the applicable units are given in the above table. For example, if the flow rate is given in U.S.gpm and the pressures are psia, N1 has a value of 1.00. If the flow rate is m3/hr and the pressures are kPa, the N1constant becomes 0.0865.2. All pressures are absolute.3. Pressure base is 101.3 kPa (1.013 bar)(14.7 psia).

5. Solve for required Cv, using the appropriateequation.

For volumetric flow rate units:

Cv ��q

N1�FP�

� � �P1��P

2

Gf

��(17)

For mass flow rate units:

Cv � w

N6�FP� � � � (P

1�� P

2)�� ��(18)

In addition to Cv, two other flow coefficients, Kvand Av, are used, particularly outside of NorthAmerica. The following relationships exist:

KV � (0.865)(CV)

AV � (2.40 10�5)(CV)

6. Select the valve size using the appropriate flowcoefficient table and the calculated Cv value.

Determining Piping Geometry Factor(Fp)Determine an Fp factor if any fittings such asreducers, elbows, or tees will be directly attachedto the inlet and outlet connections of the controlvalve that is to be sized. When possible, it isrecommended that Fp factors be determinedexperimentally by using the specified valve inactual tests.

Calculate the Fp factor using the followingequation:

Fp � �1� �KN

2

�Cv

d2�2�

�1�2

��(19)

where,

N2 = Numerical constant found in the EquationConstants table

d = Assumed nominal valve size

Cv = Valve sizing coefficient at 100% travel for theassumed valve size

3−8

In the above equation, the “K” term is thealgebraic sum of the velocity head losscoefficients of all of the fittings that are attached tothe control valve.

�K � K1� K

2� KB1 � KB2��(20)

where,

K1 = Resistance coefficient of upstream fittings

K2 = Resistance coefficient of downstream fittings

KB1 = Inlet Bernoulli coefficient

KB2 = Outlet Bernoulli coefficient

The Bernoulli coefficients, KB1 and KB2, are usedonly when the diameter of the piping approachingthe valve is different from the diameter of thepiping leaving the valve, whereby:

KB1�or�KB2 � 1� �dD�4��(21)

where,

d = Nominal valve size

D = Internal diameter of piping

If the inlet and outlet piping are of equal size, thenthe Bernoulli coefficients are also equal, KB1 =KB2, and therefore they are dropped from theequation.

The most commonly utilized fitting in control valveinstallations is the short-length concentric reducer.The equations for this fitting are as follows:

For an inlet reducer:

K1 � 0.5��1� d2

D2�2��(22)

For an outlet reducer:

K2 � 1.0��1� d2

D2�2��(23)

For a valve installed between identical reducers:

K1 � K2 � 1.5��1� d2

D2�2��(24)

Determining Maximum Flow Rate(qmax)Determine either qmax or ΔPmax if it is possible forchoked flow to develop within the control valvethat is to be sized. The values can be determinedby using the following procedures:

qmax � N1�FL�Cv�� � �P

1�� FF�Pv

Gf

��(25)

Values for FF, the liquid critical pressure ratiofactor, can be obtained from figure 3-3, or from thefollowing equation:

FF � 0.96� 0.28 �� � �Pv

Pc

��(26)

Values of FL, the recovery factor for rotary valvesinstalled without fittings attached, can be found inpublished coefficient tables. If the given valve is tobe installed with fittings such as reducer attachedto it, FL in the equation must be replaced by thequotient FLP/Fp, where:

FLP � �K1

N2

�Cv

d2�2 � 1

FL �2�

�1�2

��(27)

and

K1 = K1 + KB1

where,

K1 = Resistance coefficient of upstream fittings

KB1 = Inlet Bernoulli coefficient

Note: See the procedure for determining Fp, thepiping geometry factor, for definitions of the otherconstants and coefficients used in the aboveequations.)

Determining Allowable Sizing PressureDrop (�Pmax)ΔPmax (the allowable sizing pressure drop) can bedetermined from the following relationships:

For valves installed without fittings:

�Pmax(L) � FL �

2��P

1� FF�Pv

���(28)

3−9

Figure 3-3. Liquid Critical Pressure Ratio Factor for Water

For valves installed with fittings attached:

�Pmax(LP) � �FLP

FP

�2�P1� FF�PV

���(29)

where,

P1 = Upstream absolute static pressure

P2= Downstream absolute static pressure

Pv = Absolute vapor pressure at inlet temperature

Values of FF, the liquid critical pressure ratiofactor, can be obtained from figure 3-3 or from thefollowing equation:

FF � 0.96� 0.28Pv

Pc

��(30)

An explanation of how to calculate values of FLP,the recovery factor for valves installed with fittingsattached, is presented in the preceding proceduredetermining qmax (the maximum flow rate).

Once the ΔPmax value has been obtained from theappropriate equation, it should be compared withthe actual service pressure differential (ΔP = P1 −

P2). If ΔPmax is less than ΔP, this is an indicationthat choked flow conditions will exist under theservice conditions specified. If choked flowconditions do exist (ΔPmax < P1 − P2), then stepfive of the procedure for sizing valves for liquidsmust be modified by replacing the actual servicepressure differential (P1 − P2) in the appropriatevalve sizing equation with the calculated ΔPmaxvalue.

Note: Once it is known that choked flow conditionswill develop within the specified valve design(ΔPmax is calculated to be less than ΔP), a furtherdistinction can be made to determine whether thechoked flow is caused by cavitation or flashing.The choked flow conditions are caused by flashingif the outlet pressure of the given valve is less thanthe vapor pressure of the flowing liquid. Thechoked flow conditions are caused by cavitation ifthe outlet pressure of the valve is greater than thevapor pressure of the flowing liquid.

Liquid Sizing Sample ProblemAssume an installation that, at initial plant start-up,will not be operating at maximum designcapability. The lines are sized for the ultimatesystem capacity, but there is a desire to install acontrol valve now which is sized only for currently

3−10

anticipated requirements. The line size is 8-inches,and an ANSI Class 300 globe valve with an equalpercentage cage has been specified. Standardconcentric reducers will be used to install the valveinto the line. Determine the appropriate valve size.

1. Specify the necessary variables required tosize the valve:

� Desired valve design is an ANSI Class 300globe valve with equal percentage cage and anassumed valve NPS 3.

� Process fluid is liquid propane

� Service conditions are q = 800 gpm

P1 = 300 psig = 314.7 psia

P2 = 275 psig = 289.7 psia

ΔP = 25 psi

T1 = 70°F

Gf = 0.50

Pv = 124.3 psia

Pc = 616.3 psia

2. Use an N1 value of 1.0 from the EquationConstants table.

3. Determine Fp, the piping geometry factor.

Because it is proposed to install a NPS 3 valve inan 8-inch line, it will be necessary to determine thepiping geometry factor, Fp, which corrects forlosses caused by fittings attached to the valve.

From Equation 19,

Fp � �1� �KN

2

�Cv

d2�2�

�1�2

where,

N2 = 890, from the Equation Constants Table

d = 3 inches, from step one

Cv = 121, from the flow coefficient table for anANSI Class 300, NPS 3 globe valve with equalpercentage cage.

To compute ΣK for a valve installed betweenidentical concentric reducers:

�K � K1� K

2

� 1.5��1� d2

D2�2

� 1.5��1� (3)2

(8)2�2

� 1.11

where,

D = 8 inches, the internal diameter of the pipingso,

Fp � �1� 1.11

890�12132�2��1�2

� 0.90

4. Determine ΔPmax (the allowable sizingpressure drop)

Based upon the small required pressure drop, theflow will not be choked (ΔPmax > ΔP).

5. Solve for Cv, using equation 17.

Cv �q

N1FP

P1�P

2

Gf

� 800

(1.0)(0.90)25

0.5

� 125.7

6. Select the valve size using the flow coefficienttable and the calculated Cv value.

The required Cv of 125.7 exceeds the capacity ofthe assumed valve, which has a Cv of 121.

Although, for this example, it may be obvious thatthe next larger size (NPS 4) would be the correctvalve size, this may not always be true, and arepeat of the above procedure should be carriedout. This is assuming that a NPS 4 valve, Cv =203. This value was determined from the flowcoefficient table for an ANSI Class 300, NPS 4globe valve with an equal percentage cage.

Recalculate the required Cv using an assumed Cvvalue of 203 in the Fp calculation.

3−11

where,

�K � K1� K

2

� 1.5�1� d2

D2�2

� 1.5�1� 16

64�2

� 0.84

and

Fp � �1� �KN

2

�Cv

d2�2�

�1�2

� �1� 0.84

890�20342�2��1�2

� 0.93

and

Cv �q

N1Fp

P1�P

2

Gf

� 800

(1.0)(0.93)25

0.5

� 121.7

This solution indicates only that the NPS 4 valve islarge enough to satisfy the service conditionsgiven. There may be cases, however, where amore accurate prediction of the Cv is required. Insuch cases, the required Cv should be determinedagain using a new Fp value based on the Cv valueobtained above. In this example, Cv is 121.7,which leads to the following result:

Fp � �1� �KN

2

�Cv

d2�2�

�1�2

� �1� 0.84

890�121.7

42�2��1�2

� 0.97

The required Cv then becomes:

Cv �q

N1Fp

P1�P

2

Gf

� 800

(1.0)(0.97)25

0.5

� 116.2

Because this newly determined Cv is close to theCv used initially for this recalculation (116.2 versus121.7), the valve sizing procedure is complete,and the conclusion is that a NPS 4 valve openedto about 75% of total travel should be adequate forthe required specifications.

Sizing for Pulp StockThe behavior of flowing pulp stock is different fromwater or viscous Newtonian fluids. It is necessaryto account for this behavior when determining therequired valve size. Methods have beendeveloped to aid in determining correct valve sizefor these types of applications.

The pulp stock sizing calculation uses thefollowing modified form of the basic liquid sizingequation (equation thirteen, above):

Q � Cv�Kp �P

where,

ΔP = sizing pressure drop, psid

Cv = valve flow coefficient

Kp = pulp stock correction factor

Q = volumetric flow rate, gpm

The root of this calculation is the pulp stockcorrection factor, Kp. This factor is the ratio of thepulp stock flow rate to water flow rate under thesame flowing conditions. It, therefore, modifies therelationship between Q, Cv, and ΔP to account forthe effects of the pulp stock relative to that forwater. The value of this parameter, in theory,depends on many factors such as pulp stock type,consistency, freeness, fiber length, valve type andpressure drop. However, in practice, it appearsthat the dominant effects are due to three primaryfactors: pulp type, consistency and pressuredifferential. Values of Kp for three different pulpstock types are shown in figure 3-4 through 3-6.

3−12

Once the value of the pulp stock correction factoris known, determining the required flow coefficientor flow rate is equivalent to basic liquid sizing. Forexample, consider the following:

Q = 1000 gpm of 8% consistency Kraft pulp stock

ΔP = 16 psid

P1 = 150 psia

Kp = 0.83 from figure 3-5

therefore, Cv �Q

Kp �P � 1000

0.83 16 � 301

Figure 3-4. Pulp Stock Correction Factors for Kraft Pulp

E1377

3−13

Figure 3-5. Pulp Stock Correction Factors for Mechanical Pulp

E1378

Figure 3-6. Pulp Stock Correction Factors for Recycled Pulp

E1379

3−14

www.Fisher.com

Chapter 4

Cavitation and Flashing

Severe Liquid Flow SizingProper control valve sizing is important tosuccessful plant operation. However, sizing is notalways straightforward. At times, it involvesconsidering phenomena beyond that of generalservice. Selecting the appropriate control valvecan be extremely critical to the complete processloop. Liquid sizing for severe flow service,including events involving cavitation or flashing,must be closely examined in order to obtainsuccessful plant operation.

Sizing for severe flow service applications can beexplained by expanding upon base liquid sizingknowledge. The following sections will build uponthe basic liquid sizing equations presented inchapter 3 in order to study liquid fluid behaviorsinvolved with choked flow, cavitation, flashing,viscous flow, and sizing for pulp stock. In addition,discussion of considerations in selecting theappropriate control valves for cavitating andflashing services will take place.

Choked FlowThe equation illustrated below (chapter 3, equation14) would imply that, for a given valve, flow couldbe continually increased to infinity by simplyincreasing the pressure differential across thevalve.

Q � Cv

P1� P

2

G�

(31)

In reality, the relationship given by this equationholds for only a limited range. As the pressuredifferential is increased a point is reached wherethe realized mass flow increase is less thanexpected. This phenomenon continues until noadditional mass flow increase occurs in spite of

Figure 4-1. Typical Flow Curve Showing RelationshipBetween Flow Rate Q and Imposed Pressure

Differential �P

A3442 / IL

increasing the pressure differential (figure 4-1).This condition of limited maximum mass flow isknown as choked flow. To understand more aboutwhat is occurring, and how to correct it whensizing valves, it is necessary to revisit some of thefluid flow basics discussed in chapter 3.

Recall that, as a liquid passes through a reducedcross-sectional area, velocity increases to amaximum and pressure decreases to a minimum.As the flow exits, velocity is restored to its originalvalue while the pressure is only partially restoredthus creating a pressure differential across thedevice. As this pressure differential is increased,the velocity through the restriction increases(increasing flow) and the vena contracta pressuredecreases. If a sufficiently large pressuredifferential is imposed upon the device, theminimum pressure may decrease to, or below, thevapor pressure of the liquid under theseconditions. When this occurs the liquid becomesthermodynamically unstable and partiallyvaporizes. The fluid now consists of a liquid andvapor mixture that is no longer incompressible.

4−2

While the exact mechanisms of liquid choking arenot fully confirmed, there are parallels betweenthis and critical flow in gas applications. In gasflows, the flow becomes critical (choked) when thefluid velocity is equal to the acoustic wave speedat that point in the fluid. Pure incompressible fluidshave high wave speeds and, practically speaking,they do not choke. Liquid-to-gas or liquid-to-vapormixtures, however, typically have low acousticwave speeds (actually lower than that for a puregas or vapor), so it is possible for the mixturevelocity to equal the sonic velocity and choke theflow.

Another way of viewing this phenomenon is toconsider the density of the mixture at the venacontracta. As the pressure decreases, so does thedensity of the vapor phase, hence, the density ofthe mixture decreases. Eventually, this decreasein density of the fluid offsets any increase in thevelocity of the mixture to the point where noadditional mass flow is realized.

It is necessary to account for the occurrence ofchoked flow during the sizing process so thatundersizing of a valve does not occur. In otherwords, knowing the maximum flow rate a valvecan handle under a given set of conditions isnecessary. To this end, a procedure wasdeveloped which combines the control valvepressure recovery characteristics with thethermodynamic properties of the fluid to predictthe maximum usable pressure differential, i.e. thepressure differential at which the flow chokes.

A pressure recovery coefficient can be defined as:

Km �P

1� P

2

P1 � Pvc

(32)

Under choked flow conditions, it is establishedthat:

Pvc � rc�Pv

(33)

The vapor pressure, Pv, is determined at inlettemperature because the temperature of the liquiddoes not change appreciably between the inletand the vena contracta. The term “rc” is known asthe critical pressure ratio, and is anotherthermodynamic property of the fluid. While it isactually a function of each fluid and the prevailingconditions, it has been established that data for avariety of fluids can be generalized, according tofigure 4-2 or the following equation, withoutsignificantly compromising overall accuracy:

Figure 4-2. Generalized rc Curve

A3443 / IL

rc � FF � 0.96� 0.28Pvc

Pc

�(34)

The value of Km is determined individually by testfor each valve style and accounts for the pressurerecovery characteristics of the valve.

By rearranging equation sixteen, the pressuredifferential at which the flow chokes can bedetermined is known as the allowable pressuredifferential:

(P1� P

2)allowable � Km(P1

� rc�Pv)

(35)

When this allowable pressure differential is used inthe equation below (equation 14 from chapter 3),the choked flow rate for the given valve will result.

Q � Cv

P1� P

2

G�

If this flow rate is less than the required serviceflow rate, the valve is undersized. It is thennecessary to select a larger valve, and repeat thecalculations using the new values for Cv and Km.

The equations supplied in the sizing standard are,in essence, the same as those presented in thischapter, except the nomenclature has beenchanged. In this case:

Qmax � N1FLCv

P1� FF�Pv

G�

(36)

4−3

Figure 4-3. Typical Cavitation Damage

W1350

where:

FL = Km�

FF = rc

N1 = units factor

CavitationClosely associated with the phenomenon ofchoked flow is the occurrence of cavitation. Simplystated, cavitation is the formation and collapse ofcavities in the flowing liquid. It is of specialconcern when sizing control valves because if leftunchecked, it can produce unwanted noise,vibration, and material damage.

As discussed earlier, vapor can form in the vicinityof the vena contracta when the local pressure fallsbelow the vapor pressure of the liquid. If the outletpressure of the mixture is greater than the vaporpressure as it exits the valve, the vapor phase willbe thermodynamically unstable and will revert to aliquid. The entire liquid-to-vapor-to-liquid phasechange process is known as “cavitation,” althoughit is the vapor-to-liquid phase change that is theprimary source of the damage. During this phasechange a mechanical attack occurs on thematerial surface in the form of high velocitymicro-jets and shock waves. Given sufficientintensity, proximity, and time, this attack canremove material to the point where the valve nolonger retains its functional or structural integrity.figure 4-3 shows an example of such damage.

Cavitation and the damage it causes are complexprocesses and accurate prediction of key eventssuch as damage, noise, and vibration level isdifficult. Consequently, sizing valves for cavitationconditions requires special considerations.

Figure 4-4. Comparison of High and Low Recovery Valves

A3444

The concept of pressure recovery plays a key rolein characterizing a valve’s suitability for cavitationservice. A valve that recovers a significantpercentage of the pressure differential from theinlet to the vena contracta is appropriately termeda high recovery valve. Conversely, if only a smallpercent is recovered, it is classified as a lowrecovery valve. These two are contrasted in figure4-4. If identical pressure differentials are imposedupon a high recovery valve and a low recoveryvalve, all other things being equal, the highrecovery valve will have a relatively low venacontracta pressure. Thus, under the sameconditions, the high recovery valve is more likelyto cavitate. On the other hand, if flow througheach valve is such that the inlet and venacontracta pressures are equal, the low recoveryvalve will have the lower collapse potential(P2−Pvc), and cavitation intensity will generally beless.

Therefore, it is apparent that the lower pressurerecovery devices are more suited for cavitationservice.

The possibility of cavitation occurring in any liquidflow application should be investigated bychecking for the following two conditions:

1. The service pressure differential isapproximately equal to the allowable pressuredifferential.

2. The outlet pressure is greater than the vaporpressure of the fluid.

4−4

Figure 4-5. Pressure Profiles for Flashing and Cavitating Flows

A3445

If both of these conditions are met, the possibilityexists that cavitation will occur. Because of thepotentially damaging nature of cavitation, sizing avalve in this region is not recommended. Specialpurpose trims and products to control cavitationshould be considered. Because of the greatdiversity in the design of this equipment, it is notpossible to offer general guidelines for sizingvalves with those specialized trims. Please refer tospecific product literature for additionalinformation.

Cavitation in Pulp StockCavitation behavior in low consistency pulp stock(less than 4%) is treated as equivalent to that ofwater. Generally, pulp stock consistency greaterthan 4% is not known to be problematic, as thestock itself absorbs the majority of the energyproduced by the cavitating microjets.

FlashingFlashing shares some common features withchoked flow and cavitation in that the processbegins with vaporization of the liquid in the vicinityof the vena contracta. However, in flashingapplications, the pressure downstream of thispoint never recovers to a value that exceeds thevapor pressure of the fluid. Thus, the fluid remainsin the vapor phase. Schematic pressure profilesfor flashing and cavitating flow are contrasted infigure 4-5.

Flashing is of concern not only because of itsability to limit flow through the valve, but alsobecause of the highly erosive nature of theliquid-vapor mixture. Typical flashing damage issmooth and polished in appearance (figure 4-6) in

Figure 4-6. Typical Flashing Damage

W2842

stark contrast to the rough, cinder-like appearanceof cavitation (figure 4-3).

If P2 < Pv, or there are other service conditions toindicate flashing, the standard sizing procedureshould be augmented with a check for chokedflow. Furthermore, suitability of the particular valvestyle for flashing service should be establishedwith the valve manufacturer. Selection guidelineswill be discussed later in the chapter.

Viscous FlowOne of the assumptions implicit in the sizingprocedures presented to this point is that of fullydeveloped, turbulent flow. Turbulent flow andlaminar flow are flow regimes that characterize thebehavior of flow. In laminar flow, all fluid particlesmove parallel to one another in an orderly fashionand with no mixing of the fluid. Conversely,turbulent flow is highly random in terms of localvelocity direction and magnitude. While there iscertainly net flow in a particular direction,instantaneous velocity components in all directionsare superimposed on this net flow. Significant fluidmixing occurs in turbulent flow. As is true of manyphysical phenomena, there is no distinct line ofdemarcation between these two regimes. Thus, athird regime of transition flow is sometimesrecognized.

The physical quantities which govern this flowregime are the viscous and inertial forces; thisratio is known as the Reynolds number. When theviscous forces dominate (a Reynolds numberbelow 2,000) the flow is laminar, or viscous. If theinertial forces dominate (a Reynolds numberabove 3,000) the flow is turbulent, or inviscid.

Consideration of these flow regimes is criticalbecause the macroscopic behavior of the flowchanges when the flow regime changes. Theprimary behavior characteristic of concern in sizingis the nature of the available energy losses. Inearlier discussion it was asserted that, under theassumption of inviscid flow, the available energy

4−5

Figure 4-7. Viscous Flow Correction Factors

A3446

losses were proportional to the square of thevelocity.

In the laminar flow regime, these same losses arelinearly proportional to the velocity; in thetransitional regime, these losses tend to vary.Thus, for equivalent flow rates, the pressuredifferential through a conduit or across arestriction will be different for each flow regime.

To compensate for this effect (the change inresistance to flow) in sizing valves, a correctionfactor was developed. The required Cv can bedetermined from the following equation:

Cvreq�d

� FR�Cvrated

(37)

The factor FR is a function of the Reynoldsnumber and can be determined from a simplenomograph procedure, or by calculating theReynolds number for a control valve from thefollowing equation and determining FR from figure4-7.

Rev �N4FdQ

�FL1�2�Cv1�2� 1

N2

(FL)2�Cv

d22 1�

1�4

(38)

To predict flow rate, or resulting pressuredifferential, the required flow coefficient is used inplace of the rated flow coefficient in theappropriate equation.

When a valve is installed in a field pipingconfiguration which is different than the specifiedtest section, it is necessary to account for theeffect of the altered piping on flow through thevalve. (Recall that the standard test section

consists of a prescribed length of straight pipe upand downstream of the valve.) Field installationmay require elbows, reducers, and tees, which willinduce additional losses immediately adjacent tothe valve. To correct for this situation, two factorsare introduced:

� Fp

� Flp

Factor Fp is used to correct the flow equationwhen used in the incompressible range, whilefactor Flp is used in the choked flow range. Theexpressions for these factors are:

Fp � ��KN2

�Cv

d22 1�

�1�2

(39)

FIp � FL�FL�2KI

N2

�Cv

d22 1�

�1�2

(40)

The term K in equation 39 is the sum of all losscoefficients of all devices attached to the valveand the inlet and outlet Bernoulli coefficients.Bernoulli coefficients account for changes in thekinetic energy as a result of a cross-sectional flowarea change. They are calculated from thefollowing equations.

KBinlet

� 1� (d�D)4

(41a)

KBoutlet

� (d�D)4 � 1

(41b)

Thus, if reducers of identical size are used at theinlet and outlet, these terms cancel out.

The term “KI” in equation 40 includes the losscoefficients and Bernoulli coefficient on the inletside only.

In the absence of test data or knowledge of losscoefficients, loss coefficients may be estimatedfrom information contained in other resources.

The factors Fp and FI would appear in flowequations 31 and 36 respectively as follows:

For incompressible flow:

Q � FpCv

P1� P

2

G�

(42)

4−6

Figure 4-8. The implosion of cavitation vapor cavities is rapid, asymmetric and very energetic. The mechanics of collapse give rise to high velocity liquid jets, which impinge on metallic surfaces. Ultimately, the metal fatigues and

breaks away in small pieces.

E0111

For choked flow:

Qmax � FICv

P1� FF�Pv

G�

(43)

Valve Material DamageCavitation damage is usually the mosttroublesome side effect plaguing the control valveindustry. It does not take many examples of suchdamage to fully demonstrate the destructivecapabilities of cavitation.

Typically, cavitation damage is characterized byan irregular, rough surface. The phrase“cinder-like appearance” is used frequently todescribe cavitation damage. It is discernible fromother types of flow damage such as erosion andflashing damage which are usually smooth andshiny in appearance. This next section will dealwith cavitation damage, although most of thecomments can also apply to flashing damage. Acomparison of figures 4-3 and 4-6 illustrates thesedifferences.

While the results of cavitation damage are all toofamiliar, the events and mechanisms of thecavitation damage process are not known orunderstood completely in spite of extensive studyover the years. There is general agreement,however, on a number of aspects of the processand consistency in certain observations.

Cavitation damage has been observed to beassociated with the collapse stage of the bubbledynamics. Furthermore, this damage consists oftwo primary events or phases:

1. An attack on a material surface as a result ofcavitation in the liquid.

2. The response or reaction of the material to theattack.

Any factor that influences either of these eventswill have some sort of final effect on the overalldamage characteristics.

The attack stage of the damage process has beenattributed to various mechanisms, but none ofthem account for all the observed results. Itappears that this attack involves two factors thatinteract in a reinforcing manner:

1. Mechanical attack

2. Chemical attack

There is evidence indicating the almost universalpresence of a mechanical attack component whichcan occur in either of two forms:

1. Erosion resulting from high-velocity microjetsimpinging upon the material surface.

2. Material deformation and failure resulting fromshock waves impinging upon the material surface.

In the first type of mechanical attack a small,high-velocity liquid jet is formed during theasymmetrical collapse of a vapor bubble. Iforientation and proximity of the jets is proper, adamaging attack occurs on the metal surface asshown in figure 4-8. This is the most probableform of mechanical attack. High-speedcinematography, liquid drop impingementcomparisons, and various analytical studiessupport its presence.

The second type of mechanical attack, shockwave impingement, does not appear to be as

4−7

dominant. Analytical estimations of vapor bubblecollapse pressures do not suggest that the shockwaves are on a damaging order of magnitude —at least during the initial collapse. Experimentalstudies bear this out. They also reveal thatresulting collapse pressures increase in magnitudewith subsequent rebound collapses and becomepotentially damaging.

The other primary component of attack, chemicalattack, is perhaps more significant because itinteracts with the mechanical component ratherthan acting by itself. After a period of mechanicalattack, many of the protective coatings of amaterial (films, oxides, etc.) are physicallyremoved, making the base material morevulnerable to chemical attack.

Just as a number of variables have an affect onthe behavior of individual cavities, a number ofvariables influence the degree and extent ofmaterial damage. The principal variables thatinfluence cavitation damage include air content,pressure, velocity and temperature.

Air content impacts cavitation damage primarilythrough its effect on cavity mechanics. Again, twoopposing trends are evident on increasing theamount of air. Adding air supplies more entrainedair nuclei which, in turn, produce more cavitiesthat can increase the total damage. After a point,however, continued increases in air contentdisrupt the mechanical attack component andeffectively reduce the total damage.

Pressure effects also exhibit two opposing trends.Given a fixed inlet pressure P1, decreasing thebackpressure P2 tends to increase the number ofcavities formed, which creates a worse situation.However, a lower backpressure also creates alower collapse pressure differential (P2 − Pv),resulting in a decrease in the intensity of thecavitation.

An additional pressure effect, unrelated to theabove, concerns the location of damage. As thebackpressure is changed, the pressure required tocollapse the cavities moves upstream ordownstream depending upon whether thepressure is increased or decreased, respectively.In addition to a change in the severity of the totaldamage, there may be an accompanying changein the physical location of the damage whenpressure conditions are altered.

It should now be apparent that the cavitation andflashing damage process is a complex function of:

1. Intensity and degree of cavitation (cavitationattack)

2. Material of construction (material response)

3. Time of exposure

While the above-mentioned influences have beenobserved, they remain to be quantified. Often,experience is the best teacher when it comes totrying to quantify cavitation damage.

NoiseAlthough the noise associated with a cavitatingliquid can be quite high, it is usually of secondaryconcern when compared to the material damagethat can exist. Therefore, high intensity cavitationshould be prevented to decrease the chance ofmaterial damage. If cavitation is prevented, thenoise associated with the liquid flow will be lessthan 90 dBA.

For a flashing liquid, studies and experience haveshown that the noise level associated with thevalve will be less than 85 dBA, regardless of thepressure drop involved to create the flashing.

Cavitation / Flashing DamageCoefficients and Product SelectionCavitation in control valves can be an applicationchallenge. It is important to understand theguidelines when selecting an appropriate valveand trim. Experience, knowledge of wherecavitation problems begin, and the effect of valvesize and type, are all useful in deciding whichvalve and trim can be used.

TerminologyFL: Pressure recovery coefficient. A valveparameter used to predict choked flow.

ΔPmax: Allowable sizing pressure drop. Thelimiting pressure drop beyond which any increasein pressure drop brought about by decreasing P2will not generate additional flow through the valve.Therefore the valve is “choked”. Per equation 28of chapter 3:

�Pmax(L) � FL

2

(P1� FF�Pv)

where,

P1 = Upstream absolute static pressure

Pv = Absolute vapor pressure at inlet temperature

4−8

FF = the liquid critical pressure ratio factor. Can beobtained from the following equation:

FF � 0.96� 0.28Pv

Pc

�Kc: Cavitation coefficient. A valve parameterdependent upon valve style and trim. It predictsthe beginning of cavitation related damage andvibration problems for a particular valve/trim style.

�PCavitation � Kc�(P1� Pv)

Ar: Application ratio. A cavitation index which isdependent upon the actual service conditions. Itindicates the presence of flashing or potentiallycavitating services.

Ar � (�PFlow)���(P1� Pv)

Ki: Incipient cavitation coefficient. A valveparameter which predicts the point of initialgeneration and collapse of vapor bubbles.(Specific values of Ki are generally not available).

�PIncipient�Cav. � Ki�(P1� Pv)

Valve Selection Coefficient Criteria andSelection Procedure1. Determine ΔPFlowing (ΔPFlow)

2. Calculate Ar

a. If Ar ≥ 1.0, the service is flashing.

b. If Ar ≤ 1.0, the service is potentiallycavitating.

3. Use ΔPFlow and Ar in conjunction with the Kcvalues of valve trim ΔP limits and Kc indices, aswell as other valve selection criteria (P1, temp.,style, etc.), to make a valve selection.

The cavitation coefficient (Kc) is based upon valvetype and pressure drop limit. Select a valve/trimthat has a ΔP limit higher than the service ΔPFlowand a Kc higher that the service Ar.

Application GuidelinesGuidelines (including Ar and Kc ratios) weredeveloped to help select the proper valveconstruction when cavitation is present. Theseguidelines are intended to provide valve selectionsfree of cavitation related material and vibrationdamage over the long term. The guidelines do not

indicate an absence of cavitation. Thus, noise dueto cavitation may still be present. If noise is aconcern, use hydrodynamic noise prediction toassist in selecting a valve.

The following restrictions apply to theseguidelines:

� Water only

� Customer requirements that may require useof different guidelines

Examples:

� Long maintenance intervals

� Very low noise requirements

� Different fluids

� Corrosive an/or erosive environment

� Installation limitations

� Valve usage rate

These guidelines will aid in selecting a valve andtrim designed to help prevent cavitation damageand thus offer long term valve life in potentiallycavitating services.

For detailed cavitating service valve selectionguidelines, please contact your local sales office.

Additional Guidelines� For all valve styles and sizes, applying

backpressure to the valve can eliminate cavitation.This solution is most effective when the serviceconditions do not vary widely.

� Fluids information:

— Cold water is the most common problemfluid.

— Pure component fluids, similar to water,can also cause problems.

— Fluid mixtures, like that of pulp stock, canbe less damaging even when the numbersindicate cavitation is present. Experience ismost useful here.

These guidelines have been constructed from abroad base of experience. There are undoubtedlyexceptions to these guidelines and, as always,

4−9

recent experience should be used to select thebest valve for specific applications.

Hardware Choices for FlashingApplicationsIt was stated previously that flashing is a liquidflow phenomenon that is defined by the system,and not by the valve design. Therefore, sinceflashing cannot be prevented by the control valve,all that can be done is to prevent flashing damage.There are three main factors that affect theamount of flashing damage in a control valve:

1. Valve design

2. Materials of construction

3. System design

Valve DesignWhile valve design has no bearing upon whetherflashing does or does not occur, it can have alarge impact on the intensity of flashing damage.Generally, there are two valve designs that aremore resistant to flashing damage.

An angle valve with standard trim in the flow downdirection and with a downstream liner is perhapsthe best solution to preventing flashing damage.figure 4-9 shows a typical angle valve for flashingservice.

This construction is an excellent choice becauseflashing damage occurs when high velocity vaporbubbles impinge on the surface of a valve. Anangle valve reduces the impingement by directingflow into the center of the downstream pipe, notinto the valve body. If damage does occur, thedownstream liner can be replaced much moreeconomically than the valve body.

A rotary plug style of valve is also an excellentchoice for medium to low pressure flashingapplications. This valve can be installed with theplug facing the downstream side of the body(figure 4-10) so when flashing occurs, it does sodownstream of the valve. In some cases, a spoolpiece of sacrificial pipe is used to absorb theflashing damage.

Materials of ConstructionThere are several factors that determine theperformance of a given material in a particularflashing and/or cavitating situation including thematerials’ toughness, hardness, and its corrosion

Figure 4-9. Fisher EAS valve with outlet liner isused for flashing service. The liner resists

erosion and protects the body.

W0970

RESTRICTED-TRIM ADAPTOR

LINER

resistance in the application environment. Within agiven material family (e.g. the 400-series stainlesssteels), hardness is a fairly accurate method forranking materials. However, when comparingmaterials from different families, hardness doesnot correlate with overall resistance to damage.For example, cobalt-chromium-tungsten basedalloy 6 has much more resistance to cavitationand flashing than either hardened type 410 or 17-4stainless steels, even though they all exhibitroughly the same hardness. In fact, alloy 6 equalsor exceeds the performance of many materialswith a hardness of 60 HRC and higher. Thesuperior performance of alloy 6 is attributed to abuilt-in “energy-absorbing” mechanism shared bya number of cobalt-base alloys.

Materials commonly used for flashing andcavitating services are alloy 6 (solid and overlays),nickel-chromium-boron alloys (solid and overlays),hardened 440C stainless steel, hardened 17-4stainless steel, and hardened 410/416 stainlesssteel.

Because the standard materials used in valvebodies are relatively soft, selection for cavitation andflashing resistance must rely upon factors other thanhardness. In general, as the chromium andmolybdenum contents increase, the resistance todamage by both cavitation and flashing increase.Thus, the chromium-molybdenum alloy steels have

4−10

Figure 4-10. Rotary plug valves, such as the V500 Vee-Ball valve(reverse flow trim direction, trim level 3) have excellenterosion resistance and perform well in flashing service

W8359

better resistance than the carbon steels, and thestainless steels have even better resistance than thechromium-molybdenum alloy steels.

In the past, ASME SA217 grade C5 was the mostcommonly specified chromium-molybdenum alloysteel. However, because of the poor casting,welding, and manufacturing characteristics of C5,ASME SA217 grade WC9 has become a morepopular alternative. Experience indicates that WC9performs on par with C5 in cavitation and flashingservices despite its lower chromium content(2-1/4% vs. 5%). This is apparently because itshigher molybdenum content (1% vs. 1/2%) makesup for the lower chromium content.

ASTM A217 grade C12A is becoming morecommon in the power industry. This material hasexcellent high temperature properties, and istypically used at temperatures exceeding 1000°F(538°C). Its higher chromium and molybdenumcontents (9% Cr, 1% Mo) would indicate excellentcavitation resistance.

While angle bodies are a better choice for flashingapplications than globe bodies, they are also amore economical choice in most cases. This isbecause carbon steel bodies can be used in anangle valve with an optional hardened downstreamliner (17-4PH SST or alloy 6) because only thedownstream portion of the valve will experiencethe flashing liquid (see figure 4-9). If a globe valveis used, it is better to use achromium-molybdenum alloy steel body becausethe flashing will occur within the body itself.

Figure 4-11. Location of a control valve can oftenbe changed to lengthen its life or allow use of lessexpensive products. Mounting a heater drain valve

near the condenser is a good example.

E0864

System DesignThis section discusses system design where it isassumed flashing will occur. The optimum positionof the valve in a flashing service can have a greatimpact on the success of that valve installation.

Figure 4-11 shows the same application with theexception of the location of the control valve.These figures are fairly representative of a valvethat controls flow to a condenser. In the topillustration, the flashing will occur in thedownstream pipe between the control valve andthe tank. Any damage that occurs will do so in thatdownstream piping area. In the bottom illustration,the flashing will occur downstream of the valvewithin the tank.

Because the tank has a much larger volumecompared to the pipe, high velocity fluid

4−11

impingement on a material surface will not occuras there is essentially no material surface. Thissystem design will help prevent flashing damage.

Hardware Choices for CavitatingApplicationsThe design of a control valve greatly affects theability of a valve to control cavitation. This sectiondiscusses the theories behind each type of trimdesign that is used for cavitation control and alsoreviews each type of Fisher trim used to controlcavitation.

The design theories or ideas behind the varioustrim designs include:

� Tortuous path

� Pressure drop staging

� Expanding flow area

� Drilled hole design

� Characterized cage

� Separation of seating and throttling locations

� Cavitation control in lieu of prevention

Tortuous PathProviding a tortuous path for the fluid through thetrim is one way to lower the amount of pressurerecovery of that trim. Although this tortuous pathcan be in the form of drilled holes, axial flowpassages or radial flow passages, the effect ofeach design is essentially the same. The use of atortuous path design concept is used in virtuallyevery cavitation control style of hardware.

Pressure Drop StagingThis approach to damage control routes flowthrough several restrictions in series, as opposedto a single restriction. Each restriction dissipates acertain amount of available energy and presents alower inlet pressure to the next stage.

A well-designed pressure-staging device will beable to take a large pressure differential whilemaintaining the vena contracta pressure above thevapor pressure of the liquid, which prevents theliquid from cavitating.

Figure 4-12. In Cavitrol trim, the pressure drop isstaged in two or more unequal steps. Staging isaccomplished by increasing the flow area from

stage to stage. This stepped reduction allows fullpressure drop without the vena contracta pressure

falling below the vapor pressure of the liquid.

A2149-1

For the same pressure differential then, the venacontracta pressure in conventional trim will belower than for the staged trim, and the liquid willbe more prone to cavitate.

Trims that dissipate available energy have anadditional advantage. If the design pressuredifferential is exceeded and cavitation does occur,the intensity will be less. This is because thepressure that causes the collapse of cavities (i.e.,the recovered pressure) will be less.

Expanding Flow AreasThe expanding flow area concept of damagecontrol is closely related to the pressure dropstaging concept. Figure 4-12 shows a pressureversus distance curve for flow through a series offixed restrictions where the area of eachsucceeding restriction is larger than the previous.Notice that the first restriction takes the bulk of thepressure drop, and the pressure drop throughsuccessive sections decreases.

4−12

Figure 4-13. By combining the geometric effects ofthick plates and thin plates, it is possible to design

a flow passage that optimizes capacity andrecovery coefficient values. These carefullydesigned passages are used exclusively in

Cavitrol cages.

E0113−1

In the last restriction, where cavitation is mostlikely to occur, the pressure drop is only a smallpercentage of the total drop, and the pressurerecovery is substantially lowered.

The expanding flow area concept requires fewerpressure drop stages to provide the samecavitation protection as the equal area concept.Because the pressure drop of the last stage israther low compared to the total pressure drop, ifcavitation does occur, the intensity and cavitationdamage will be much less.

Drilled Hole DesignDrilled hole cages are used in the Fisher Cavitrol�cavitation control trim line to provide a tortuouspath, pressure drop staging, and expanding flowarea. The design of each particular drilled hole hasa significant impact on the overall pressurerecovery of the valve design.

Figure 4-13 illustrates a cross section of threetypes of drilled holes that could be used in acavitation control cage. The thin plate design is aninefficient flow device, but it does provide a highFL

2 and, therefore, a low pressure recovery. Thethick plate design provides an efficient design, butalso provides a high pressure recovery as denotedby a low FL

2 value.

The Cavitrol trim hole design is a balance betweenthe thick plate and the thin plate hole designs. It

provides relatively high flow efficiency whilemaintaining a high FL

2, which results in a lowpressure recovery. This design represents theoptimal choice between capacity and cavitationcontrol.

Another benefit of this type of drilled hole design isthat the vena contracta point is further from theexit of the hole when compared to a straightthrough drilled hole. Consequently, if pressurerecovery above the vapor pressure occurs(cavitation), it will do so further away from theexternal wall of the cage, and the amount ofdamage will be smaller.

One disadvantage of cavitation control trims is thepotential for flow passages to become pluggedwith sand, dirt or other debris. Particulate ladenflow is common to water injection applications.The flowing media often times contains smallparticulate that can plug the passages, restrictingor totally stopping flow through the valve. If thispotential exists, the particles must be removedfrom the flow stream, usually by filtration or analternative approach to cavitation should be taken.

An alternative is to use a trim that is designed toallow the particulate to pass, but still controlcavitation. The Fisher Dirty Service Trim (DST)has been designed to allow particles up to 3/4” tobe passed and to control cavitation up to pressuredrops of 4000 psi. This trim has been usedextensively in produced water injection, waterinjection pump recirculation, and other liquid flow,particulate containing, high pressure dropapplications.

Characterized CageThe characterized cage design theory has evolvedfrom the fact that “capacity is inversely related to adesign’s ability to prevent cavitation.” In thoseapplications where the pressure drop decreasesas the flow rate increases, characterized cagescan be used to optimize cavitation prevention andcapacity.

For a Cavitrol III trim design, as the travelincreases, the cage design changes. It begins asa pressure-staging design and then develops intoa straight-through hole design. Consequently, thecavitation control ability of this trim design isgreatest at low travels and decreases withincreasing valve plug travel.

Care should be taken to employ characterizedcages only in applications where the pressuredrop decreases as travel increases.

4−13

Figure 4-14. Cavitrol IV trim provides cavitationprotection at pressures to 6500 psi. It uses expanding flow areas to affect a four-stage

pressure drop. All significant pressure drop is taken downstream of the shutoff seating surface.

W3668−1

Separate Seating and ThrottlingLocations

In a modern power plant, most cavitatingapplications require a control valve to not onlyprovide cavitation control, but also provide tightshutoff. The best way to accomplish this is toseparate the throttling location from the seatinglocation as shown in figure 4-14. The seatingsurface of the plug is upstream of the throttlinglocation, and the upper cage is designed such thatit takes very little pressure drop. The seatingsurface experiences relatively low flow velocitiesas velocity is inversely related to pressure. Arecent technological advancement has been toimplement the use of a softer seating materialrelative to the material of the plug. This allows fora slight deformation of the seating material, whichprovides much better plug/seat contact and, as aresult, greatly enhanced shutoff capability. Valves

utilizing this soft seating material are capable ofproviding Class VI shutoff.

Cavitation Control HardwareAlternativesIn the previous sections, theories behind moderntypes of cavitation control hardware werediscussed. This section presents alternatives tothe, sometimes, costly cavitation hardware.Guidelines are also presented to help determinewhen cavitation control hardware is required orwhen other alternatives can be employed.

System DesignCorrect liquid system design is the mosteconomical way to prevent the damaging effectscaused by cavitation without applying cavitationprevention control valves. Unfortunately, even thebest system design is likely to need cavitation typecontrol valves, but by applying certain designfeatures, the complexity of these control valvesmay be simplified.

The most common and oldest method ofdesigning a liquid flow system where largepressure drops must occur is to use a standardtrim control valve with a downstreambackpressure device. Although these devicescome in various sizes, shapes, and designs, theyall perform the same function of lowering thepressure drop across the control valve by raisingits downstream pressure.

Because the downstream pressure of the valve isincreased, the vena contracta pressure isincreased. If the backpressure device is sizedcorrectly, the vena contract pressure will not fallbelow the vapor pressure, and cavitation will notoccur.

While this is a simple and cost-effective way toprevent cavitation damage in the control valve,there are several serious considerations to look atbefore using a downstream backpressure device.

� A larger valve may be required to pass therequired flow as the pressure drop is lowered.

� Although cavitation may not occur at thecontrol valve, it may occur at the backpressuredevice.

� The backpressure device can only be sizedfor one condition. If other conditions exist, the

4−14

backpressure provided may allow cavitation tooccur.

� If the backpressure device becomes worn,the backpressure will decrease and cavitation inthe valve may occur.

Another disadvantage that is rarely mentionedoccurs when a valve is opened against a highupstream pressure. Until the flow reaches thebackpressure device and stabilizes, the valve willexperience the entire pressure drop of the system.Although this may only occur for a short period oftime, the potential for damage exists.

In the instance of rotary valves, air injection(known as aspiration) also can be used tominimize the effects of cavitation in a system. Withthis method, air is injected upstream of the venacontracta. The dispersed air acts as a buffer whenthe vapor bubbles implode so that the intensity ofthe cavitation is lowered. Unfortunately, thelocation of the vena contracta, the amount of air tobe injected, etc. are hard to quantify.

Because air is being injected into the system, thismethod of cavitation control is usually used onlarge valves dumping to a tank or pond or wheresolids in the system prevent the use of othercavitation control devices.

Cavitation is an interesting but destructivephenomenon. Preventing cavitation is the mostacceptable way of limiting potential for damage.Proper application of available products, basedupon sizing equations and field experience, willprovide long term success.

SummaryThe past two chapters have indicated that afundamental relationship exists between keyvariables (P1, P2, Pv, G, Cv, Q) for flow through adevice such as a control valve. Knowledge of anyfour of these allows the fifth to be calculated orpredicted. Furthermore, adjustments to this basicrelationship are necessary to account for specialconsiderations such as installed pipingconfiguration, cavitation, flashing, choked flow,and viscous flow behavior. Adherence to theseguidelines will ensure correct sizing and optimumperformance.

It is important to understand that pulp stock flowexhibits characterizations that closely resemblethose of water. Guidelines for hindering the effectsof cavitation are based upon process testing usingwater. One must consider that a pulp stockmulti-phase flow may result in less severe damagewhen compared to that of water for flashing,cavitation, or turbulent flow. However, it must benoted that pulp stock can lead to other issuessuch as erosion and corrosion, depending onprocess make-up and the materials used in theprocess. Therefore, it is important to understandthe process media, as well as firm processconditions, in order to ensure the correct valve isproperly sized and selected for the given severeservice application.

As noted throughout the chapter, it is evident thatsevere flow phenomena through a control valvecan occur under the proper conditions. In general,the most common liquid severe serviceapplications involve either cavitation or flashing. Itis important to have a basic understanding of bothliquid service incidents as presented in thischapter.

www.Fisher.com

Chapter 5

Gas Sizing

This chapter addresses the six-step procedure forsizing control valves for compressible flow usingthe standardized ISA procedure. All six steps areoutlined below, and must be accounted for whensizing a valve for compressible flow. Steps threeand four are involved in determining specific sizingfactors that may or may not be required in thesizing equation depending on the serviceconditions of the application. When steps threeand/or four are required, refer to the appropriatesection of the book referenced below.

Standardized ISA Procedure1. Specify the necessary variables required tosize the valve as follows:

� Desired valve design (globe, butterfly, ball)

� Process fluid (air, natural gas, steam, etc.)

� Appropriate service conditions (q, or w, P1,P2 or �P, T1, Gg, M, k, Z, and �1)

The ability to recognize the appropriate terms for aspecific valve sizing application is gained throughexperience sizing valves for different applications.Refer to the notations table in chapter three forany new or unfamiliar terms.

2. Determine the equation constant, N.

N is a numerical constant contained in each of theflow equations to provide a means for usingdifferent systems of units. Values for these variousconstants and their applicable units are given inthe equation constants table 5-2 at the end of thischapter.

Use N7 or N9 when sizing a valve with a specifiedflow rate in volumetric units (scfh or m3/h).Selecting the appropriate constant depends uponthe specified service conditions. N7 is used only

when specific gravity, Gg, has been specifiedalong with the other required service conditions.N9 is used only when the molecular weight, M, ofthe gas has been specified.

Use N6 or N8 when sizing a valve with a specifiedflow rate in mass units (lb/h or kg/h). In this case,N6 is used only when specific weight, �1, has beenspecified along with the other required serviceconditions. N8 is used only when the molecularweight, M, of the gas has been specified.

3. Determine Fp, the piping geometry factor.

Fp is a correction factor that accounts for anypressure losses due to piping fittings such asreducers, elbows, or tees that might be attacheddirectly to the inlet and outlet connections of thecontrol valve. If such fittings are attached to thevalve, the Fp factor must be considered in thesizing procedure. If no fittings are attached to thevalve, Fp has a value of one and drops out of thesizing equation.

For rotary valves with reducers, other valvedesigns and fitting styles refer to the determiningpiping geometry section of chapter three todetermine the appropriate Fp value.

4. Determine Y, the expansion factor.

Y � 1� x3Fk�xT

where,

Fk = k/1.4, the ratio of specific heats factor

k = Ratio of specific heats

x = �P/P1

xT = The pressure drop ratio factor for valvesinstalled without attached fittings. Moredefinitively, xT is the pressure drop ratio required

5−2

to produce critical, or maximum, flow through thevalve when Fk = 1.0

When the control valve to be installed has fittings,such as reducers or elbows attached to it, theireffect is accounted for in the expansion factorequation by replacing the xT term with a newfactor xTP. A procedure for determining the xTPfactor is described in the following section:Determining xTP, the Pressure Drop Ratio Factor.

Note: Conditions of critical pressuredrop are realized when the value of xbecomes equal to or exceeds theappropriate value of the product ofeither Fk*xT or Fk*xTP at whichpoint::

y � 1� x3Fk�xT

� 1� 1�3 � 0.667

In actual service, pressure drop ratios can, andoften will exceed the indicated critical values. Atthis point, critical flow conditions develop. Thus,for a constant P1, decreasing P2 (i.e., increasing�P) will not result in an increase in the flow ratethrough the valve. Therefore, the values of xgreater than the product of either Fk*xT or Fk*xTPmust never be substituted in the expression for Y.This means that Y can never be less than 0.667.This same limit on values of x also applies to theflow equations introduced in the next section.

5. Solve for the required CV using the appropriateequation.

For volumetric flow rate units —

� when specific gravity, Gg, of the gas has been specified:

Cv �q

N7�Fp�P1�Y x

Gg�T1�Z

�� when molecular weight, M, of the gas has

been specified:

Cv �q

N9�Fp�P1

�Y x

M�T1�Z

�For mass flow rate units —

� when specific weight, �1, of the gas has been specified:

Cv � w

N6FpY x�P

1��

1�

� when molecular weight, M, of the gas has been specified:

Cv � M

N8�Fp�P1

�Yx�M

T1�Z

�6. Select the valve size using the appropriate flowcoefficient table using the calculated CV value.

Determining xTP, the Pressure DropRatio FactorWhen the control valve is to be installed withattached fittings such as reducers or elbows, theiraffect is accounted for in the expansion factorequation by replacing the xT term with a newfactor, xTP.

xTP �xTFp �

2�1� xT�Ki

N5

�Cv

d� 22

�1

where,

N5 = numerical constant found in the equationconstants table

d = assumed nominal valve size

CV = valve sizing coefficient from flowcoefficient table at 100% travel for the assumedvalve size

Fp = piping geometry factor

xT = pressure drop ratio for valves installedwithout fittings attached. xT values are includedin the flow coefficient tables.

In the above equation, Ki is the inlet head losscoefficient, which is defined as:

Ki � K1� KB1

where,

K1 = resistance coefficient of upstream fittings(see the procedure: Determining Fp, the PipingGeometry Factor, which is contained in Chapter3: Liquid Valve Sizing

KB1 = Inlet Bernoulli coefficient (see theprocedure: Determining Fp, the PipingGeometry Factor, which is contained in chapterthree: Liquid Valve Sizing

Compressible Fluid Sizing SampleProblem No. 1Assume steam is to be supplied to a processdesigned to operate at 250 psig. The supply

5−3

source is a header maintained at 500 psig and500�F. A 6-inch line from the steam main to theprocess is being planned. Also, make theassumption that if the required valve size is lessthan 6 inches, it will be installed using concentricreducers. Determine the appropriate ED valve witha linear cage.

1. Specify the necessary variables required tosize the valve.

� Desired valve design—ANSI Class 300 EDvalve with a linear cage. Assume valve size is 4inches.

� Process fluid—superheated steam

� Service conditions—

w = 125,000 lb/h

P1 = 500 psig = 514.7 psia

P2 = 250 psig = 264.7 psia

�P = 250 psi

x = �P/P1 = 250/514.7 = 0.49

T1 = 500�F

�1 = 1.0434 lb/ft3 (from properties ofsaturated steam table)

k= 1.28 (from properties of saturated steamtable)

2. Determine the appropriate equation constant,N, from the equation constants table 3-2 inchapter three.

Because the specified flow rate is in mass units,(lb/h), and the specific weight of the steam is alsospecified, the only sizing equation that can beused is that which contains the N6 constant.

Therefore, N6 = 63.3

3. Determine Fp, the piping geometry factor.

Fp � �1� �KN

2

�Cv

d22

�1�2

where,

N2 = 890, determined from the equationconstants table

d = 4 in.

Cv = 236, which is the value listed in the flowcoefficient table 4-2 for a NPS 4 ED valve at100% total travel.

and

�K � K1� K

2

� 1.5�1� d2

D22

� 1.5�1� 42

622

� 0.463

Finally:

Fp ��� 1� 0.463

890�(1.0)(236)

(4)22

���

�1�2

� 0.95

4. Determine Y, the expansion factor.

Y � 1� x3Fk�xTP

where,

Fk �k

1.40

� 1.28

1.40

� 0.91

x � 0.49�(As�calculated�in�step�1.)

Because the size 4 valve is to be installed in a6-inch line, the xT term must be replaced by xTP.

xTP �xTFp�

2�1� xT�Ki

N5

�Cv

d22

�1

where,

N5 = 1000, from the equation constants table

d = 4 inches

Fp = 0.95, determined in step three

xT = 0.688, a value determined from theappropriate listing in the flow coefficienttable

5−4

Cv = 236, from step three

and

Ki � K1� KB1

� 0.5�1� d2

D22 � �1� �d

D4

� 0.5�1� 42

622 � �1� �4

64

� 0.96

where D = 6 inches

so:

XTP � 0.69

0.952�1� (0.69)(0.96)

1000�236422�1

� 0.67

Finally:

Y � 1� x3�Fk�xTP

� 1� 0.49

(3)�(0.91)�(0.67)� 0.73

5−5

Table 5-1. Abbreviations and TerminologySymbol Symbol

Cv Valve sizing coefficient P1 Upstream absolute static pressure

d Nominal valve size P2 Downstream absolute staticpressure

D Internal diameter of the piping Pc Absolute thermodynamic criticalpressure

Fd Valve style modifier,dimensionless

Pv Vapor pressure absolute of liquid atinlet temperature

FF Liquid critical pressure ratio factor,dimensionless

ΔP Pressure drop (P1-P2) across thevalve

Fk Ratio of specific heats factor,dimensionless

ΔPmax(L) Maximum allowable liquid sizingpressure drop

FL Rated liquid pressure recoveryfactor, dimensionless

ΔPmax(LP) Maximum allowable sizing pressuredrop with attached fittings

FLP Combined liquid pressure recoveryfactor and piping geometry factorof valve with attached fittings(when there are no attachedfittings, FLP equals FL),dimensionless

q Volume rate of flow

FP Piping geometry factor,dimensionless

qmax Maximum flow rate (choked flowconditions) at given upstreamconditions

Gf Liquid specific gravity (ratio ofdensity of liquid at flowingtemperature to density of water at60�F), dimensionless

T1 Absolute upstream temperature(degree K or degree R)

Gg Gas specific gravity (ratio ofdensity of flowing gas to density ofair with both at standardconditions(1), i.e., ratio ofmolecular weight of gas tomolecular weight of air),dimensionless

w Mass rate of flow

k Ratio of specific heats,dimensionless

x Ratio of pressure drop to upstreamabsolute static pressure (ΔP/P1),dimensionless

K Head loss coefficient of a device,dimensionless

xT Rated pressure drop ratio factor,dimensionless

M Molecular weight, dimensionless Y Expansion factor (ratio of flowcoefficient for a gas to that for aliquid at the same Reynoldsnumber), dimensionless

N Numerical constant Z Compressibility factor,dimensionless

γ1 Specific weight at inlet conditions

υ Kinematic viscosity, centistokes

1. Standard conditions are defined as 60�F (15.5�C) and 14.7 psia (101.3kPa).

5−6

Table 5-2. Equation Constants(1)

N w q p(2) � T d, D

N1

0.08650.8651.00

- - -- - -- - -

m3/hm3/hgpm

kPabarpsia

- - -- - -- - -

- - -- - -- - -

- - -- - -- - -

N20.00214

890- - -- - -

- - -- - -

- - -- - -

- - -- - -

- - -- - -

mminch

N50.00241

1000- - -- - -

- - -- - -

- - -- - -

- - -- - -

- - -- - -

mminch

N6

2.7327.363.3

kg/hkg/hlb/h

- - -- - -- - -

kPabarpsia

kg/m3

kg/m3

lb/ft3

- - -- - -- - -

- - -- - -- - -

N7(3)

Normal ConditionsTN = 0�C

3.94394

- - -- - -

m3/hm3/h

kPabar

- - -- - -

deg Kdeg K

- - -- - -

Standard ConditionsTs = 15.5�C

4.17417

- - -- - -

m3/hm3/h

kPabar

- - -- - -

deg Kdeg K

- - -- - -

Standard ConditionsTs = 60�F

1360 - - - scfh psia - - - deg R - - -

N8

0.94894.819.3

kg/hkg/hlb/h

- - -- - -- - -

kPabarpsia

- - -- - -- - -

deg Kdeg Kdeg R

- - -- - -- - -

N9(3)

Normal ConditionsTN = 0�C

21.22120

- - -- - -

m3/hm3/h

kPabar

- - -- - -

deg Kdeg K

- - -- - -

Standard ConditionsTs = 15.5�C

22.42240

- - -- - -

m3/hm3/h

kPabar

- - -- - -

deg Kdeg K

- - -- - -

Standard ConditionsTS = 60�F

7320 - - - scfh psia - - - deg R - - -

1. Many of the equations used in these sizing procedures contain a numerical constant, N, along with a numericalsubscript. These numerical constants provide a means for using different units in the equations. Values for thevarious constants and the applicable units are given in the above table. For example, if the flow rate is given in U.S.gpm and the pressures are psia, N1 has a value of 1.00. If the flow rate is m3/hr and the pressures are kPa, the N1constant becomes 0.0865.2. All pressures are absolute.3. Pressure base is 101.3 kPa (1.013 bar)(14.7 psia).

5−7

Table 5-3. Flow Coefficient Table

Gas or Liquid Flow Modified EqualPercentage Characteristic

ValveSize,

Inches

MinimumThrottling

CV(1)

Coefficients

Valve Rotation, Degrees

10 20 30 40 50 60 70 80 90

8 86.7

Cv 47.3 126 236 382 604 972 1600 3000 4960

KV 40.9 109 204 330 522 841 1380 2600 4290

FL 0.79 0.87 0.91 0.91 0.85 0.81 0.73 0.63 0.63

Fd 0.37 0.64 0.78 0.88 0.94 0.97 0.98 0.99 1.00

XT 0.44 0.64 0.77 0.77 0.67 0.51 0.38 0.20 0.13

10 136

Cv 74.1 197 369 598 946 1520 2510 4700 7770

KV 64.1 171 320 517 818 1320 2170 4060 6720

FL 0.79 0.87 0.91 0.91 0.85 0.81 0.73 0.63 0.63

Fd 0.37 0.64 0.78 0.87 0.94 0.97 0.99 0.99 1.00

XT 0.44 0.64 0.77 0.77 0.67 0.51 0.38 0.20 0.13

12 196

Cv 107 284 532 861 1360 2190 3610 6760 11 200

KV 92.2 246 460 745 1180 1890 3120 5850 9670

FL 0.79 0.87 0.91 0.91 0.85 0.81 0.73 0.63 0.63

Fd 0.39 0.67 0.79 0.87 0.93 0.97 0.99 1.00 1.00

XT 0.44 0.64 0.77 0.77 0.67 0.51 0.38 0.20 0.13

16 347

Cv 189 505 945 1530 2420 3890 6410 12 000 19 900

KV 164 437 818 1320 2090 3370 5540 10 400 17 200

Fd 0.38 0.64 0.79 0.87 0.93 0.97 0.99 0.99 1.00

FL 0.79 0.87 0.91 0.91 0.85 0.81 0.73 0.63 0.63

XT 0.44 0.64 0.77 0.77 0.67 0.51 0.38 0.20 0.13

20 542

Cv 296 788 1480 2390 3780 6080 10 000 18 800 31 000

KV 256 681 1280 2070 3270 5260 8660 16 200 26 800

Fd 0.42 0.66 0.79 0.87 0.93 0.97 0.99 1.00 1.00

FL 0.79 0.87 0.91 0.91 0.85 0.81 0.73 0.63 0.63

XT 0.44 0.63 0.76 0.76 0.66 0.50 0.38 0.20 0.131. Valves should not be required to throttle at a Cv less than the minimum throttling Cv.

5−8

Table 5-4. Representative Sizing Coefficients for ED Single-Ported Globe Style Valve Bodies

Valve Size(inches)

Valve Plug Style Flow Characteristic Port Dia.(in.)

RatedTravel(in.)

CV FL XT FD

1/2 Post Guided Equal Percentage 0.38 0.50 2.41 0.90 0.54 0.61

3/4 Post Guided Equal Percentage 0.56 0.50 5.92 0.84 0.61 0.61

1

Micro Form

Cage Guided

Equal Percentage

LinearEqual Percentage

3/81/23/4

1 5/161 5/16

3/43/43/43/43/4

3.074.918.84

20.617.2

0.890.930.970.840.88

0.660.800.920.640.67

0.720.670.620.340.38

1 1/2

Micro-Form

Cage Guided

Equal Percentage

LinearEqual Percentage

3/81/23/4

1 7/81 7/8

3/43/43/43/43/4

3.205.18

10.239.235.8

0.840.910.920.820.84

0.650.710.800.660.68

0.720.670.620.340.38

2 Cage Guided LinearEqual Percentage

2 5/162 5/16

1 1/81 1/8

72.959.7

0.770.85

0.640.69

0.330.31

3 Cage Guided LinearEqual Percentage

3 7/16 1 1/2 148136

0.820.82

0.620.68

0.300.32

4 Cage Guided LinearEqual Percentage

4 3/8 2 236224

0.820.82

0.690.72

0.280.28

6 Cage Guided LinearEqual Percentage

7 2 433394

0.840.85

0.740.78

0.280.26

8 Cage Guided LinearEqual Percentage

8 3 846818

0.870.86

0.810.81

0.310.26

www.Fisher.com

Chapter 6

Control Valve Noise

Noise has always been present in control valves.It is a natural side effect of the turbulence andenergy absorption inherent in control valves. Thischapter will address how noise is created, why itcan be a problem, and methods to attenuate noisecreated in control valves.

The major problem with industrial noise is its affecton humans. Companies usually build town borderstations on sites remote from residentialdevelopments. Isolation, however, is not alwayspossible, and noise prevention is a must.

The U.S. Occupational Safety and Health Act(OSHA) establishes maximum permissible noiselevels for all industries whose business affectsinterstate commerce. These standards relateallowable noise levels to the permissible exposuretime. Notice in table 6-1 that the maximumpermissible levels depend upon the duration ofexposure. For example, the maximum sound levela person should be exposed to for an eight hourday is 90 dBA. These maximum sound levels havebecome the accepted noise exposure standard formost regulatory agencies. Thus, they havebecome the standard by which much noisegenerating equipment has been specified andmeasured.

Table 6-1. Maximum Permissible Noise LevelsDuration of Exposure

(Hours)Maximum Sound Pressure

(dBA)

16 85

8 90

4 95

2 100

1 105

1/2 110

1/4 115

Decibels (dB) are a measure to give an indicationof loudness. The “A” added to the term indicatesthe correction accounting for the response of thehuman ear. The sensitivity of our ears to soundvaries at different frequencies. Applying this “A”correction is called weighting, and the correctednoise level is given in dBA.

The A-weighting factor at any frequency isdetermined by how loud noise sounds to thehuman ear at that particular frequency comparedto the apparent loudness of sound at 1000 hertz.At 1000 hertz the A-weighting factor is zero, so ifthe sound pressure level is 105 dB, we say itsounds like 105 dB.

On the other hand, if we listen to a sound at 200hertz with a sound pressure level of 115 dB, itsounds more like 105 dB. Therefore, we say thatthe A-weighted loudness of the noise with a soundpressure level of 115 dB is 105 dBA.

Essentially, if two or more sounds with differentsound pressure levels and frequencies sound likethe same loudness, they have the same dBA,regardless of what their individual, unweightedsound pressure levels may be.

The effect of A-weighting on control valve noisedepends upon the flowing medium since eachdevelops its own characteristic spectrum. Noiselevels for hydrodynamic noise, or liquid flow noise,have appreciable energy at frequencies below 600hertz. When the levels are A-weighted, it makesthe low frequency terms more meaningful and thegovernment standards somewhat more difficult tomeet.

On the other hand, aerodynamic noise levelsproduced by steam or gas flow are the same ineither dB or dBA. This is because aerodynamicnoise occurs primarily in the 1000 to 8000 hertzfrequency range. The human ear has a fairly flatresponse in the frequency range of 600 to 10,000

6−2

hertz, and the A-weighting factor is essentiallyzero in this range. Thus, there is negligibledifference between the dB and dBA ratings.

Sound CharacteristicsAnalyzing noise, in the context of piping andcontrol valves, requires consideration of its origin.This indicates how the noise will propagate.Generally speaking, noise originates from either aline source or a point source.

A sound level meter is used to determine soundpressure levels. Readings for line source noiselevels are normally measured one meter from thepipe’s surface and at a point one meterdownstream from the valve outlet. Measurementsshould be made in an unobstructed free field areawith no sound reflecting surfaces nearby.

Line source noise levels are radiated from thepiping in the form of an imaginary cylinder, thepipe centerline as the axis. As you move awayfrom the pipeline, the sound pressure leveldecreases inversely to the changes in surfacearea of the imaginary cylinder. The followingequation defines the sound pressure level (LpA) atdistances other than one meter from the pipelinesurface:

LpA � F� 10 log1� rR� r

where,LpA = sound pressure levelF = noise level at one meter from the pipe surfacer = pipe radius in meters based on the pipe outside diameter

R = distance in meters from the pipe surface

What this equation tells us is that the soundpressure level decreases dramatically as thedistance from the pipeline increases. Keep in mindthat this equation determines the noise levelradiated only by the pipeline. Other noise sourcescould combine with the pipeline noise source toproduce greater overall sound pressure level.

The other type of noise source needed to bediscussed is point source. Point source noisemeasurements are taken at a three meter distancein the horizontal plane through the source. Ventapplications are typical examples of point sourcenoise. Point source noise levels are radiated in theform of an imaginary sphere with the source at thecenter of the sphere. As you move away from apoint source, the sound pressure level decreasesinversely in proportion to the changes in thesurface area of the imaginary sphere. Theequation that defines the sound pressure level atdistances other than three meters from the pointsource and below a horizontal plane through thepoint source is:

LpA � F� 20 log3

R

where,LpA = the sound pressure level

F = the noise level at three meters from the source

R = the distance in meters from the source

This procedure determines the noise level radiatedonly by the point source. Other noise sourcescould combine with the point source noise toproduce a greater overall sound pressure level.

Combining Noise SourcesThe noise level in a certain area is the result ofcombining all of the noise generated by everynoise source in the vicinity. The methodology ofcombining sources is important to prediction andactually lies at the root of noise abatementtechnology.

To determine the resultant noise level of two noisesources, it is necessary to combine two sources ofenergy. The energy, or power, of two sourcescombines directly by addition. The power levelsmust be calculated separately and thenlogarithmically combined as one overall noisesource. The sources can be line, point, or acombination of both. Table 6-2 simplifies theprocess of combining two known noise levels.

6−3

Table 6-2. Combined Noise CorrectionsdBA1 - dBA2 dBA Adder to Loudest Noise Source

0 3.01

1 2.54

2 2.12

3 1.76

4 1.46

5 1.2

6 <1To use table 6-2:1. Determine the noise level of each source at the point where you want todetermine the combined noise level.2. Determine the arithmetic dB difference between the two sources at thelocation of interest.3. Find the difference between the two sources in the table.4. Read across the table to find the dB factor to be used. Add this factor to thelouder of the two sources. This value is the combined dB of the two sources.

Let’s put this table to work to illustrate how noisesources combine. Two interesting examples helpillustrate how sound levels combine:

1. When two noise sources with equal soundpressure levels of 90 dB are combined, thecorrection factor is 3.01. Therefore, the resultantcombined noise level is 93 dB.

2. If two sources have considerably differentnoise levels, say 95 dB and 65 dB, the correctionfactor is nearly zero. Therefore, the combinednoise level is essentially the same as the louder ofthe two sources, that is, 95 dB. This leads us tothe first rule of noise control: Preventing orcontrolling the loudest noise sources first.

While this appears obvious, in practice it is not theeasiest path.

Sources of Valve NoiseControl valves have long been recognized as acontributor to excessive noise levels in many fluidprocess and transmission systems. The majorsources of control valve noise are mechanicalvibration noise, aerodynamic noise, andhydrodynamic noise.

Mechanical noise generally results from valve plugvibration. Vibration of valve components is a resultof random pressure fluctuations within the valvebody and/or fluid impingement upon the movableor flexible parts. The most prevalent source ofnoise resulting from mechanical vibration is thelateral movement of the valve plug relative to theguiding surfaces. The sound produced by this typeof vibration normally has a frequency less than1500 hertz and is often described as a metallicrattling. In these situations, the physical damage

incurred by the valve plug and associated guidingsurfaces is generally of more concern than thenoise emitted.

Another source of mechanical vibration noise isresonant vibration, which occurs when a valvecomponent resonates at its natural frequency.Resonant vibration produces a single-pitched tonenormally having a frequency between 3000 and7000 hertz. This type of vibration produces highlevels of mechanical stress that may producefatigue failure of the vibrating part. Valvecomponents susceptible to natural frequencyvibration include contoured valve plugs with hollowskirts and flexible seals.

The noise caused by the vibration of valvecomponents is usually of secondary concern, and,ironically, may even be beneficial because it giveswarning when conditions exist that could producevalve failure. Noise resulting from mechanicalvibration has for the most part been eliminated byimproved valve design. Most modern controlvalves employ cage guiding and more precisebearings to eliminate vibration problems. Testinghelps isolate and eliminate resonant frequencyproblems before installation.

The second type of noise is hydrodynamic noise.Hydrodynamic noise results from liquid flow and iscaused by the implosion of vapor bubbles formedin the cavitation process. Vapor bubble formationoccurs in valves controlling liquids when theservice conditions are such that the local staticpressure, at some point within the valve, is lessthan or equal to the liquid vapor pressure.Localized areas of low static pressures within thevalve are a result of the pressure-to-velocity-headinterchange that occurs at the valve venacontracta. When the vapor bubbles movedownstream, they encounter pressures higherthan the vapor pressure and collapse. The rapidimplosion can result in severe damage to adjacentvalve or pipeline surfaces, and generate highnoise levels.

Hydrodynamic noise sounds similar to that ofgravel flowing through a pipe. Intense cavitationcan cause noise levels as high as 115 dBA, butsuch cavitation would not be tolerated becausecavitation damage would drastically shorten theoperating life of the installation. Therefore, controlvalve damage is normally of more concern thanthe noise produced in cavitating services.

Aerodynamic noise is generated by the turbulenceassociated with control of gas, steam, or vapors.While generally thought of as accompanying highcapacity, high pressure systems, damaging noise

6−4

levels can be produced in a two-inch line with aslittle as a 200 psi pressure drop. Major sources ofaerodynamic noise are the stresses or shearforces present in turbulent flow.

Some of the sources of turbulence in gastransmission lines are obstructions in the flowpath, rapid expansion or deceleration ofhigh-velocity gas, and directional changes in thefluid stream. Specific areas that are inherentlynoisy include headers, pressure regulators, linesize expansions, and pipe elbows.

Aerodynamic noise is generally considered theprimary source of control valve noise. There areseveral reasons for this:

� This type of noise has its highest energycomponents at the same frequencies where thehuman ear is most sensitive - between 1000 and8000 hertz.

� Large amounts of energy can be convertedto aerodynamic noise without damaging the valve.In the past, the noise was considered a nuisance,but as long as the valve did its job, it was not ofmajor concern. Today, with increasing focus onenvironmental issues, including noise, there areguidelines on the amount of noise a valve can emitin a given workplace. Research has also shownthat sustained noise levels above 110 decibelscan produce mechanical damage to controlvalves.

High noise levels are an issue primarily becauseof OSHA’s standards for permissible noise limitsand the potential for control valve damage above110 dBA. Additionally, loud hydrodynamic noise issymptomatic of the more dangerous anddestructive phenomenon known as cavitation.

Noise PredictionIndustry leaders use the InternationalElectrotechnical Commission standard IEC534-8-3. This method consists of a mix ofthermodynamic and aerodynamic theory andempirical information. This method allows noiseprediction for a valve to be based only upon themeasurable geometry of the valve and the serviceconditions applied to the valve. There is no needfor specific empirical data for each valve designand size. Because of this pure analytical approachto valve noise prediction, the IEC method allowsan objective evaluation of alternatives.

The method defines five basic steps to noiseprediction:

1. Calculate the total stream power in the processat the vena contracta.

The noise of interest is generated by the valve inand downstream of the vena contracta. If the totalpower dissipated by throttling at the venacontracta can be calculated, then the fraction thatis noise power can be determined. Because poweris the time rate of energy, a form of the familiarequation for calculating kinetic energy can beused. The kinetic energy equation is:

Ek � 1�2mv2

where,m = massv = velocity

If the mass flow rate is substituted for the massterm, then the equation calculates the power. Thevelocity is the vena contracta velocity and iscalculated with the energy equation of the first lawof thermodynamics.

2. Determine the fraction of total power that isacoustic power.

This method considers the process conditionsapplied across the valve to determine theparticular noise generating mechanism in thevalve. There are five defined regimes dependentupon the relationship of the vena contractapressure and the downstream pressure. For eachof these regimes an acoustic efficiency is definedand calculated. This acoustic efficiencyestablishes the fraction of the total stream power,as calculated in step one, which is noise power. Indesigning a quiet valve, lower acoustic efficiencyis one of the goals.

3. Convert acoustic power to sound pressure.

The final goal of the IEC prediction method is todetermine the sound pressure level at a referencepoint outside the valve where human hearing is aconcern. Step two delivers acoustic power, whichis not directly measurable. Acoustic or soundpressure is measurable and, therefore, hasbecome the default expression for noise in mostsituations. Converting from acoustic power to thesound pressure uses basic acoustic theory.

4. Account for the transmission loss of the pipewall and restate the sound pressure at the outsidesurface of the pipe.

Steps one and three are involved with the noisegeneration process inside the pipe. There are

6−5

times when this is the area of interest, but thenoise levels on the outside of that pipe are theprime requirement. This method must account forthe change in the noise as the reference locationmoves from inside the pipe to outside the pipe.

The pipe wall has physical characteristics, due toits material, size, and, shape, that define how wellthe noise will transmit through the pipe. Thefluid-borne noise inside the pipe interacts with theinside pipe wall causing the pipe wall to vibrate,then the vibration transmits through the pipe wallto the outside pipe wall, and there the outside pipewall interacts with the atmosphere to generatesound waves. These three steps of noisetransmission are dependent upon the noisefrequency. The method represents the frequencyof the valve noise by determining the peakfrequency of the valve noise spectrum. It alsodetermines the pipe transmission loss as afunction of frequency. The method then comparesthe internal noise spectrum to determine howmuch the external sound pressure will beattenuated by the pipe wall.

5. Account for distance and calculate the soundpressure level at the observer’s location.

Step four delivers the external sound pressurelevel at the outside surface of the pipe wall.Again, basic acoustic theory is applied to calculatethe sound pressure level at the observer’slocation. Sound power is constant for any givensituation, but the associated sound pressure levelvaries with the area of distributed power. As theobserver moves farther away from the pipe wall,the total area of distributed sound powerincreases. This causes the sound pressure levelto decrease.

Methods to Attenuate NoiseWith increasing interest in the environmentalimpact of all aspects of industry, there areincreasing demands for noise abatementprocedures and equipment.

In a closed system, (not vented to theatmosphere) noise becomes airborne only bytransmission through the valves and adjacentpiping that contains the flowstream. The soundfield in the flowstream forces these solidboundaries to vibrate, causing disturbances in thesurrounding air to propagate as sound waves.

Figure 6-1. Whisper Trim I cage used for reducingaerodynamic noise

W1257/IL

Noise control techniques fall into one of two basiccategories:

� Source treatment

� Path treatment

While preventing noise at the source is thepreferred approach to noise control, it issometimes economically or physically impracticaldue to particular application requirements. Pathtreatment is then a reasonable approach. Thereare also instances when source treatment alonedoes not provide sufficient noise reduction; pathtreatment is then used as a supplement.

In any event, the decision to use sourcetreatment, path treatment, or a combination ofboth should be made only after the applicationrequirements and alternative approaches havebeen thoroughly analyzed.

Source TreatmentThe Fisher Whisper Trim� I cage, illustrated infigure 6-1 , is interchangeable with standard trim inmany globe valves. It uses many narrow parallelslots designed to minimize turbulence and providea favorable velocity distribution in the expansionarea of the valve. It provides a multitude of lownoise flowpaths, which combine to produce lessoverall noise than standard cages. A Whisper TrimI cage is most efficient when the ratio of pressuredrop to inlet pressure is equal to or less than 0.65(that is, ΔP/P1 is less than or equal to 0.65). Inaddition, this approach is most effective when themaximum downstream velocity of the fluid is equalto or less than half the sonic velocity of that fluid.This style of cage will provide up to 18 dBAattenuation versus a standard cage with littlesacrifice in flow capacity.

6−6

Figure 6-2. Valve with Whisper Trim I and Inline Diffuser Combination

W2618

When the pressure drop ratio exceeds 0.65, theWhisper Trim I cage loses its effectiveness.Diffusers, used in conjunction with the WhisperTrim I cage to divide the overall pressure drop intotwo stages, can extend the useful capability andalso improve noise performance (figure 6-2). Thediffuser provides a fixed restriction, whichincreases backpressure to the valve therebyreducing the pressure drop across the valve. Thisdecreases the pressure drop ratio which in turndecreases the sound pressure level. The use of adiffuser allows the Whisper Trim I cage to remainwithin its most efficient P/P1 range. Diffusers areonly effective for the condition they are sized for.They are not effective in throttling applications. Atthis optimum condition they can provide up to anadditional attenuation of 25 dBA.

When pressure drop ratios are high, a WhisperTrim III cage (figure 6-3) may be used. Fluid flowsfrom the inside of the cage out through manyorifices. The performance of these cages isclosely tied to spacing of these orifices. As thepressure drop ratio increases, the centerlinedistance to hole diameter of these orifices alsoneeds to increase to prevent jet recombination.Therefore, as the level of the Whisper Trim IIIcage increases, so does the centerline distance tohole diameter. For many applications involvinghigh pressure drop ratios, a baffle is installedoutside the cage. For very high pressure drop

Figure 6-3. Whisper Trim III

W9039

ratios a baffle is often used to act on the fluid jetsexiting from the cage to further reduce turbulence.Cages similar to the Fisher Whisper Trim III cagecan reduce control valve noise by as much as 30dBA. These cages are most effective when themaximum downstream velocity of the fluid is equalto or less than 0.3 of the sonic velocity of thatfluid.

6−7

Figure 6-4. WhisperFlo Technology

W7065

W7056

Fisher WhisperFlo� trim (figure 6-4) is well-suitedfor applications that have high noise levels andrequire large Cvs. It is effective in applications thathave a pressure drop ratio up to 0.99. When apressure drop ratio of .94 or higher is expected,and WhisperFlo is desired, the noise calculationswill be performed by the engineering experts atEmerson Process Management. This design is amulti-path, two-stage design that has thecapability of reducing noise up to 50 dBA. The keyfactor behind this attenuation is allowing thepressure to recover between stages. This allowsfor the pressure drop ratio of the second stage tobe less than the pressure drop ratio of the firststage. In achieving this, along with specialpassage shapes, the frequency is shifted to ahigher spectrum, velocities are managed, and thejets maintain independence.

All of the Whisper Trim cages and WhisperFlotrims are designed for sliding stem valves. Inapplications requiring rotary valves that have highnoise, an attenuator, diffuser, or combination thereof may be applied. Applications with ball valvescan apply an attenuator to obtain up to 10 dBAreduction in noise. These attenuators aredesigned to reduce both aerodynamic andhydrodynamic noise. With butterfly valves you canonly attenuate aerodynamic noise utilizing aninline diffuser. As mentioned above, thesediffusers can provide up to a 25 dBA reduction innoise.

Figure 6-5. Vee-Ball Noise Attenuator

W6116

For control valve applications operating at highpressure ratios (ΔP/P1 is greater than 0.8), aseries approach can be very effective inminimizing the noise. This approach splits the totalpressure drop between the control valve and afixed restriction (such as a diffuser) downstream ofthe valve. In order to optimize the effectiveness ofthe diffuser, it must be designed for each uniqueinstallation so that the noise levels generated bythe valve and diffuser are equal.

Control systems venting to atmosphere aregenerally very noisy, as well. This is because ofthe high pressure ratios and high exit velocitiesinvolved. In these applications, a vent silencermay be used to divide the total pressure dropbetween the actual vent and an upstream controlvalve (figure 6-6). This approach quiets both thevalve and the vent. A properly sized vent silencerand valve combination can reduce the overallsystem noise level by as much as 60 dBA.

Path TreatmentPath treatment can be applied where sourcetreatment is more expensive, or in combinationwith source treatment where source treatmentalone is inadequate. Path treatment consists ofincreasing the resistance of the transmission pathto reduce the acoustic energy that is transmittedto the environment. Common path treatmentsinclude the use of:

� Heavy walled pipe

The noise attenuation possible with heavy-walledpipe varies with the size and schedule used. As anexample, increasing a pipeline from schedule 40to schedule 80 may reduce sound levels byapproximately 4 dB.

6−8

Figure 6-6. Valve and Vent Diffuser Combination

W2672

� Acoustical or thermal insulation

The noise level near the valve can be lowered byapplying insulation to absorb the noise. Insulationabsorbs much of the noise that would normallyreach the atmosphere, but does not absorb any ofthe noise going up or down inside the pipe walls.

Thermal insulation can give 3 to 5 dBA noisereduction per inch of insulation thickness to amaximum attenuation of 12 to 15 dBA. Acousticalinsulation can give 8 to 10 dBA noise reductionper inch of blanket type insulation. The maximumattenuation that should be expected is 24 to 27dBA.

Path treatments such as heavy-walled pipe orexternal insulation can be a very economical andeffective technique for localized noise abatement.However, they are effective for localized noisereduction only. That is, they do not reduce thenoise in the process stream, but only shroud itwhere the treatment is used. Noise propagates forlong distances via the fluid stream and theeffectiveness of the treatment ends where thetreatment ends.

� Silencers

The silencer differs from other path treatments inthat it does actually absorb some of the noiseenergy. Therefore, it reduces sound intensity bothin the working environment and in the pipeline. Ingas transmission systems, in-line silencerseffectively dissipate the noise within the fluidstream and attenuate the noise level transmittedto the solid boundaries. Where high mass flowrates and/or high pressure ratios across the valveexist, an in-line silencer is often the most realisticand economical approach to noise control. Use ofabsorption-type in-line silencers can providealmost any degree of attenuation desired.However, economic considerations generally limitnoise attenuation to approximately 35 dBA.

Hydrodynamic NoiseThe primary source of hydrodynamic noise iscavitation. Recall that cavitation is the formationand subsequent collapse of vapor bubbles in aflowing liquid. This phenomenon sounds similar tothat of gravel flowing down the pipe.

Source treatment for noise problems associatedwith control valves handling liquid is directedprimarily at eliminating or minimizing cavitation.Cavitation and its associated noise and damagecan often be avoided at the design stage of aproject by giving proper consideration to serviceconditions. However, where service conditions arefixed, a valve may have to operate at pressureconditions normally resulting in cavitation. In suchinstances, noise control by source treatment canbe accomplished by using one of several methods;multiple valves in series, a special control valve, orthe use of special valve trim that uses the seriesrestriction concept to eliminate cavitation.

Cavitrol Trim is a source treatment solution as iteliminates cavitation across the control valve. Thisis achieved by staging the pressure drop acrossthe valve so the pressure of the fluid never dropsbelow its vapor pressure (figure 6-7). Cavitrol Trimis only effective in clean processes. If a processcontains particulate, it will require Dirty ServiceTrim (DST). DST also operates on the concept ofstaging the pressure drops (figure 6-8).

While path treatment of aerodynamic noise isoften an economical and efficient alternative, pathtreatment of hydrodynamic noise is not generallyrecommended. This is because the physicaldamage to control valve parts and piping producedby cavitation is generally a much more seriousissue than the noise generated.

6−9

Figure 6-7. Cavitrol III Trim

W2479

Figure 6-8. NotchFlo DST Trim forFisher Globe Style Valves

W8538

However, if cavitation damage can be eliminatedusing the special trims discussed, it becomespractical to use the path treatment method tofurther reduce the local noise caused by thecavitating liquid. This may be accomplishedthrough the use of heavy-walled pipe andacoustical or thermal insulation.

Much technology now exists for predicting andcontrolling noise in the industrial environment.

Prediction techniques accurately alert the designerto the need for noise control. When it is a problem,a variety of solutions are available ranging fromsimple path insulation to sophisticated controlvalves which eliminate noise at the source.

Two-Phase NoiseAs the properties of the fluids vary, the noisegeneration, propagation, and pipe excitationprocesses area are all affected. Acoustical wavespeed and the density of the fluid are keyconsiderations. In an all gas or all liquidapplication, these are reasonably predictable atany point from the inlet of the downstream piping.

However, for a multiphase fluid, eitherone-component or two-component, there can betremendous variations in these importantparameters. In fact, at the vena contracta wherethe velocities are greatest, the phases mayseparate and form annular flow, with the gas andthe liquid phases having different velocities. Thispossibility makes the noise generation processnearly impossible to model.

Between the vena contracta and the downstreampiping, the phases may be re-oriented to ahomogenous mixture. Propagation of a pressurewave in this region would be again nearlyimpossible to model, as even if it is perfectlyhomogenous, the void fraction would be constantlychanging with pressure.

Wave speed and density are also important indetermining the efficiency with which a sound fieldis coupled to the pipe wall to cause vibrations andsubsequent external noise radiation.

Emerson engineers have conducted field studieson applications where flashing noise was presentin an attempt to quantify the problem, if indeedthere was one. After an extensive search therewere not any applications which were considerednoise problems, nor have any surfaced since.

Based upon this experience, two conclusions weremade:

1. 1. A technically appropriate two-phase noiseprediction method does not exist

2. Two-phase, or pure flashing, applications donot create noise problems.

6−10

Control Valve Noise SummaryThe requirement for noise control is a function oflegislation to protect our wellbeing and to preventphysical damage to control valves and piping.

Noise prediction is a well defined science. Actualresults will be within 5 dBA of predicted levels.

Prediction is based upon contributions for:

� Pressure drop

� Flow rate

� P/P1 and trim style

� Piping and insulation

� Downstream pressure

Noise reduction is accomplished in two generalways:

1. Source treatment, which acts upon the amountof noise generated

2. Path treatment, which blocks transmission onnoise to the environment.

There are two common source treatments:

1. Valve noise trim is based on principles ofdividing the flow to create many small noisesources which combine to a lower level than asingle large flow noise. Diffusers used with controlvalves share pressure drop creating two lowernoise sources which again combine to an overalllower level.

2. Path treatment involves use of insulation orabsorptive devices to lower the sound level whichreaches observers.

Hydrodynamic noise from liquid flow streams canmainly be traced to cavitation. In this case,damage from the cavitation is of more concernthan the noise. Appropriate treatment of thecavitation source should be initiated throughstaging the pressure drop.

Two-phase, or pure flashing, applications do notcreate noise problems, and there is no technicallyappropriate two-phase noise prediction method.

www.Fisher.com

Chapter 7

Steam Conditioning

IntroductionPower producers have an ever-increasing need toimprove efficiency, flexibility, and responsivenessin their production operations. Changes resultingfrom deregulation, privatization, environmentalfactors, and economics are combining to alter theface of power production worldwide. Thesefactors are affecting the operation of existingpower plants and the design of future plantsresulting in a myriad of changes in the designsand operating modes of future and existing powerplants. 3-

Competing in today’s power market requiresheavy emphasis on the ability to throttle backoperations during non-peak hours in order tominimize losses associated with power pricesfalling with demand. These changes areimplemented in the form of increased cyclicaloperation, daily start and stop, and faster ramprates to assure full load operation at daily peakhours.

Advanced combined cycle plants are nowdesigned with requirements including operatingtemperatures up to 1500°F, noise limitation inurbanized areas, life extension programs,cogeneration, and the sale of export steam toindependent customers. These requirementshave to be understood, evaluated, andimplemented individually with a minimum of costand a maximum of operational flexibility to assureprofitable operation.

Great strides have been made to improve heatrates and increase operational thermal efficiencyby the precise and coordinated control of thetemperature, pressure, and quality of the steam.Most of the steam produced in power and processplants, today, is not at the required conditions forall applications. Thus, some degree of

conditioning is warranted in either control ofpressure and/or temperature, to protectdownstream equipment, or desuperheating toenhance heat transfer. Therefore, the sizing,selection, and application of the properdesuperheating or steam conditioning systems arecritical to the optimum performance of theinstallation.

Thermodynamics of SteamHighly superheated steam, (i.e. 900 - 1100°F) isusually generated to do mechanical work such asdrive turbines. As the dry steam is expandedthrough each turbine stage, increasing amounts ofthermal energy is transformed into kinetic energyand turns the turbine rotor at the specified speed.In the process, heat is transferred and work isaccomplished. The spent steam exits the turbineat greatly reduced pressure and temperature inaccordance with the first law of thermodynamics.

This extremely hot vapor may appear to be anexcellent source for heat transfer, but in reality it isjust the opposite. Utilization of superheated steamfor heat transfer processes is very inefficient. It isonly when superheated steam temperatures arelowered to values closer to saturation that its heattransfer properties are significantly improved.Analysis has shown that the resultant increase inefficiency will very quickly pay for the additionaldesuperheating equipment that is required.

In order to understand why desuperheating is soessential for optimization of heat transfer andefficiency, we must examine the thermodynamicrelationship of temperature and the enthalpy ofwater. Figure 7-1 illustrates the changes of statethat occur in water over a range of temperatures,at constant pressure, and relates them to theenthalpy or thermal energy of the fluid.

7−2

Figure 7-1. Temperature enthalpy diagram forwater. Note that the greatest amount of thermal

energy input is used to vaporize the water.Maximum efficiency in heat transfer requires

operation at near saturation temperature to recoverthis energy.

E0117

300

200

100

32

0

0 500 1000

Btu added to 1 pound of water

Tem

per

atau

re, d

ef F

1/2 Btu per degree

Ice heating at about

144 Btu to melt ice

Water heating at 1 Btu per degree

Atmospheric pressure

Evaporation at 14.7 psi

Evaporation at more than 14.7 psi

All data for 1 lb. water

which water cannot exist as a liquid

These lines curve and meet at 705.4 deg F the critical temperature, above

Ste

am s

up

erh

eati

ng

at

abo

ut

0.4

Btu

per

deg

ree

212 deg F

970 Btu to

boil water

In the lower left portion of the graph, the water isfrozen at atmospheric pressure and below 32°F.At this point, heat is being rejected from the wateras it maintains its solid state. As heat is graduallyadded the ice begins to change. Addition of heatto the ice raises the temperature and slows therate of heat rejection. It requires approximately1/2 BTU of thermal energy to be added to a poundof ice to raise its temperature 1°F. Upon reaching32°F, the addition of more heat does notimmediately result in an increase in temperature.Additional heat at this point begins to melt the iceand results in a transformation of state from asolid to a liquid. A total of 144 BTUs is required tomelt one pound of ice and change it to water at32°F.

Once the phase change from a solid to a liquid iscomplete, the addition of more heat energy to thewater will again raise its temperature. One BTU ofheat is required to raise the temperature of onepound of water by 1°F. This relationship remainsproportionate until the boiling point (212°F) isreached. At this point, the further addition of heatenergy will not increase the temperature of thewater. This is called the saturated liquid stage.

Figure 7-2. Temperature enthalpy diagram forwater showing that saturation temperature varies

with pressure. By choosing an appropriatepressure, both correct system temperature and

thermal efficiency can be accommodated.

Tem

per

atau

re, d

ef F

T-H DIAGRAMWATER

LIQUID800 PSIA

14.7 PSIA

LIQUID-VAPORVAPOR

ENTHALPY, BTU/LBME0118

The water begins once again to change state, inthis case from water to steam. The completeevaporation of the water requires the addition of970 BTUs per pound. This is referred to as thelatent heat of vaporization, and is different at eachindividual pressure level. During the vaporizationprocess the liquid and vapor states co-exist atconstant temperature and pressure. Once all thewater, or liquid phase, has been eliminated wenow have one pound of steam at 212°F. This iscalled the saturated vapor stage. The addition offurther thermal energy to the steam will now againincrease the temperature. This process is knownas superheating. To superheat one pound ofsteam 1°F requires the addition of approximately0.4 BTUs of thermal energy.

The potential thermal energy release resultingfrom a steam temperature change differssignificantly depending on temperature andsuperheat condition. It is much more efficient, ona mass basis, to cool by addition of ice rather thanby the addition of cold fluids. Similarly, it is moreefficient to heat with steam at temperatures nearthe saturation temperature rather than in thesuperheated region. In the saturated region muchmore heat is liberated per degree of temperaturechange than in the superheated range because

7−3

production of condensate liberates the enthalpy ofevaporation, the major component of the totalthermal energy content. The temperature-enthalpy diagram in figure 7-2 is generalized toshow the thermodynamic relationship at variouspressures.

The graph in figure 7-2 illustrates three distinctphases (i.e., liquid, vapor, and liquid-vapor) andhow enthalpy relates to temperature in each phaseat constant pressure. The rounded section in themiddle of the graph is called the ”steam dome”and encompasses the two-phase, liquid-vaporregion. The left boundary of the steam dome iscalled the saturated liquid line. The right boundaryline is the saturated vapor line. The twoboundaries meet at a point at the top of the domecalled the critical point. Above this point, 3206 psiand 705°F, liquid water will flash directly to drysteam without undergoing a two-phasecoexistence. When conditions exceed this criticalpoint they are considered to be existing in thesupercritical state.

In the lower left side of the graph, the saturatedliquid line intersects the temperature axis at 32°F.At this point we have water and a defined enthalpyof 0 BTU/LB. As heat is added to the system, thetemperature and enthalpy rise and we progress upthe saturated liquid line. Water boils at 212°F at14.7 psia. Thus, at 212°F and 180 BTU/LB, wenote a deviation from the saturated liquid line.The water has begun to boil and enter a newphase; Liquid-Vapor.

As long as the liquid stays in contact with thevapor, the temperature will remain constant asmore heat is added. At 1150 BTU/LB (at 14.7 psi)we break through to the saturated vapor line.Thus, after inputting 970 BTU/LB, all of the waterhas been vaporized and enters the pure vaporstate. As more heat is added, the temperaturerises very quickly as the steam becomessuperheated.

Why Desuperheat?Desuperheating, or attemperation as it issometimes called, is most often used to:

� Improve the efficiency of thermal transfer inheat exchangers

� Reduce or control superheated steamtemperatures that might otherwise be harmful toequipment, process or product

� Control temperature and flow with loadvariation

Dry superheated steam is ideally suited formechanical work. It can be readily converted tokinetic energy to drive turbines, compressors andfans. However, as the steep temperature-enthalpy line slope would indicate, the amount ofheat output per unit of temperature drop is verysmall. A heat exchanger using superheatedsteam would have to be extremely large, use greatquantities of steam, or take tremendoustemperature drops. A 10°F drop in temperatureliberates only 4.7 BTU per pound.

If this same steam had been desuperheated tonear saturation the thermal capabilities would begreatly enhanced. The same 10°F drop intemperature would result in the release of over976 BTU of heat. This illustrates the obviousadvantages of desuperheating when thermalprocesses are involved. Only by desuperheatingthe superheated steam is it possible toeconomically retrieve the energy associated withvaporization. By changing steam pressure, thesaturation temperature can be moved to match thetemperature needs of the process and still gainthe thermal benefits of operating near saturation.

The previous discussion centered on why wesuperheat steam (to do mechanical work) andwhen it should be desuperheated back tosaturation (to heat). There are many situationswhen saturated steam suddenly andunintentionally acquires more superheat than thedownstream process was designed toaccommodate. This “unintentional” superheatproduces the same thermal inefficienciesmentioned previously. In this case, we are talkingabout the sudden expansion and temperaturechange associated with a pressure reducing valve.Take the following steam header conditions forexample:

Conditions: P1 = 165 psia T1 = 370°F Enthalpy = 1198.9 BTU/LB

Saturation temperature at 165 psia is 366°F.Therefore, the steam has only 4°F of superheatand would be excellent for heat transfer. Assumethat another thermal process requires somesteam, but at 45 psia rather than 165 psia. Thesimple solution is to install a pressure reducingvalve. Since throttling devices, such as valvesand orifices, are isenthalpic (constant enthalpyprocesses) the total heat content of the steam willnot change as flow passes through the restriction.

7−4

After the valve, the steam will have the followingconditions:

Conditions: P2 = 45 psia Enthalpy = 1198.9 BTU/LB

Referencing a set of steam tables, we see that atthe above conditions the steam temperature is328°F giving the impression that it has cooled.However, from the steam tables we see that thesaturation temperature for 45 psia steam has alsodropped to 274°F. The net result is that our steamnow has 54°F of superheat (328°F - 274°F). Useof this steam for heat transfer could beuneconomical and return on investment on adesuperheater would be most favorable.

Desuperheating

In this section we will briefly discuss the processof desuperheating. The need to desuperheat isusually performed simply to control the steamtemperature, or heat content, of the flowing vapormedia. Depending on the process downstream ofthe main steam source, a desuperheater will beutilized to transform the steam into a medium thatis more efficient for heat transfer or just moreconducive for interaction with its surroundingcomponents. One means of accomplishing this iswith a direct contact heat transfer mechanism.This can easily be achieved by the use of a singlespray injection nozzle that, when properly placed,diffuses a calculated quantity of liquid into theturbulent flow stream. Vaporization of the liquidphase proceeds while mass, momentum, andenergy transfer occurs, and the resultant vaporexits the process at the desired temperature orheat content level.

Desuperheaters

A desuperheater is a device that injects acontrolled amount of cooling water into asuperheated steam flow in an effort to reduce orcontrol steam temperature (figure 7-3).Desuperheaters come in various physicalconfigurations and spray types that optimizeperformance within specified control andinstallation parameters. Selection should alsoalways include attention to those details that wouldprovide the most economic solution withoutsacrificing required performance.

Figure 7-3. Insertion style desuperheater injects acontrolled amount of cooling water into super-

heated steam flow.

E0865

The success of a particular desuperheater stationcan rest on a number of physical, thermal, andgeometric factors. Some of the factors are quiteobvious and others are more obscure, but they allhave a varying impact on the performance of theequipment and the system that it is installed in.Considerable research has been conducted intothe characteristics of desuperheaters and thetransformation of spraywater to vapor. Thefindings are of considerable interest to both designand process engineers. In the next severalsections, we will discuss these findings and howthey relate to the desuperheating system as awhole.

The most important factor is the selection of thecorrect desuperheater type for the respectiveapplication. Units come in all shapes and sizesand use various energy transfer and mechanicaltechniques to achieve the desired performancecriteria and optimize the utilization of the systemenvironment. These design criteria include:

� Mechanically Atomized − Fixed and VariableGeometry Spray Orifice

� Geometrically Enhanced

� Externally Energized

The mechanically atomized style of desuperheateris the most popular and simplistic style thatprovides nominal performance over a wide rangeof flow and conditions. These models are of theinternally energized variety. The atomization andinjection of the spray water is initiated by thepressure differential between the spraywater andthe steam. The DMA, fixed geometry sprayorifice, units are the simplest and by design havea constant area flow path. These units are highly

7−5

Figure 7-4. The DMA/AF desuperheater utilizes variable-geometry, back-pressure

activated spray nozzles.

W6310-1

dependent on the pressure differential and thusprovide levels of performance that arecommensurate with the magnitude of thedifference. Obviously, the larger the water/steamdifferential the better the unit will perform (i.e.,penetration velocity, flow variation and dropletsize). Since the equipment turndown is usuallylimited to 4:1, it is best suited for near steady loadapplications.

An upgrade from the fixed geometry unit is theDMA/AF (figure 7-4) variable geometry nozzledesuperheater. Here the actual flow geometry ofthe unit is varied to maintain an optimumdifferential across the discharge orifice. As aresult of this change, the level of flow variation isgreatly enhanced and so is the performance.Equipment turndowns can exceed 40:1, thusmaking this style a good choice for medium tohigh load change applications.

Another form of mechanically atomizeddesuperheater is the DVI, Geometric Enhancedstyle, (figure 7-5). Here, the unit is supplied a highpressure recovery flow restriction that alters flowgeometry and helps to keep the level of turbulenceand kinetic energy at a high level during all phasesof the units operation due to an increased velocityat the point of spray water injection. Thisincreased level of surrounding energy helps to

Figure 7-5. The DVI desuperheater injectsspraywater in the outlet of the venturi section,

assuring excellent mixing and rapid atomization.

W6313-1

impart energy transfer to the droplets and assistsin break-up, mixing, and vaporization. This style isbest suited for medium turndown applicationstypically around 15:1.

The last group of desuperheater units utilizes anexternal energy source for the atomization of thespraywater. The most common medium is a highpressure steam source. In this case, the highlevels of kinetic energy are provided by a criticalpressure reduction in the desuperheatersprayhead. The critical drop is used to shear thewater into a fine mist of small droplets, which isideal for vaporization, as shown in figure 7-6. Thistype of system can provide a very high degree offlow variation without requiring a high pressurewater supply. Applications requiring turndownranges greater then 40:1 utilize this type ofequipment for best performance. In addition to anexternal spraywater control valve, the system willalso require an atomizing steam shut-off valve(figure 7-7).

Other factors that have a large amount of impacton the performance of a desuperheating systeminclude:

� Installation Orientation

� Spray Water Temperature

� Spray Water Quantity

� Pipeline Size

� Equipment vs. System Turndown

7−6

Figure 7-6. The DSA desuperheater useshigh-pressure steam for rapid and complete

atomization of spraywater in low-velocity steamlines.

W6311-2

Installation orientation is often overlooked, but acritical factor in the performance of the system.Correct placement of the desuperheater can havemore impact on the operation than the style of theunit itself. For most units, the optimum orientationis in a vertical pipeline with the flow direction up.This flow direction is ideal, as the natural flowdirection of the injected water tends to be in thecounter direction due to effect of gravity. The roleof gravity in this orientation will suspend thedroplets in the flow longer while they are beingevaporated, thus shortening the requireddownstream distance or efficient mixing.

Other orientation factors that are of concerninclude downstream pipefittings, elbows, and anyother type of pipeline obstruction that can providea point for water impingement or fallout.

Spraywater temperature can have an great impacton the desuperheater performance. While it goesagainst logical convention, hotter water is betterfor cooling. As the temperature increases andmoves closer to saturation, its flow and thermalcharacteristics are improved and impact mostsignificantly the following:

� Surface Tension

� Drop Size Distribution

� Latent Heat of Vaporization

� Vaporization Rate

Improvement in all these areas will act to improvethe overall performance of the system, as thespraywater will evaporate and mix with the steamat a faster rate.

The quantity of water to be injected will, as withany mass flow calculation, have a directlyproportionate affect on the time for vaporization.

The heat transfer process is time dependent; thus,the quantity of spray water will increase the timefor complete vaporization and thermal stability.

Another concern for proper system performance ispipeline size. Pipe size should be determined in aneffort to balance the velocity of the steam flow.Steam traveling at a fast rate will require longerdistances to effectively cool, as heat transfer is afunction of time. Steam traveling at low velocitywill not have enough momentum to suspend waterdroplets long enough for evaporation. As a result,water will fall out of the steam flow to collect alongthe bottom of the pipe, and it will not cool thesteam effectively. Ideal velocity is typically in therange of 250 ft/sec to 300 ft/sec.

As the pipeline size increases to limit steamvelocity, more attention must be paid to thepenetration velocity of the spray and the coveragein the flow stream. Experience shows that singlepoint injection type desuperheaters will haveinsufficient nozzle energy to disperse throughoutthe entire cross-sectional flow area of the pipeline.As a result, the spray pattern collapses andthermal stratification occurs (i.e., sub-cooledcenter core within a superheated outer jacket.)

This condition normally is eliminated after the flowstream undergoes several direction changes,although this is not always possible within thelimits of the control system or process. Properplacement of high-energy TBX-T (figure 7-8)multi-nozzle steam coolers in the larger pipelineswill normally prevent thermal stratification.

The most over used and misunderstood word inthe field of desuperheating is “turndown.” Whenapplied to a final control element, such as a valve,it is a simple ratio of the maximum to minimumcontrollable flow rate. Turndown is sometimesused interchangeably with rangeability; however,the exact meaning differs considerably when it

7−7

Figure 7-7. The DSA desuperheater utilizes two external control valves: a spraywater unit and an atomizing steam valve.

DSA DESUPERHEATERC0817 / IL

Figure 7-8. TBX-T Cooler

W8786-1

comes to actual performance comparisons. Sincea desuperheater is not a final control element itsperformance is linked directly to its systemenvironment; thus, the actual turndown is more afunction of system parameters rather than basedon the equipment’s empirical flow variations. Oncethis is understood, it is obvious that even a gooddesuperheater cannot overcome the limitations ofa poorly designed system. They must beevaluated on their own merits and weighedaccordingly.

A final design parameter for all insertion typedesuperheaters is its ability to withstand highlevels of thermal cycling. Due to the nature ofoperation of today’s plants, desuperheaters shouldbe designed with the intent to operate under dailycycling environments. Exposure to frequent dailycycling can lead to thermal fatigue and

desuperheater failure if the unit is not designed forthe operation. Design upgrades for this applicationconsist of thermal liners to reduce thermal loadsand structural optimization to reduce inducedvibration at stress sensitive welds.

To summarize the requirements to correctly size adesuperheater, the following system and operatinginformation is required:

� Minimum and Maximum Steam Flow

� Steam Pressure and Temperatures

� Cooling Water Pressure and Temperature

� Required System Turndown Ratio

� Pipe Size and System Layout

� Planned Mode of Operating

Steam Conditioning ValvesSteam conditioning valves representstate-of-the-art control of steam pressure andtemperature by integrally combining both functionswithin one control element unit. These valvesaddress the need for better control of steamconditions brought on by increased energy costsand more rigorous plant operation. Steamconditioning valves also provide bettertemperature control, improved noise abatement,and require fewer piping and installationrestrictions than the equivalent desuperheater andpressure reduction station.

7−8

Figure 7-9. The TBX utilizes an externalspraywater manifold with multiple nozzles for

moderate to large volume applications.

W8493-1

Steam conditioning valve designs can varyconsiderably, as do the applications they arerequired to handle. Each has particularcharacteristics or options that yield efficientoperation over a wide range of conditions andcustomer specified requirements.

The TBX steam-conditioning valve (figure 7-9)combines pressure and temperature control in asingle valve. Finite element analysis (FEA) andcomputational fluid dynamic (CFD) tools wereused in its development to optimize the valve’soperating performance and overall reliability. Therugged design of the TBX proves capable ofhandling full mainstream pressure drops, while itsflow-up configuration, in conjunction with WhisperTrim technology, prevents the generation ofexcessive noise and vibration.

The simplified trim configuration used in the TBXaccommodates rapid changes in temperature asexperienced during a turbine trip. The cage iscasehardened for maximum life and is allowed toexpand during thermally induced excursions. Thevalve plug is continuously guided and utilizescobalt-based overlays both as guide bands and toprovide tight, metal-to-metal shutoff against theseat.

Figure 7-10. Detail of AF Spray Nozzle.

W8494-1

The TBX incorporates a spraywater manifolddownstream of its pressure reduction stage. Themanifold features variable geometry, backpressureactivated AF nozzles that maximize mixing andquick vaporization of the spraywater.

The AF nozzle (figure 7-10) was developedoriginally for condenser dump systems in whichthe downstream steam pressure can fall below thesaturation level. In these instances, thespraywater may flash and significantly change theflow characteristic and capacity of the associatednozzle at a critical point in the operation.

Spring loading of the valve plug within the AFnozzle prevents any such changes by forcing theplug to close when flashing occurs. With flashing,the compressibility of the fluid changes, and thenozzle spring will force closure andre-pressurization of the fluid leg. Once this isdone, the fluid will regain its liquid properties andreestablish flow to the condenser.

The TBX injects the spray water towards thecenter of the pipeline and away from the pipe wall.The number of injection points varies byapplication. With high differentials in steampressure, the outlet size of the valve increasesdrastically to accommodate the larger specificvolumes. Correspondingly, an increased numberof nozzles are arranged around the circumferenceof the outlet making for a more even and completedistribution of the spray water (figure 7-11).

The simplified trim arrangement in the TBXpermits extending its use to higher pressureclasses (through ANSI Class 2500) and operatingtemperatures. Its balanced plug configuration

7−9

Figure 7-11. The TBX showing externalspraywater manifold.

W8520

provides Class V shutoff and a linear flowcharacteristic.

The TBX typically uses high-performance,pneumatic piston actuators in combination withFIELDVUE Digital Valve Controllers to achieve fullstroke in less than two seconds while maintaininghighly accurate step response. The FIELDVUEinstruments along with AMS ValveLink� softwareprovide a self-diagnostic capability that givesanswers about valve performance. The currentvalve/actuator signature (seat load, friction, etc.)can be compared against previously storedsignatures to identify performance changes beforethey cause process control problems.

When piping dictates, the TBX valve can beprovided as separate components, allowingpressure control in the valve body andtemperature reduction in a downstream steamcooler. The steam cooler is equipped with a watersupply manifold (multiple manifolds are alsopossible). The manifold provides cooling waterflow to a number of individual spray nozzles thatare installed in the pipe wall of the cooler section.The result is a fine spray injected radially into thehigh turbulence of the axial steam flow.

Installation GuidelinesInstallation of desuperheaters and steamconditioning valves is key to long term success

and performance. It is best to installdesuperheaters in a straight run of horizontal orvertical pipe. Installation in elbows is also possible,but it can affect system turndown and thermalstratification due to momentum caused changes inthe velocity profile.

Momentum forces the majority of the steam flowto the outside surfaces of the bend. This results ina low velocity void on the inside of the elbow. Ifhigh turndowns are not required, this installation issatisfactory since the voids would rarely be belowminimum velocity at maximum flow. As the flow isreduced, however, these areas may lose theirability to perform as required to desuperheat thesteam.

Other installation parameters that are always ofinterest to the piping designer are how muchstraight run of pipe is required and where thetemperature sensor should be located. Both arethermally derived questions and require thermallyderived answers. It is desirable to have thethermal sensor as close as possible to thedesuperheater in order to reduce the signal lagtime. It is also desirable not to have any pipingcomponents (e.g., elbows or tees) that woulddetract from the thermal process.

The following equations provide guidelines fordesigning a proper system. These equationsrelate to time required for complete vaporizationand mixing.

Downstream Straight Pipe Requirements (SPR):SPR (ft) = 0.1 Sec. x Maximum Steam Velocity(ft/sec)

Downstream Temperature Sensor Distance (TS):15% Spraywater or less:TS (ft) = 0.2 Sec. x Maximum Steam Velocity(ft/sec)

Greater than 15% Spraywater:TS (ft) = 0.3 Sec. x Maximum Steam Velocity(ft/sec)

Temperature control is not limited to receiving asignal from a downstream temperature sensor.Another valid alternative is feed-forward control.

Feedforward control is accomplished using analgorithm that is characterized specifically to thevalve installed in the application. The algorithm isprogrammed into the distributed control system toprovide an accurate calculation of the spray waterthat is required to reduce the steam enthalpy andtemperature to the desired outlet set point. Thealgorithm requires input of upstream temperature

7−10

and pressure as well as the position of the valve.Upstream and spraywater enthalpies are thendetermined using an inherent steam table withinthe DCS. The total spraywater required iscalculated from a heat balance using the finalenthalpy into the condenser. This method oftemperature control is a practical solution forapplications that do not have enough downstreampipe distance for accurate measurement by atemperature sensor.

Turbine Bypass SystemsThe most severe and critical application of anysteam conditioning installation is that of the turbinebypass.

The concept of the turbine bypass has beenaround for a long time; however, its applicationand importance has broadened significantly inrecent years. Steam turbine bypass systemshave become essential to today’s power plantperformance, availability, responsiveness, andmajor component protection.

The following will concentrate on the generalapplication of bypass systems as used in fossilfueled utility power plants. The closedwater/steam heat cycle of such typical units maybe comprised, but not limited to, sub- orsuper-critical pressures, to single, double, or triplereheat sections and to condensation at or nearambient temperatures. The steam generatingprinciples where such bypass systems areemployed include natural or assisted circulationdrum boilers, combined circulation boilers, andonce-through boilers. The turbine may be ofsingle or double shaft design and operated eitherat fixed inlet pressure or on sliding pressure.

Bypass System BenefitsJust how beneficial a bypass system proves to bedepends upon many factors (e.g., plant size,mode of operation, age of existing components,size of the condenser, main fuel type, controlphilosophy, etc). However, the main benefits forthe application of a comprehensive bypass systemin the 25-100% size range are:

� The matching of steam and heavy turbinemetal component temperatures during thestartup and shutdown phase. This has provento be of major economic significance in terms offuel savings and the thermal protection of criticalheavy wall boiler and turbine components. By

limiting temperature differentials during turbineadmission the effects of thermal fatigue areminimized and longevity of componentsmaximized. This is especially important for lifeextension programs where the role andjustification of the bypass system may be centeredsolely on this aspect.

� The ability to avoid a boiler trip followinga full load rejection. A boiler (HRSG) / turbineunit with a bypass can withstand a completesystem load rejection and remain available forrapid reloading after the disturbance has beenremoved. This important advantage for systemflexibility and operating efficiency can make thedifference between a more costly and timeconsuming warm start and a hot start.

� Reduction in solid-particle erosion ofturbine components. The loss of material fromthe boiler tubing and internals is most prevalentduring commissioning startup and after the unithas been shutdown for an expended period oftime. Thermal transients assist in the dislodgingof scale, oxides, and weldments within the boilercircuit to form an abrasive steam flow that, overtime, could accelerate the wear of sensitiveturbine blades and seriously affect operatingefficiencies and maintenance costs. Damage canbe reduced or eliminated by routing the steamthrough the bypass system.

� Independent operation of the boiler andturbine set. The ability to operate the boilerwithout the turbine, at any load up to the limit ofthe bypass capacity, can be surprisingly useful foroperational or testing purposes. For example, allboiler controls and firing systems can be testedand fine-tuned independent of the turbineoperation. This significantly reduces both costand time relating to initial commissioning of theplant, retrofitting and checking equipmentperformance, and system troubleshooting.

General System DescriptionA complete and comprehensive turbine bypasssystem can be comprised of many inter-linked andcoordinated components. These include thebypass valves, spray water control valves, controlsystem, and the actuation and positioning system.For this discussion, we will center our attention onthe bypass valves themselves.

The bypass system incorporates the dualoperating function of steam conditioning valves(i.e., for the controlled reduction of both pressureand temperature). The bypass valve incorporates

7−11

Figure 7-12. Turbine Bypass System

E0866

the latest technology in pressure-reducing/lownoise trim to handle the flow and reduction ofpressure energy to acceptable levels. However,since steam throttling in a control valve is anisenthalpic process, desuperheating is required tocontrol the discharge temperature and enthalpylevels. As a result, the valves are equipped with aspecial spraywater injection system that producesa finely atomized and evenly distributed waterinterface for rapid vaporization and steamtemperature control.

The bypass system can be supplied with one ortwo control inputs depending on the role it plays inthe control scheme. If the valve is used solely forstartup and shutdown, it will receive a singlemodulating control signal to position the trim as afunction of the startup and shutdown curves forthe respective unit. If the valve must also act torelieve pressure during a turbine trip or loadrejection, an additional discrete input is includedthat will ramp open the valve quickly to apredetermined position, before reverting to amodulating configuration in accordance with theboiler control requirements. Fast positioning speedand resultant alternate flow path are critical tocounteract the pressure build-up resulting from theisolation of the boiler piping circuit when theturbine valves close in this trip situation.

High Pressure BypassDuring startup, shutdown, or on turbine trip, theHP bypass system directs steam from thesuperheater outlet to the cold reheat line, therebybypassing the HP turbine section (figure 7-9). Themajor advantages of such an action have beengenerally outlined above. However, more specificduties are:

1. Pressure and temperature controlled bypass of the HP turbine section.

2. Controlled main steam pressure build-up in the boiler.

3. Cooling of the reheat section of the boiler.4. Prevention of the opening of spring-loaded

HP safety valves during minor disturbances.5. Avoidance of condensate loss and noise from

blowing safety valves.6. Protection of the boiler against exceeding

design pressures.

The failure mode of the HP bypass system is verydependent on local design codes and theperformance scenario for the system. If it isdesigned as a safety bypass system and replacesthe standard safety relief valve function, the valvesmust always fail in the open position. However, ifthe standard safety relief valves are in place, thevalve is normally required to fail closed, especiallyin over-temperature situations on drum boilers.

7−12

Control of the HP bypass is normally initiated viafeedback input signals from the main steampressure and the cold reheat temperature. Theratio of steam to spraywater is normally inverselyproportional to the respective valve position,especially during startup and shutdown. This isbecause startup conditions normally require largevalve Cvs, due to the large specific volumesassociated with low pressures at hightemperatures, even though flow is greatlyreduced.

During trip conditions, the opposite is true, andlarge quantities of spraywater are required atlower valve openings. For this situation, specialcontrol algorithms usually are incorporated into thecontrol system to provide independent feedforwardcontrol. This is especially important during a tripsequence where time of response is critical tomaintain system integrity, performance, andcomponent protection.

Spraywater for cooling is normally obtained fromthe boiler feed pump discharge and is regulated byan external spraywater control valve that isproperly sized to handle the required flow andpressure drop.

Hot-Reheat and Low PressureBypassDuring startup, shutdown, or on turbine trip, theHRH and LP bypass systems direct steam fromthe hot reheat line to the condenser, thusbypassing the IP and LP turbine sections (figure7-12). The major advantages of such an actionhave been generally outlined above. However,more specific duties are:

1. Pressure and temperature controlled bypassing of the IP and LP turbines.

2. Controlling pressure build-up in the boiler reheat section.

3. Prevention of condensate losses during loadtrips and minor disturbances.

4. Protecting the condenser against excessive pressure, temperature, and enthalpy excursions during bypass operation.

In contrast to the HP bypass, the HRH and LPbypass valves only fail closed as a failure mode.While it is important to control the hot reheatpressure, it is even more critical to protect thecondenser against damage from uncontrolled orimproper admission of steam. The condensermanufacturer interfaces specific condenser controlpermissives with these bypass control systems. If

Figure 7-13. TBX WhisperFlo Sparger.

W8684-2

any of these permissives is not met or is exceededduring bypass operation, the valve is quickly shut.These permissives include, but are not limited to:

1. Condensate level high2. Condenser temperature high3. Condenser pressure high4. Spraywater pressure low5. Loss of coolant

Another added challenge of the HRH and LPbypass system is to properly control the amount ofbackpressure on the bypass valves. A condenseror condenser duct, which is downstream of thesebypass valves, typically operates at a vacuum inthe range of 1 - 3 psia. Given this scenario, it iscrucial to create backpressure in order to maintaina desired velocity within reasonable pipe sizes.

A second challenge to this application is to createthese desired conditions while minimizing thenoise generated by this process. Dumping highvelocity steam into a low pressure, thin wallcondenser/turbine exhaust duct requires carefulevaluation in order to assure steam jets do notconverge. Hole spacing within the sparger andsparger placement within the duct are critical formaintaining low noise levels.

A typical bypass to condenser installation requiresa steam conditioning valve to control pressure andtemperature, a spraywater valve to regulate thewater supply, and a downstream TBX sparger tocreate backpressure. Low noise WhisperFlo trim

7−13

alternatives are also available for the TBX sparger(figure 7-13).

Control of the HRH and LP bypass valvesnormally is initiated via feedback input signalsfrom the hot reheat steam pressure and thespecified condenser inlet temperature/enthalpy.The steam entering the condenser must becontrolled specifically to guard against excessivethermal expansion of the tubing and shell. As inthe case of the HP bypass, the ratio of steam tospray water is normally inversely proportional,especially during startup and shutdown. Inaddition, the dual role of the HRH and LP bypasssystem in controlling the thermal admissionparameters to the condenser normally results inthe requirement for a prescribed amount ofover-spray.

This situation is compounded by the closeproximity of these valves to the condenser. Thismakes any kind of feedback temperature controlalmost impossible considering the quantity ofspray water to be vaporized and the short distanceavailable to measure the process. It is highlyrecommended that feedforward control algorithmsbe incorporated into the control system to provideindependent feedforward control for thespraywater admission.

Spraywater for cooling is normally obtained fromthe condensate boost pump discharge and isregulated by a properly sized external spraywatercontrol valve.

Bypass Size

A comprehensive bypass system includes HPbypass, HRH bypass and LP bypass valves.However, they may or may not be sized for thesame capacity. There are many variables that caninfluence the required size of each bypass system.

Bypasses for once-through boiler plants aregenerally designed for 100% of full-load steam tosuit startup and part-load operation. Ifconventional safety valves are omitted, 100%bypass capacity is essential.

Bypass capacity for drum boiler plants involveseveral different issues. Some argue that 100%capacity bypasses are worthwhile, but experiencehas proven that bypasses with capacities ofbetween 25 - 70% normally are sufficient to handlemost operating and trip conditions.

For temperature matching in a drum plant duringhot startup only, it may be possible to use abypass of only 30% when firing with oil and40-50% for coal. Overall, these values areconsidered the lowest practical load for the boilerunder automatic control.

On bypass applications requiring the control of afull turbine trip, the values increase to 40% on gasand oil-fired drum units and up to 70% for coal. Inselecting the bypass capacity, it is important toconsider all control systems and plant componentsand their ability to turn down instantaneously fromfull to auxiliary load.

Note also that if the high pressure bypass capacityexceeds approximately 50%, and the low pressurebypass passes all the steam to the condenser,then condenser duty during bypass operation ismore severe than during normal, full-load turbineoperation. This fact may limit bypass capacity,especially on systems being retrofit to existingplants.

Starts, Trips, Load Rejection,Two-Shift OperationThe worth of a turbine bypass and the flexibility,added efficiency, and responsiveness are nevermore apparent than during starts, trips, or loadrejections. Modern bypass systems operateduring:

� Cold starts

� Warm starts

� Hot starts

� Load rejection

� Quick turbine shutdown

� Two-shift operation

Bypass valves and systems that are designedcorrectly have noteworthy advantages for theseindividual modes. They are detailed as follows:

Cold StartsA cold start typically occurs after the unit has beendown for over a week. Preheating of the systemis required as first stage and reheat temperaturesare normally below 200°F. The bypass systempermits involvement of the furnace, superheaters,and reheater very early in the steam/water cycle.

7−14

This is important in the production of steam puritybefore the turbine start. Steam flow through thesuperheater and reheater enhances the tubecooling effect, thereby allowing greater latitude ingas and steam temperatures. During the startup,thermal stresses are controlled while achieving thefastest possible loading rate. Depending on thesize of the bypass system, the unit can typicallybe brought on line in 4.5 - 9 hours.

Warm StartsA warm start is indicative of a weekend shutdown.In this case, the HP turbine casing is usuallyabove 450°F. As with the cold start, the steamtemperature can be controlled to permit thematching of steam and metal temperatures underall operating conditions. Expected startup time isbetween 2.5 - 5 hours.

Hot StartsA hot start is usually associated with a minordisturbance that created a unit trip. The bypassallows the boiler to remain on line until thedisturbance is cleared and the unit can bereloaded in the shortest possible time, which isusually between 1 - 2 hours.

Load Rejection/Quick RestartDuring load rejection, the bypass system providesthe necessary control and flow path for unitrunback to minimum load and for theestablishment of a definitive course of action (i.e.,complete shutdown or quick restart). All systemsare protected, and a minimum of condensate islost.

Two-Shift OperationTwo-shift operation may become necessary if autility grid has a number of large base-loadedunits, which are not as maneuverable as thesmaller fossil fueled units used for peakingpurposes. This would require that the smallerunits be shutdown every night and restarted everymorning, which is a very material-life consumingmeans of operating. Once again, the bypasssystem provides a means for the efficient andtimely matching of steam and metal temperatures.This allows the efficient startup of the units everymorning without thermally stressing thecomponents, yet it increases unit efficiency andavailability.

Chapter 7 — Steam ConditioningSummary

The implementation of a properly designed turbinebypass system can be beneficial and instrumentalin the pursuit of increased efficiency, flexibility,and responsiveness in the utility power plant.Component life can be extended as the ability toregulate temperatures between the steam andturbine metal is enhanced. Commissioning timeand cost can be reduced through independentboiler and turbine operation. The magnitude ofreturn on investment hinges on the specificapplication mode, style or service of plant, andequipment supplied. While not discussed here,this logic applies as well to combined cycle plants,cogeneration facilities, and industrial powerfacilities.

Short Notes:

� A desuperheater is a device that sprays aprecisely controlled amount of water into a steamline to modify steam temperature.

� System parameters and required turndownare the most influential parameters indesuperheater selection.

� Desuperheating is done primarily to improveefficiency of thermal transfer devices and toprovide temperature protection for process,product and equipment.

� Another reason to desuperheat is to controlthe “unintentional superheat” created by pressurereduction valves.

� Proper installation is key to bestperformance. Guidelines for piping geometry andplacement of downstream temperature sensorsare available.

� Steam conditioning is the process ofcombining pressure reduction and desuperheatinginto a single control element.

� Turbine bypass systems are beneficial andinstrumental for achieving high efficiency,flexibility, and responsiveness in today’s powerplants.

www.Fisher.com

Chapter 8

Process Overview

The modern pulp and paper mill is a complexmanufacturing process involving many variedtypes of operation. Many factors influence the typeof process used by a specific mill. Some of thesefactors include: type of wood available (hardwoodor softwood), type of paper or paperboardproduced, age of mill, and availability of anabundant supply of water. This sourcebook willfocus primarily upon the Kraft or sulfate pulpingprocess as illustrated in figure 1.

Wood PreparationWood preparation is a series of steps thatconverts logs to a suitable form for use in the pulpmill. This area of the mill is commonly referred toas the woodyard.

Logs from the forest are usually received from atruck, rail car, or barge. Large overhead cranesare used to unload and sort the logs into piles forlong or short logs. Logs may pass through aslasher, which cuts the logs into segments, if acertain length is required.

The next step involves debarking, which removesboth dirt and bark from the logs. The mostcommon method employed is mechanicaldebarking via a barking drum. Logs are fed intothe rotating cylinder and the rotating/tumblingaction rubs the bark from the logs. The bark fallsout of the cylinder via slots and debarked logs exitthe opposite end of the cylinder. Bark is used asfuel for the power boiler or log boiler.

Following debarking, the logs are fed to thechipper. The chipper uses high speed rotatingblades to reduce the logs to chips of a suitablesize for pulping. Chips are then screened foracceptable sizes by passing them over a set of

vibratory screens. The rejects are returned forfurther chipping and acceptable chips are stored inlarge outdoor piles or silos for pulp mill use.

PulpingPulping is the process of separating the woodchips into fibers for paper manufacture. This isaccomplished primarily by mechanical, chemical,or combined mechanical/chemical processes.Some mills that produce various grades of paperhave both mechanical and chemical pulpingprocesses.

Mechanical pulping, or the groundwood process,involves pressing logs against a rotatinggrindstone and washing away the torn fibers withwater. This process is a large consumer of electricpower due mainly to the grindstone motor. Thistype of pulp is used primarily for the production ofnewsprint grade paper.

More modern methods of mechanical pulpinginvolve shredding and grinding of wood chipsbetween the rotating disk of a refiner. The productis referred to as refiner mechanical pulp (RMP).Variations of this process involve pretreating ofwood chips with steam and/or chemicals. This iscommonly referred to as thermo-mechanical pulp(TMP) or chemithermomechanical pulp (CTMP).

The majority of pulping processes in NorthAmerica are chemical processes. The mostcommon are the sulfate and sulfite processes. Ofthese two, the sulfate or Kraft process is thedominant process. The Kraft cooking process ispart of a larger process called the Kraft recoverycycle. A typical Kraft recovery cycle is illustrated infigure 2.

The Kraft process involves cooking the woodchips under pressure in an alkaline solution of

8−2

sodium hydroxide (NaOH) and sodium sulfide(Na2S). This solution, known as white liquor,breaks down the glue-like lignin which binds thecellulose wood fibers together. Cellulose fibers areused to form a paper sheet on the paper machine.

The primary piece of equipment in the pulp mill isthe cooking vessel or digester. The digester is avessel in which wood chips and white liquor aresteam heated to a predetermined pressure andtemperature. The objective is to remove as muchlignin as possible without decreasing fiber(cellulose) strength.

Both batch and continuous processes are used tocook wood chips. The batch process involvesfilling a vessel with wood chips and white liquor.The contents are then heated to a predeterminedcooking temperature and pressure via direct orindirect steam heating. After a prescribed cookingtime, the contents are blown to a holding tank andthe process repeated.

As its name implies, the continuous digester has afairly constant input of chips and outflow of pulpfibers. The chips are usually preheated in asteaming vessel before they are conveyed to thedigester. As the chips move down through thedigester (vertical type), they are successivelyheated, cooked, and washed prior to cooling anddischarge to the blow tank. Indirect steam heatingof the cooking liquor is used to control thetemperature in each section of the digester. Mostof the newer pulp mills have favored thecontinuous digester cooking process over thebatch process.

Following cooking, the pulp from the blow tankmust be washed to remove residual cookingchemicals. This is sometimes referred to as brownstock washing. The resulting process stream fromwashing the stock is referred to as weak blackliquor.

For many years, the standard method of washinghas been a series of rotary vacuum washers. Thepulp and wash filtrate (black liquor) flow in acountercurrent sequence with clean water usedonly for the final washing stage. This allows anincrease in wash solids as it flows toward the firststage washer and a decrease in pulp solids as itmoves to the last stage washer.

As mills face growing economic and environmentalpressures, new methods of washing have beendeveloped. Some of these systems include rotary

pressure filter systems, continuous digesterwashing, and pressure diffusion washers. Each ofthese systems are designed to achieve clean pulpand reclaim cooking chemicals with less washwater.

A final step in the pulping process involvespassing the pulp through knotter and screenequipment. These steps may occur before, after,or split around brown stock washing. The knottersremove uncooked wood chips, knots, and fiberbundles called shives. Screening is removal ofother tramp rejects such as rocks, steel, plastic orconveyor parts.

Recovery of Kraft Pulping LiquorsDue to the high cost of pulping chemicals (sodiumand sulfur), effective recovery following cookinghas an important economic impact on milloperating cost. The primary purpose of the Kraftrecovery cycle (see figure 2) is to reclaim thesechemicals and regenerate them to cooking liquorform. A secondary objective is efficient heatrecovery and steam generation from thecombustion of wood organics in black liquor fuel.This complex process will be covered in threesteps: evaporation, burning, and causticizing orregeneration.

EvaporationWeak black liquor from the brown stock washerscontains spent cooking chemicals, wood organicssuch as lignin, and water. At this stage, the solidscontent is typically 12 - 18%. Before burning,water must be evaporated to raise the solidscontent to 65 - 70%. The bulk of this task iscommonly accomplished by multiple-effectevaporators.

A set of evaporators commonly consists of sixvessels and interconnecting pumps and piping.Steam is used as a heating medium to evaporatewater from the liquor. Steam typically flowscountercurrent to the liquor for maximumeconomy. Since the vessels operate at differentpressures, the vapors from one vessel serve asthe steam supply for the next vessel. Liquortypically leaves the evaporators at 50% solids with5 - 6 lbs. of water evaporated per pound of steamused.

8−3

BurningBlack liquor from the evaporators at 50 - 55%solids cannot be burned in the recovery boiler.Further evaporation to 65 - 70% solids must beattained prior to combustion. This is accomplishedby evaporator-like vessels called concentrators orby direct contact evaporators (cascade or cyclonetype) which use boiler flue gas for evaporation. Ifdirect contact evaporators are used (olderdesigns), air is mixed with the black liquor in theblack liquor oxidation system prior to directheating. This helps prevent the release of odorousgases due to direct heat contact. Most new boilersuse concentrators for final evaporation sinceindirect steam heating emits fewer odors. This iscommonly referred to as low odor design.

The recovery boiler is one of the largest and mostexpensive pieces of hardware in the mill. It is theheart of the chemical recovery process. The heavyblack liquor is sprayed into the furnace forcombustion of organic solids. Heat liberated fromburning serves to produce steam in the watercircuit and reduce sulfur compounds to sulfide.The molten sodium compounds accumulate toform a smelt bed on the furnace floor. The moltensmelt, consisting primarily of sodium carbonate(Na2CO3) and sodium sulfide (Na2S), flows bygravity to the dissolving tank. The dissolving tankis filled with a water solution (weak wash) to coolthe smelt. This solution is called green liquor andis transferred to the causticizing area.

CausticizingGreen liquor is sent to the causticizing area fortransformation to white liquor for cooking. Theprocess begins with clarification of the green liquorto remove impurities called dregs. Clarified greenliquor is then mixed with lime in the slaker to formwhite liquor. The lime (CaO) activates theconversion of Na2CO3 in the green liquor to formsodium hydroxide (NaOH) for white liquor. Toallow time for a complete reaction, the white liquorpasses to a series of agitated tanks calledcausticizers.

A second chemical reaction resulting from theaddition of CaO is the precipitation of lime mud(CaCO3). The lime mud is removed from the whiteliquor by filtration or gravity settling and theclarified white liquor is stored for digester chipcooking.

Lime mud filtered from the white liquor is washedto remove residual cooking chemicals. The wash

water, or weak wash, is sent to the recovery boilerdissolving tank. The washed lime mud is sent tothe lime kiln where heat is added for conversion tolime. This calcined lime, along with purchasedmake-up lime, is used to supply the slaker.

Although many variations exist, this completes atypical Kraft recovery cycle as illustrated in figure2. The next step is the preparation of the pulp forpaper making.

BleachingThe primary objective of bleaching is to achieve awhiter or brighter pulp. If a mill produces brownpaper such as linerboard, a bleaching sequence isnot required. However, if white paper such aswriting or magazine paper is produced, bleachingis required. Bleaching removes the lignin whichremains following digester cooking. Lignin is thesource of color and odor for pulp.

The bleach plant has recently evolved to the mostcontroversial area of pulp and paper productiondue to the formation of dioxin from chlorinebleaching. Environmentalists claim dioxin in pulpmill effluent is contaminating rivers while otherstudies indicate levels of dioxin in effluent are toolow to pose any danger. Nevertheless, manytechnology changes have occurred in the pastdecade that have significantly reduced dioxinemissions.

Bleaching practices prior to 1980 used largeamounts of chlorine to achieve the desired level ofbrightness. Although other stages using sodiumhydroxide, chlorine dioxide, and hypochlorite wereused, chlorine was the prime bleaching agent.Following each stage, washing was required toremove residuals. Large quantities of water wereused following the chlorination stage where themajority of toxic byproducts are formed. Variousmethods of post bleaching treatment of effluentwere used with mixed success.

Beginning in the mid-1980’s, mills began makingsignificant changes due to increasingenvironmental awareness. One change is theincreased use of oxygen (O2) delignification priorto bleaching with chlorine and chlorine dioxide.This provides lignin removal with the benefit ofchlorine-free effluent. This also allows for lesschemical use in subsequent bleaching stages.

A second change is extensive reuse of washerfiltrate to reduce fresh water usage. This reducesthe amount of effluent to be treated prior todischarge from the mill. Some modern plants use

8−4

totally enclosed pressure diffusion washersfollowing O2 delignification to further reduce toxiceffluent.

Another change involves increased substitution ofchlorine dioxide for chlorine gas. Chlorine dioxidedoes not release the chlorine ions responsible forforming dioxin. Although chlorine dioxide is moreexpensive to produce, it requires 2.5 - 3 times lessto bleach the same amount of pulp. Someprocesses which use O2 delignification prior tobleaching have achieved 100%.

Though evolution has caused dioxin emissions todecrease overtime, changes such as this willcontinue to take place in the future. Federal andstate regulator agencies continue to disagree onallowable emission limits. Future technology willcontinue to move toward zero discharge limits fordioxins and other by-products of the bleachingprocess.

Stock PreparationPulp, as produced in the pulp mill, is not suitablefor manufacture of most grades of paper.Properties must be added to the pulp which willaid in uniform sheet distribution and bonding offibers. Two major steps used to impart desiredproperties are stock proportioning and mechanicaltreatment by beating and refining pulp fibers.

Stock proportioning involves the addition of othertypes of pulp and chemical additives to achieve adesired grade of paper. Different pulps, along withwater, are added to achieve proper consistency.Consistency is defined as the percentage byweight of dry pulp fiber in a combination of pulpand water. Typical consistencies in the papermachine range from 1/2 - 3%. “Broke” pulp mayalso be added to the mixture. “Broke” pulpconsists of paper breaks and trim ends from thepaper machine which have been beaten in a brokerepulper.

Various chemical additives are required to aid inproper sheet formation and drainage of water.Some additives and their effects are:

� Starch — improves paper strength andsurface “feel” at the dry-end of the paper machine

� Alum — pH control and chemical retentiononto pulp fibers

� Fillers — common fillers are clay, calciumcarbonate, or titanium dioxide. These particlesserve to fill gaps between fibers to produce asmoother, brighter sheet.

These are a few of the many additives that maybe used. The chemical additives and various pulpare mixed in a blending chest and the mixture iscommonly referred to as furnish.

Another step in pulp treatment involvesmechanical action. The two most commontreatments are referred to as beating and refining.This action tends to separate and shear pulpfibers which increases paper strength and allowsfibers to more easily absorb water and additives.Two basic types of refiners are conical and diskrefiners. Both types consist of rotating elementsand a stationary housing to provide shearingaction. Refining is often done in two stages. Onestage involves treatment of virgin pulp fiber onlyand a second stage for treatment of virgin fiber,broke, and chemical additives.

Finally, the mixture of pulp and chemicals from theblending chest is pumped through refiners to themachine chest. The machine chest is alsosupplied by the save-all. The save-all screensfibers from white water drained from the papersheet into the wire pit.

Paper MachineFollowing stock preparation, the furnish is sent tothe machine room for final sheet forming. Eventhough different types of paper machines exist formanufacture of various grades of paper, most allperform the same basic functions which can bedivided into two broad categories. The “wet end” iswhere the pulp and water solution is spread onto amoving wire and dewatered to form sheet. Thesheet then moves to the “dry end” for furtherevaporation of water and smoothing of the sheet.The type of machine used most today is theFourdrinier paper machine.

The portion of the wet end that supplies stock tothe machine is referred to as the approachsystem. Primary components in this system arethe wire pit, machine chest, fan pump, basisweight valve, screens, and cleaners. The systeminvolves the fan pump accepting a mixture of whitewater from the wire pit and stock from themachine chest. The basis weight valve controlsthe flow of stock from the machine chest to the fanpump. The mixture of stock and white water is

8−5

pumped through screens and cleaners to the headbox. Screen and cleaners remove undesirableparticles such as dirt, grit and clumps of fibers, orchemical additives. The next step involving thehead box and forming wire actually begins theformation of a paper sheet. The head box acceptsstock and white water from the fan pump and isrequired to deliver a uniform flow onto the movingmachine wire. Most modern head boxes arepressurized. Proper control for achieving an evenand uniform outflow is critical for proper sheetformation. The continuous fine screen wireprovides for the formation of a mat of fibers anddrainage of water. Modern wires run at speeds of3000 - 5000 feet per minute. Due to the high wirespeeds, drainage is aided by a vacuum systemfound under the wire screen.

After initial sheet formation and dewatering, thesheet moves to the dry end of the machine. Fromthe machine wire, the sheet is transferred to thepress section. The press section rolls provide amechanical means for water removal and pressurewhich consolidates and smooths the sheet. Thesheet is conveyed through the various presses ona felt or synthetic fabric. The fabric provides fortransfer of pressing forces onto the paper andvolume space for removal of water and air. Apaper sheet leaving the presses is typically at 30 -35% consistency (70% water).

The paper sheet is now transferred to the dryersection where heat is applied to evaporate themoisture content to 5 - 10%. The system consistsof a series of large diameter cylinders that areinternally steam heated. The paper sheet isconveyed by a synthetic fiber over the cylinderswhere moisture is evaporated and carried away bya ventilation system. Condensate formed in thedryer cylinders is removed by siphons andreturned to the powerhouse. Even drying acrossthe entire sheet is a major challenge in thissection.

Following drying, the paper is sent to the calendarwhere large roll presses consolidate the paper toits final thickness and smoothness. The calendarstack consists of hard cylinders capable ofproviding high compression forces. Paper from thecalendar is fed onto spools and rolled into largereels. These reels are then processed to meetcustomer size specifications by the winder and rollfinishing areas prior to shipment.

UtilitiesSo far, attention has focused on processing of theprimary raw ingredient, wood. However, other rawmaterials such as water and electricity, asillustrated in figure 3, play important roles in theproduction of paper.

A paper mill requires large volumes of water foruse throughout the process. Although a fewprocesses can use raw water directly from thesource, most users require a higher quality ofwater. Most water requires treatment in asedimentation basin followed by filtration toremove suspended solids and other impurities.The degree of treatment required depends on thesource of the water such as a river, lake, or well.

Additional treatment for removing dissolvedminerals is required for water used in boilers.Failure to remove these deposits results inbuild-up of sludge and scale which eventuallyleads to operational problems in the boilers. Themost common method employed to remove thedissolved minerals is with ion-exchange resinscalled demineralizers. Demineralized waterproduction has a high capital and operating cost.

Water required for boilers demands specialtreatment. In addition to demineralizers, furthertreatment involving mechanical deaeration andchemical additives is required to remove oxygen.This ultra-pure water is used to produce steam inboth the power and recovery boilers. Steamproduced is used for both process heating andgenerating electricity with steam turbinegenerators. Dual use of fuel energy is calledcogeneration. Since the cost of makingdemineralized water is high, it is important thatclean steam condensate be returned to the boilerfor reuse. A typical return rate is about 50%.

The other basic raw material, electrical power, istypically provided by a combination of own-makeand a tie to the local utility. Own-make electricity isproduced via high pressure superheated steamfrom the power and recovery boilers fed to steamturbines. The turbines extract energy from thesteam which, in turn, drives an electricalgenerator. The power boiler produces steam fromburning wood waste such as bark and issupplemented with coal or oil. In most cases,approximately 185 pound steam is sent to thedigester and turbine while 80 pound steam is usedin the steam room. Additionally, the recoveryboiler burns black liquor as fuel.

Since a mill typically does not produce enoughpower to meet all of its electrical requirements, a

8−6

tie is usually established with the local utility. Thisalso allows the mill to remain in operation if theirelectrical production is curtailed or downcompletely. Most mills try to use as littlepurchased electricity as possible.

Waste TreatmentAn important consideration for modern pulp andpaper mills involves the effective treatment ofwater and air waste streams. Increasedenvironmental awareness has led to stringentemission limits. This aspect of pulp and papermills could be one of the most controversial andcapital intensive areas in the future.

The primary concern for water is treatment ofeffluent which is returned to the source (river orlake). Water used in areas such as the pulp milland bleach plant picks up contaminants whichwould make it harmful for fish and people. Thewaste effluent is typically treated in sedimentation

clarifiers and/or aeration lagoons to removecontaminants. Although some methods are highlyeffective, future trends will be toward closedsystems with no effluent waste stream.

Air pollution from pulp and paper mills involvesboth particulate and odor emissions. The majorsources of particulate emissions involve powerand recovery boilers. Fine particulate from varioussodium compounds are emitted from recoveryboilers and coarse particulate from burning woodwaste in the power boiler. Treatment forparticulate typically involves collection devicessuch as scrubbers or electrostatic precipitators.

Although odor emissions in general are notdangerous to the public, resentment due to thesmell requires pollution control application. Thevarious sulfur gases causing the odor are referredto a TRS (Total Reduced Sulfur). Since odorpollution is difficult to treat, in-process methodsresulting in less generation of odorous gases ispreferred. However, absorption of gases with wetscrubbers is often used to achieve finalabatement.

8−7

C0810 / IL

Figure 8-1. Kraft Pulp and Paper Mill Process Overview

8−8

C0811 / IL

Figure 8-2. Kraft Recovery Cycle

8−9

C0812 / IL

Figure 8-3. Utilities

8−10

www.Fisher.com

Chapter 9

Pulping

Pulping is the process of converting wood materialto separate pulp fibers for paper making.Processes range from purely mechanical, in whichthe wood is ground into fibers by refiners orgrindstones, to chemical processes, in which thefibers are separated by chemically degrading anddissolving the lignin that binds fibers together. Inmany cases, mills will produce various grades ofpaper having both mechanical and chemicalpulping processes.

Mechanical Pulping

Stone Groundwood

Process:The most basic type of mechanical pulping isknown as stone groundwood (SGW), and hasbeen virtually unchanged since its development inthe 1840s. This process involves rotatingmanufactured grindstones to be pressed againstsmall wood logs that are oriented parallel to theaxis of the stone where a typical modern SGWplant will consist of only four to six grinders tosupply a large paper machine.

The quality of the produced pulp (strength anddrainage properties) depends primarily on thesurface characteristics of the stone. Water isadded to wash away the torn fibers. Virtually allstones are artificially manufactured using a hardgrit material, typically embedded with siliconcarbide or aluminum oxide.

This process is a large consumer of electric powerdue to the rotating grindstone, and the pulp

produced is typically used for the production ofnewsprint grade paper.

Refiner Mechanical Pulp

Process:Commercial production of refiner mechanical pulp(RMP) began in 1960. It is produced in mostmodern mills using chips rather than logs andrigged metal discs used for shredding and grindingof the wood chips. The chips are ground betweenthe rotating discs in a refiner, producing RMP.

This process is typically done in two separatestages operating in series, and produces alonger-fibered pulp than SGW. As a result, thepulp is stronger, freer, bulkier, and usuallysomewhat darker in color.

Thermomechanical Pulping

Process:The first major modification to RMP was theaddition of steam before the refiner. This is knownas thermomechanical pulping (TMP). Thesteaming serves to soften the chips, resulting inthe pulp having longer fibers and fewer shivesthan RMP. These longer fibers produce a strongerpulp than either SGW or RMP, making for astronger final sheet of paper. This process is stillemployed on a large scale to produce high-tierpulps for newsprint and board.

Referring to figure 1, chips are fed by a feedingplug and screw feeder to a presteamer, which isheated by PCV-1 to typically 15 to 30 psig and265 to 285°F. After a retention time of a couple ofminutes, the pressurized chips are fed to the

9−2

E1387

PRESTEAMER

CHIPS

FRESH STEAM

PCV-2

TMP PULP

PCV-3REFINER

CYCLONE

STEAM TOSTACK

PCV-1

PCV-4

PROCESSSTEAM

STEAM TORECOVERY

Figure 9-1. Thermomechanical Pulping Process

refiner. The refiner may then be fed with freshsteam via PCV-4, during startup, to increase thepressure to 60-75 psig or 300°F.

The refiner discharges the pulp and steam to thecyclone, which separates the steam from the pulp.The PCV-1 and PCV-2 valves control the pressurein the refiner. During production, this steam is sentto heat recovery, while during startup it goes to thesteam stack for disposal. The TMP pulp(approximately 35% solids) is discharged throughvalve PCV-3 from the first stage refiner to thesecond stage, and from there, to further treatmentin the screening and cleaning stages.

Valve Selection:

The control of clean steam from PCV-4 can beeasily accomplished by the Fisher Vee-Ballsegmented ball valve. Where fibers can build upand result in potential plugging problems, namelyPCV-1 and PCV-2, the Vee-Ball has provensuccessful with its V-notch ball, as this shearsthrough any pulp fibers. However, any use of theVee-Ball attenuator must be evaluated with care.The Fisher Control-Disk can also be used in thisservice.

The TMP discharge, or blow valve, PCV-3contains pulp at 35% consistency, andsteam/condensate. Because of the high pressuredrop of the system, this valve must withstand

erosion. One solution is the Vee-Ball with stellitedinternals and trim, including the ball seal. Thewater control valve should be a Vee-Ball to ensureoptimal control.

Valve SelectionTag Application Recommended Alternate

PCV-1,PCV-1

Steam(process)

Vee-Ballsegmented ball

Control-Disk

PCV-3 Refiner Blow Vee-Ball segmentedball w/ stellited trim

PCV-4 Steam (fresh) Vee-Ballsegmented ball

Control-Disk

- - - Water (control) Vee-Ballsegmented ball

Control-Disk

Chemithermomechanical Pulp

Process:

Wood chips can be pretreated with sodiumcarbonate, sodium hydroxide, sodium sulfide, orother chemicals prior to refining with equipmentsimilar to a mechanical mill. The conditions of thechemical treatment are much less vigorous than ina chemical pulping process since the goal is tomake the fibers easier to refine rather thanremoving the lignin as in a fully chemical process(described later in this section). Pulp made usingthese hybrid processes are known aschemithermomechanical pulp (CTMP).

9−3

Chemical PulpingChemical pulp is produced by combining woodchips and chemicals in large pressure vesselsknown as digesters (see chapter 3) where heatand the chemicals break down the lignin, whichbinds the cellulose fibers together, withoutseriously degrading the cellulose fibers. Chemicalpulp is used for materials that need to be strongeror combined with mechanical pulps to give aproduct with different characteristics.

Sulfite

Process:The sulfite process produces wood pulp, which isalmost pure cellulose fibers, by using various saltsof sulfurous acid to extract the lignin from woodchips in digesters. This process is used to makefine paper, tissue, glassine, and to add strength tonewsprint. The yield of pulp is higher than Kraftpulping as the process does not degrade lignin tothe same extent as the Kraft process, and sulfiteis easier to bleach.

Sulfite pulping is carried out between a pH of 1.5and 5, depending upon the counterion to sulfiteand the ratio of base to sulfurous acid. The pulp isin contact with the pulping chemicals for four tofourteen hours, and at temperatures ranging from265°F to 320°F, again depending upon thechemicals used.

The pulping liquor for most sulfite mills is made byburning sulfur with the correct amount of oxygento give sulfur dioxide (SO2), which is thenabsorbed into water to give sulfurous acid(H2SO3).

S + O2 → SO2 SO2 + H2O ⇔ H2SO3

Care must be given to avoid the formation of sulfurtrioxide (SO3) as this produces sulfuric acid(H2SO4) when it is dissolved in water. Thispromotes the hydrolysis of cellulose withoutcontributing to delignification (removal of lignin),and ultimately damages the cellulose fibers. Thisis one of the largest drawbacks of the sulfiteprocess, and leads the pulp fibers not being asstrong as Kraft pulp fibers.

The cooking liquor is prepared by adding thecounter ions, such as hydroxide or carbonatesalts. The relative amounts of each species

present in the liquid depends largely on theamounts of sulfurous acid used.

For monovalent hydroxides (Na+, K+, and NH4+),

MOH:

H2SO3 + MOH → MHSO3 + H2O MHSO3 + MOH → M2SO3 + H2O

For divalent carbonates (Ca2+, Mg2+), MCO3:

MCO3 + 2H2SO3 → M(HSO3)2 + CO2 + H2O M(HSO3)2 + MCO3 → 2 MSO3 + CO2 + H2O

The spent cooking liquor from the process isknown as brown or red liquor. Pulp washers, usingcountercurrent flow, remove the spent cookingchemicals and degraded lignin and hemicellulose.The extracted brown liquor is then concentrated inmultiple effect evaporators. The concentratedbrown liquor can be burned in the recovery boilerto generate steam and recover the inorganicchemicals for reuse in the pulping process, or itcan be neutralized to recover the usefulbyproducts of pulping.

The most common recovery process used ismagnesium-based sulfite pulping, called the“Magnefite” process. The concentrated brownliquor is burned in the recovery boiler, producingmagnesium oxide (MgO) and sulfur dioxide, bothof which are recovered from the flue gasescreated by the burning of the brown liquor.Magnesium oxide is recovered in a wet scrubberto give a slurry of magnesium hydroxide(Mg(OH)2).

MgO + H2O → Mg(OH)2

This magnesium hydroxide slurry is then used inanother scrubber to absorb sulfur dioxide from theflue gases, producing a magnesium bisulfite(Mg(HSO)3) solution that is clarified, filtered, andused again as the pulping liquor.

Mg(OH)2 + 2 SO2 → Mg(HSO3)2

Sulfate (Kraft)

Process:The sulfate or Kraft process is the dominantchemical process used by pulp mills today. TheKraft process involves cooking the wood chipsunder pressure in an alkaline solution of sodiumhydroxide (NaOH) and sodium sulfide (Na2S). Thissolution breaks down the glue-like lignin which

9−4

binds the cellulose wood fibers together. Thisprocess produces stronger pulp than the otherprocesses, but is darker in color than the otherpulp processes. However, the benefit to thisprocess is the wide range of fiber sources that canbe used, and the regeneration of cooking liquors.

Woodchips are fed into digesters where they areimpregnated with the cooking liquors of warmblack liquor and white liquor (see chapter 3). Thewarm black liquor is spent cooking liquor comingfrom the blowtank. White liquor is a mixture ofsodium hydroxide and sodium sulfide, produced inthe Kraft recovery process. Delignification requiresseveral hours of cooking at 265°F to 355°F. Underthese conditions, the lignin and somehemicellulose degrade to give fragments solubilityin a strongly basic liquid.

White Liquor:

NaOH ⇔ Na+ + OH− Na2S + H2O ⇔ 2Na+ + OH− + HS−

The lignin is removed by the following reaction,where the HS− ion the component that ultimatelyremoves the lignin.

E1388

The finished cooked wood chips are blown fromthe digester, and the action of the cooked woodchips hitting the walls of the blowtank produceindividual pulp fibers. The solid pulp (about 50%by weight based on dry wood chips) is thencollected and washed. The washing stagesseparate the cooking liquors from the cellulosefibers, where the pulp is brown after cooking andis known as brown stock.

The combined liquids, known as black liquor dueto its color, contains lignin fragments,carbohydrates from the breakdown ofhemicellulose, sodium carbonate, sodium sulfate,and other inorganic salts.

Recovery Process:The excess black liquor is concentrated in amultiple effect evaporator (see chapter 4) intoheavy black liquor, and burned in the recoveryboiler (see chapter 5) to recover the inorganicchemicals for reuse in the pulping process. Moreconcentrated black liquor increases the energyand chemical efficiency of the recovery cycle. Thecombustion is carried out such that sodium sulfate(Na2SO4) is reduced to sodium sulfide by theorganic carbon in the mixture:

1. Na2SO4 + 2 C → Na2S + 2 CO2

The molten salts (smelt) from the recovery boilerare dissolved in process water known as weakwash (see chapter 6). The solution of Na2CO3 andsodium sulfide results in green liquor. This liquid ismixed with calcium hydroxide (Ca(OH)2) toregenerate the white liquor in the pulping process.

2. Na2S + Na2CO3 + Ca(OH)2 ⇔ Na2S + 2 NaOH+ CaCO3

Calcium carbonate (CaCO3) precipitates from thewhite liquor and is recovered and heated in a limekiln where it is converted to calcium oxide (lime).

3. CaCO3 ⇔ CaO + CO2

Lime is reacted with water to regenerate thecalcium hydroxide used in reaction 2.

4. CaO + H2O → Ca(OH)2

The combination of reaction 1 through 4 forms aclosed cycle with respect to sodium, sulfur, andcalcium. The recausticizing process where sodiumcarbonate is reacted to regenerate sodiumhydroxide is the main reaction in the processwhere approximately 98% of the originalchemicals are regenerated.

www.Fisher.com

Chapter 10A

Digesters

Batch Digesters – Kraft PulpingKraft batch digesters have been produced inseveral different configurations, including rotating,horizontal, and spherical vessels. By far the mostprevalent configuration is the upright, cylindricalbatch digester.

Typically a batch digester is two to three and ahalf stories tall and has 2500 to 7000 cubic feet ofcapacity. The quantity of pulp produced per batchranges from five to 25 tons.

Wood chips, chemicals, and steam are combined,coked under pressure to a schedule, and thendumped to a blow tank on a batch basis. Mills withbatch digesters have been between four and 36units. Some mills have both batch and continuousdigesters.

There are two methods of heating batch digesters:

� Directly steam batch digesters (figure10A-1): These units are the least complicatedand are usually of older design. Steam at 50 to150 psi is injected at the base of the digester intodirect contact with the wood chips and cookingliquor.

� Indirect steam batch digesters (figure10A-2): Cooking liquor is extracted from thedigester through a screen to prevent removal ofwood chips or pulp. The liquor is passed throughan indirect heat exchanger and then recirculatedto the top and bottom of the digester. Chippacking, air evacuation systems, and presteamingare incorporated with indirect heating to produce amore modern batch digester design. A relativelynew derivation of the modern, indirectly steamedbatch digester is the “low energy process.” Thelow energy process batch digester is covered laterin the chapter.

A drawback to the directly steamed digester is thedilution effect from condensed steam. Indirectlysteamed digesters; however, require moremaintenance due to the screens, pumps, andexternal heat exchangers. Regardless of thedigester steaming method, the process objectiveis the same: to elevate the temperature andpressure of the chip-liquor mass such that thealkaline component in the cooking liquor candissolve the desired amount of lignin andextractives from the cellulose fiber.

Batch Digester Process ParametersThe Kraft pulping process is also known as thesulfate or alkaline process. The actual cookingliquor is a mixture of white liquor from thechemical recovery boiler and recausticizingoperations, and black liquor (spent white cookingliquor), which has been separated from previousbatches by brown stock washers. The mainconstituents in the white liquor that contribute todissolving away the lignin binder material areNaOH and Na2S. The Kraft cooking liquor isalkaline, or basic, with a starting pH above 13units. The temperature of the cooking liquor, whenadded to the digester, is typically 160 to 190°F.The temperature of the wood chips is typically 60 -80°F, but may be much colder in northernclimates. The following paragraphs describe thevarious process parameters associated with theKraft batch pulping process.

Chemical Concentration of CookingLiquorsA key process parameter in the production of Kraftpulps is the chemical strength of the white cookingliquor being added to the wood chips in a digester.To achieve a target pulp yield at a target K (orKappa) number, a specific quantity of white andblack cooking liquors must be added per unit of

10A −2

Figure 10A-1. Directly Steamed Batch Digester

10A −3

Figure 10A-2. Indirectly Steamed Batch Digester

10A −4

dry wood. However, these total liquor-to-wood andliquor-strength-to-wood ratios are difficult toenforce because on-line measurement of chipmoisture and weight have proven unreliable, andthe chemical strength of the white liquor solution isnot a stable or directly measurable variable.Sensor developments in both areas are rapidlyadvancing.

Cooking TimeA digester begins to cook slowly as soon as thewhite liquor and black liquor solutions are appliedto the wood chips (even at atmospheric pressure,i.e., before capping). Typically, a digester cookingcycle is as follows: the digester is capped,steaming is begun, and the temperature andpressure ramp up to a predetermined pressure (ortemperature). Further steaming is then regulatedto maintain the desired target pressure (ortemperature). When the target “H” factor has beenreached, the digester’s contents are blown to theblow tank.

The total length of the cooking cycle per batchdigester will depend on the desired pulp grade andthe mill’s criteria of operations. Consequently, abatch cooking cycle can range from two hours“cap-to-cap” for a hard cook (high yield), to fivehours for a soft cook (low yield).

Cooking TemperatureThe batch digester temperature is also asignificant factor in achieving cooking uniformity(delignification) throughout the mass of chips.Higher temperatures accelerate the rate ofchemical reaction between the wood chips and thecooking liquors. The quantity of rejects (i.e., knotsor partially cooked chips) in a batch is related touneven temperature distribution in the digester.For example, if poor convection mixing causestemperature differences between the bottom andtop, it is not uncommon for pulp at the bottom of adigester to be several Kappa units different frompulp at the top of a digester.

Typically, a charged batch digester at atmosphericpressure is at around 165°F to 195°F. Uponcapping and steaming, the maximum desiredcooking temperatures will range from 330°F to350°F at pressures of 100 to 120 psi. Normally,there is at least a top and a bottom temperaturesensor on each digester. A middle temperatureprobe is encouraged for improved indication oftemperature distribution. The bottom temperature

usually exceeds the top temperature due tohydrostatic liquor head on directly steamed units,because the sensor is closer to the entry point ofsteam.

Cooking PressureDigester pressure rises as the steam flow to thedigester raises the temperature of the chip andliquor mass. Batch temperature is considered tobe the key variable, but batch pressure is aneasier and faster variable to measure than a“representative” chip and liquor mass temperature.The pressure/temperature relationship is based onsaturated steam tables. The implied digestertemperature should include a slight increment forthe boiling point rise of the cooking liquor. Theelevated boiling point over water is due to organicand inorganic solids in the liquor. Digesterpressure ranges from atmospheric at liquorcharging to a maximum of 100 to 120 psi for theextended cooking period.

Digester pressure causes the cooking liquor tomore readily impregnate the wood chips so thatthe delignification reactions proceed from theinside of the chip to the outside of the chip, as wellas vice-versa. Pressure in a batch digester rangesfrom atmospheric at liquor changing to amaximum of 100-120 psi for the extended cookingperiod.

Pressure ProfileThe batch digester cooking cycle is usuallyrepresented in text books by a graph of theinternal digester pressure vs. time, such as thatshown in figure 10A-3. In real life, the cookingcycle pressure profile is never this rigid. The timeinterval at each different phase of the cook canvary significantly from one grade of pulp toanother and from mill to mill. Figure 10A-4 showsa more realistic representation of the pressureprofile over the entire cooking cycle. Figure 10A-5shows the steam demand profile required tocomplete this representative cooking cycle. Thevertical axis defining the amount of steamdemanded is not labeled because different pulpgrades or different sized digesters will requiredifferent quantities of steam. However, a typical3-hour cooking cycle at 100 psig consumes from4000 to 6500 pounds of steam per ton of pulpproduced. Therefore, a 7-ton digester couldconsume about 40,000 pounds of steam percooking cycle.

10A −5

Figure 10A-5. Steam Demand Profile

Figure 10A-4. Actual Batch Cooking Cycle

Figure 10A-3. Theoretical Batch Cooking Cycle

10A −6

False PressureA batch digester is basically a large pressurecooker. As steam is applied to the mass of chipsand liquor, a quantity of resinous vapors aredistilled off. These vapors, along with air initiallyentrained with the chips and a small quantity ofnon-condensed steam, migrate to the top of thedigester. These vapors and gases aresystematically drawn off through the digester relief(gas off) piping. The vapors are a major source ofthe distinctive Kraft mill odor.

If the non-condensable portions of these gasesare not relieved from the digester they wouldeventually accumulate sufficiently in the top of thedigester to indicate a falsely high pressure relativeto the steam saturation temperature. Under suchconditions, the correct control action would be toreduce the steam flow to the digester. The cookwould then take place at a steam saturationtemperature corresponding to say, 102 psiginstead of the 100 psig target. The resulting batchwould be very undercooked and possibly ruined.Therefore, the non-condensable portion of therelieved gases must be removed from the digesterin order to maintain the correcttemperature-pressure relationship.

OverpressureOverpressure of a digester means that the actualdigester pressure is above the desired target.Overpressure may result from trying to maintainthe proper temperature while false pressure exists(overshooting), or via exothermic reactions oncethe target pressure has been reached. Forinsurance and safety purposes, each digester willhave an upper pressure limit rating. Overpressureexposes production personnel to a hazardousenvironment and is a contributor to off-qualitypulp.

Relief or Gas OffThe previous discussions of digester overpressureand digester false pressure outlined the necessityfor relieving excess gases from a digester. Figure10A-1 shows a typical relief piping arrangement.When the gas off valve is open, the blow backvalve must be closed. This interlock must exist forboth safety and economic reasons.

In general, a relief line is connected to the neck ofthe digester through a relief screen. The screenprevents large quantities of cellulose fibers or

chips from entering the relief line piping system.Liquor, being a fluid, can readily pass through thescreen, but it is undesirable to allow any chemicalloss to occur. All relief gases and liquids pass fromtheir respective digesters into a common header.This header directs all such materials to a centralseparating device which separatesnon-condensable gases from condensable gasesand liquor, pulp, etc. The non-condensable gasesare quite odorous. These gases are usuallyscrubbed and/or burned in a lime kiln. Thecondensable gases, however, can contain, inaddition to steam, a significant quantify of avaluable byproduct, i.e., crude turpentine. Three tofour gallons of crude turpentine can be distilled offand recovered per ton of resinous southern pinepulp produced.

Blow BackBlow back is basically a short reversal of steamflow through the gas off line to the top of thedigester for the purpose of cleaning the reliefscreen and/or collapsing the steam bubble, whichmay form within the chip mass in the lower area ofthe digester. The sequence is typically: (1) shut offthe high pressure steam valve to the base of thedigester, (2) shut off the gas-off valve to thecommon header, and (3) open the blow back valve(see figures 10A-1 and 10A-2). This sequenceallows high pressure steam to be briefly injectedthrough the relief lines and into the top of thedigester. The surge of high pressure steam blowsthe screen clean while the increased pressurefrom the top forces the chip mass down,collapsing the bubble. A blow back typically lastsonly 10-30 seconds, and then the valves revert tothe original status.

Blow TankWhen a digester cooking cycle is completed, theblow valve is opened to connect the digester witha common blow tank. The blow tank is a lowpressure receiving vessel, which is usuallycapable of holding several blows. Several batchdigesters producing the same grade of pulp willdischarge into the same blow tank. The highpressure in the digester will blow the entire massof chips and black liquor into the blow tank.

Typically, a blow tank is equipped with bothvacuum and pressure relief valving systemsbecause 100 to 120 psi is released when the blowvalve is opened. A significant quantity of vapor willbe flashed off the pulp and liquor as it enters theblow tank. A blow tank is not designed to

10A −7

withstand digester-like pressures and so the reliefvalve will pop if the outlet for gas-off is pluggedwith pulp. Similarly, flashed vapor to a condensingdevice can create sufficient vacuum to collapse ablow tank. The vacuum relief valve provides amargin of safety against such an occurrence.From the blow tank the brown stock is pumped towashing and screening stages.

Control Valve Selection

Digester Capping ValveThe chips are conveyed to the chip chute which ismounted directly to the capping valve (see figures10A-1 and 10A-2). One of the most importantvalves, this valve is used to automate the chipfilling operation. This is an erosive service as thechips impinge on the sides of the body and ball, sohardened materials and trim must be used. Inaddition, tight shutoff is necessary to ensure theappropriate pressure can be reached within thedigester for chip cooking.

General Service Valves

Refer to figure 10A-1.PROCESS FISHER CONTROL VALVE PRODUCT DESIGN

ValveTag #

DIRECT STEAMEDBATCH DIGESTER

V150 V300 V500 CV500 ED/ETTypical

Valve SizeApplicationDescription

ControlFunction

HV-1 Liquor fill O/O P 10’’

FV-1 White liquor to digester T S S P 8’’

FV-2 Black liquor to digester T S S P 8’’

HV-2 Blow back steam valve O/O P 2’’

PV-1 Gas off T P 3’’

TV-1 Cooking valve T S P 8’’

Refer to figure 10A-2.PROCESS FISHER CONTROL VALVE PRODUCT DESIGN

ValveTag #

INDIRECTLY STEAMEDBATCH DIGESTER

V150 V300 V500 CV500 ED/ETTypical

Valve SizeApplicationDescription

ControlFunction

HV-1 Liquor fill O/O S P 10’’

FV-1 White liquor to digester T S P 8’’

FV-2 Black liquor to digester T S P 8’’

HV-2 Blow back valve O/O P 2’’

PV-1 Gas off T P 3’’

TV-1 Indirect steam valve T S P 3’’

FV-3 Digester top recirculation T P 8’’

FV-4 Digester bottom recirculation T P 8’’

TV-2 Condensate return T P S 3’’

TV-3 Direct steam valve T S P 6’’

CODE:P = Primary selection, S = Secondary selection, T = Throttling, O/O = On/Off

10A −8

Batch Digester – Low EnergyProcessMuch has been written about the relatively newlow energy cooking process. Studies indicateimproved pulping properties and operatingefficiencies over conventional batch digesters.Some of the reported benefits of the low energyprocess are:

� Significant steam savings

� Reduced evaporator load

� Lower black liquor viscosities

� Lower alkali consumption

� Fewer brown stock washing stages

� Stronger pulp

� Lower environmental impact

The ProcessThe main difference between the low energycooking process and a conventional batch digesteris the tank farm associated with the heat recoverysystem. Accumulator tanks are added with each“stage” of process. A typical three-stage design isshown in the process schematic (see figure10A-6). Actual mill installations have duplicate setsof pumps, valves and piping to handle odd andeven digesters. Also there are typically sets of “A”,“B”, and “C” accumulator tanks. This arrangementallows for operations flexibility.

The function of the tank farm and management ofthe transfer of liquors is key to understanding thelow energy process and cooking cycle. First,empty digesters are filled with wood chips. Ifdesired, packing of the chip bed can beaccomplished with steam or liquor. Steamprovides higher compaction. Compaction of thebend increases capacity of the cook.

Cool black liquor from the atmospheric “A” tank isadded to provide a liquor pad in the bottom of thedigester. Warm black liquor from the pressurized“B” tank is then pumped into the digesterdisplacing entrained air. The discharge valves arethen closed. The warm liquor pump brings thedigester up to pressure by pre-impregnating thechips and hydraulically filling the vessel.

The chips are further heated by pumping both hotblack and hot white liquor into the digester. Thewhite liquor is preheated through an indirect heatexchanger between the “C” and “B” tanks usinghot black liquor as the heat source. The hot whiteliquor is then stored in a pressurized accumulatortank for delivery to the digesters. Warm blackliquor is displaced to the “A” tank where soap isskimmed and excess liquor is sent to the weakliquor filters.

After completion of the hot liquor fill operation, thepulp mass is generally close to the required cooktemperature and pressure. If necessary, furtherheating is done by an external liquor heater.

As the pulp is cooked, resinous vapors are givenoff. These vapors, along with any remainingentrained air, migrate to the top of the digester.These gases are systematically drawn off throughthe digester relief valve. If these gases were notdrawn off a false pressure relative to the steamsaturated temperature would be indicated. Undersuch conditions the control action might be toreduce steam to the digester and thus undercookthe pulp. From time to time between relief cycles,steam is blown back through the relief line to cleandebris off the relief screens.

Once the proper degree of cooking (H-factor) hasbeen reached, cooking is stopped by pumpingwasher filtrate into the bottom of the digester.Most of the cooking liquor remains hot and isdisplaced to the “C” tank, ready to use for the nextcook. The cooler liquor goes to “B” and “A” tanks.As a result, the pulp in the digester is washed andcooled below flash point at atmosphericconditions. This in-digester washing reduces loadon both the brown stock washers and evaporators.The pulp is then transferred to the blow tank bycold-blowing the digester with compressed air. Ahigh pressure air receiver is used for this airsupply. Since the pulp is blown cool a number ofbenefits arise including improved pulp quality andlower emissions of total reduced sulfur.

10A −9

Figure 10A-6. Typical Three-Stage Design

10A −10

Control Valve Selection

General Service Valves

Refer to figure 10A-6.PROCESS FISHER CONTROL VALVE PRODUCT DESIGN

ValveTag #

WARM BLACK LIQUORACCUMULATOR

V150 V200 V300 V500 CV500 A81Control-

Disk

TypicalValveSize

ApplicationDescription

ControlFunction

HV−4 Warm liquor to warm fill pump O/O P S 14’’

TV−2 Mill water temperature controlvalve through cooler

T P S 6’’

HV−5 Digester liquor return header O/O P S 12’’

HV−6 Digester liquor return header O/O P S 12’’

FV−2 Warm liquor flow throughliquor cooler

T P S 4’’

FV−6 Warm fill control valve T P S 10’’

Refer to figure 10A-6.PROCESS FISHER CONTROL VALVE PRODUCT DESIGN

ValveTag #

WARM BLACK LIQUORACCUMULATOR AND DISPLACEMENT

TANK V150 V200 V300 V500 CV500 A81 Control-Disk

ED/ETTypicalValveSizeApplication

DescriptionControl

Function

HV−1 Digester liquor return header O/O P S 12’’

HV−2 Displacement tank bypass O/O P S 12’’

LV−1 Cool liquor level control valveto liquor filter

T P S 10’’

TV−1 Warm liquor to cool-tempcontrol

T S P 2’’

HV−3 Cool liquor pad to warm fillpump

O/O P S 12’’

LV−2D Brown stock filtrate levelcontrol

T P S 12’’

Refer to figure 10A-6.PROCESS FISHER CONTROL VALVE PRODUCT DESIGN

ValveTag #

HOT BLACK LIQUOR ACCUMULATORAND WHITE LIQUOR ACCUMULATOR

V150 V200 V300 V500 CV500 A81Control-

Disk ED/ETTypicalValveSize

ApplicationDescription

ControlFunction

HV−7 Digester liquor return header O/O P S 10“

HV−8 Displacement liquor returnheader

O/O P S 10’’

FV−3 Cool white liquor to heatexchanger

T P S 6’’

FV−4 Hot liquor to hot fill pump T P S 8’’

FV−5 Hot white liquor to hot fillpump

T P S 10’’

10A −11

Refer to figure 10A-6.PROCESS FISHER CONTROL VALVE PRODUCT DESIGN

ValveTag #

DIGESTERV150 V200 V300 V500 CV500 A81 Control-

DiskED/ET

TypicalValveSize

ApplicationDescription

ControlFunction

HV−9 Digester hot header return O/O P S 10’’

HV−10 Digester warm header return O/O P S 10’’

PV−1 Digester main pressurecontrol valve

P 12’’

HV−11 Air to receiver to digester O/O P S 6’’

HV−12 Digester air evacuation O/O P S 12’’

FV−7 Relief to blow tank P 9’’

HV−13 Digester steam packer valve P 8’’

PV−2 Digester relief to hotaccumulator

P 3’’

HV−14 Digester relief screen blowback

P 3’’

HV−15 Digester top recirculation O/O P 12’’

TV−3 Digester sparger steam valve P 3’’

HV−16 Digester bottom recirculation O/O P 12’’

HV−17 Digester cone flush dilution O/O P 9’’

HV−18 Digester displacement fill O/O P 10’’

HV−19 Digester warm fill inlet O/O P 12’’

HV−20 Digester hot fill inlet O/O P 10’’

CODE:P=Primary Selection, S=Secondary selection, T=Throttling, O/O−=On/Off

10A −12

www.Fisher.com

Chapter 10B

Kamyr� Continuous Digesters

Kamyr� continuous digesters vary dependingupon the type of raw material, end product, thecost of chemicals, steam or power, as well asgeographical location of the mill. There are fivebasic configurations:

� Single vessel hydraulic digester (the originaldesign).

� Two vessel hydraulic digester (with separatehigh pressure impregnation vessel).

� Single vessel steam/liquor phase digester(with high pressure feeding system). Developedprimarily for sulfite cooking, it has also been usedfor Kraft, pre-hydrolysis Kraft, and neutral sulfitesemi-chemical pulp.

� Two vessel steam/liquor phase digester (withseparate high pressure impregnation vessel)

� Steam/Liquor Phase Digester (with asthmafeeder) used for pulping sawdust, shavings ornon-wood fibers such as straw, bamboo, jute, etc.

The single vessel hydraulic digester was theoriginal Kamyr digester. It is so named becausethe digester consists of a single vessel, which isoperated completely full of liquor (no vapor beingpresent). Digester heating is accomplished byheating the cooking liquor indirectly in heatexchangers using 150 to 175 psi steam andcirculating this heated cooking liquor through thedownward flowing chips in the digester vessel.

Over the years, this hydraulic digester system hasconstantly improved. The modern single vesselhydraulic digester has an improved steaming andfeeding system, which improves its tolerance ofdry and degraded chips. The combinationimproved cooking circulations and steps in thevessel diameter have made it possible to produce

a more uniform pulp, with a poor quality of chipfurnish.

The main features of a hydraulic digester are:

� Metering and steaming of the wood chips.

� Feeding chips to the digester via the highpressure feeder and top circulation loop.

� High pressure impregnation of the chips withcooking liquor.

� Heating to cooking temperature throughliquor circulation, using indirect heating.

� Hi-heat-in-digester washing followed by adiffusion washer.

The hydraulic type digester with its longimpregnation and cooking times produces a veryuniform, high quality, strong pulp and is suitablefor the production of either liner grade orbleachable grade pulps.

The two vessel hydraulic digester was originallydesigned to provide optimum pulp quality in verylarge digester systems (above 1200 TPD), but hasbeen used for smaller tonnages.

For large tonnages, the digester vessel becomeslarge in diameter. This, in turn, makes it moredifficult to ensure uniform circulation in the normalcooking liquor heating system.

In order to ensure uniform heating of the largerchip mass, an impregnation vessel is added toheat the chips to cooking temperature before theyenter the second vessel (digester). This allows forall of the chips to be heated to exactly the samecooking temperature, helping to optimize pulpuniformity.

This digester is essentially unaffected by varyingchip furnish, as there is a homogeneous chip

10B −2

mass throughout the chip column which explains agrowing use of the two vessel digester system.

Process

Regardless of the type, a Kamyr continuousdigester combines the digesting and washingprocess into one vessel so that only one washingstage is required following the blow tank. Chipsfrom the wood yard are fed to a surge bin (chipbin) in the digester building. They are thenmetered continuously through a chip meter to alow pressure feeder. The chips then fall into thesteaming vessel, and are conveyed to the chipchute and the high pressure feeder. Chips arethen sluiced to the top of the digester and the topseparator at a predetermined temperature andpressure. Cooking liquor is also fed continuouslyinto the top of the digester in the desired ratio tothe wood chips. The chips move slowly by gravityto the bottom where they are discharged as pulp.Along the way, they are heated to simulate theheating of a batch digester and its contents. Thetemperature is varied in the mid-section, orcooking zone, to suit different production rates.The bottom digester section is used as a washerand, at this point, any similarity between batch andcontinuous cooking ends.

With a vacuum drum washer, liquor is drainedfrom the pulp and replaced by water. In the Kamyrdigester wash system, the liquor is displaced byintroducing very weak black liquor from thevacuum washer, diffuser or even warm water atthe bottom of the digester, making it flowcounter-currently to the pulp mass which ismoving to the bottom. This is known as “hi-heat”diffusion washing. The up-flowing weak blackliquor wash displaces the stronger residual blackliquor, which is drawn off through a screen in thewall of the digester located about half-way up fromthe bottom, but below the cooking zone. Thisallows the pulp to be effectively washed before itis blown to the blow tank.

Although all components of the Kamyr digesterare closely interrelated, the following is a logicalstep by step explanation of controlled loops, whichwill provide an understanding of the various stagesof the Kamyr system.

PressurizingWeak black liquor (filtrate) or warm water,approximately 170�F, is pumped to the digesterthrough the cold blow pump. There, it is used tomaintain digester pressure. The digester pressurecontrol system has three major components:

� An automatic pressure control valve on aninput liquor line.

� An automatic pressure relief control valve onthe digester relief line.

� A pressure pump kick-out switch on thedigester.

The digester pressure can be controlled by movingthe set point on the input control valve to 165 psig(normal digester pressure). A rise in the set pointwould then cause the valve to open to admit moreliquor to the digester and, therefore, increase thepressure. The reverse occurs when the pressureset point is lowered. This is a rapid responsecontroller.

The secondary pressure control element consistsof the automatic relief valve, which bleeds liquorfrom the lower cooking zone header to the No. 2flash tank. The relief valve is set slightly higher (15psig) than the input valve and bleeds off liquorwhen the pressure exceeds this set point. Therelief valve should always be in a closed positionunder normal conditions. It is only used as apressure relief valve.

The third control device is primarily an emergencysafety device, which is activated when the first twocontrol devices fail. A pressure switch is mountedon the digester shell and stops the cold blow pumpwhen the pressure rises too high. Digesterpressure is normally set at 165 psig, the pressurerelief valve is set on 180 psig, and the pressureswitch at 225 psig. As the pressure continues torise, the make-up liquor pump will kick-out.

Chip FeedingProcessed chips from the wood room aretransmitted to the chip bin, which also serves as ashort term chip storage bin. This storage canfacilitate continued digester operation during smallupsets between the wood room and the digesterbuilding. Chips flow by gravity from the bin througha tapered hopper into the chip meter. The chipmeter is a rotating star feeder with seven pocketsyielding a certain volume of chips per revolution.Digester production is regulated by a variable

10B −3

Figure 10B-1. Chip Feeding

E1214

speed drive on the chip meter drive. It is importantto keep the chip meter full in order to maintain asteady feed rate. A vibrator is provided on the chiphopper for intermittent use, as required, to ensurea steady feed of chips to the chip meter.

Chips leaving the chip meter drop into a secondfeeder, called a low pressure feeder. It is simply acontinuously rotating, tapered star feeder, whichforms a seal between atmospheric pressure and15-18 psig of the next stage. Its primary function isto prevent steam leakage and to deliver chips tothe steaming vessel. Steam is injected into theends of the rotor housing in order to blow sawdustand chip fines out. The pressure remains in theempty pocket after the chips have dropped outand is relieved through a pipe connected to the top

of the chip hopper. To ensure a steady feed, thelow pressure feeder is designed so that onepocket is being filled with chips, one pocket isdischarging chips, and one pocket is relievingsteam to the chip hopper.

Valve: TV-2A Chip bin temperatureThis valve provides an alternate source of lowpressure steam to the steaming vessel where thewood chips are pre-steamed at atmosphericpressure.

� Typical process conditions:

— Fluid: Steam— T = 325 – 400�F— P = 60 – 80 psig

10B −4

Figure 10B-2. Steaming Vessel

E1215

— dP = 50 – 60 psid— Q = 2900 – 5000 lbs/hr

� Typical valve selection:

— This is specified by Kamyr as a NPS 8 toNPS 10 Fisher Vee-Ball� V150 valve, andcould need an attenuator. Carbon steel bodymaterial has been used successfully,although stainless steel would provide anadded level of durability. HD metal seats arespecified with PTFE packing, PEEKbearings, and Nitronic 50 shafts.

Valve: TV-4 Fresh steam to chip binThis valve provides hot water to the chip bin.

� Typical process conditions:

— Fluid: Hot water— T = 312�F— P = 230 psig— dP = 135 – 145 psi

� Typical valve selection:

— This valve is specified by Kamyr as aNPS 1/2 globe valve with a 300 lb. rating. Areduced port may be needed dependingupon flow requirements. A carbon steel bodyis suggested with 316 SS equal percentagetrim.

Pre-steaming and ConditioningThe steaming vessel is a normally horizontallymounted cylinder with an internal screw conveyorfor carrying chips along from the low pressurefeeder through steam to the next stage. Its main

functions are to remove gases and air from thechips, raise the temperature to approximately250�F, and bring the chips to a more uniformmoisture content. A secondary aim is to maintain apressure balance in the feeding system. Thismeans that the steam pressure (15-18 psig) in thesteaming vessel must always be higher than thevapor pressure of the liquor in the top circulationline so that the latter does not start to boil when itleaks back into the chip chute low pressure area.Removal of air and gases enables the cookingliquor to penetrate the chips more easily.

The steam is supplied from two sources: flashsteam from the No. 1 flash tank and from a freshmake-up low pressure steam header.

In order to obtain an effective pre-steaming, thesteam is introduced at the bottom of the steamingvessel (bottom steaming). There are no controlson the steam from the flash tank as it isdependent upon extraction flow from the digester.The venting of exhaust and non-condensablegases controls the amount of fresh steam usage.The exhaust line is equipped with a screen thatprevents sawdust and fine particles from beingcarried into the heat recovery area.

After passing through the steaming vessel, thechips fall from the end of the screw conveyor,down a chute known as the chip chute, into a poolof liquor. At this point, the chips start to absorbliquor. An inspection port is provided on thesteaming vessel for the monitoring of the flow ofchips into the chute.

The chip chute is a vertical pressure vessel withan internal slotted screen plate. It sits directly ontop of the high pressure feeder. The liquor in the

10B −5

chip chute is maintained at a constant level by theuse of a level control valve on the overflow line,which carries excess liquor to a surge tank knownas a level tank. The source of chip chute liquor ismainly leakage from around the high pressurefeeder and some steaming vessel condensate. Inorder to take sudden flow surges, such as whenthe high pressure feeder starts rotating or whenchips fall into the chute, the overflow control valveresponse is extremely rapid. The screen platesprevent chips from being carried with the overflowto the level tank, and keep the chips in the chipchute to feed the pockets of the high pressurefeeder.

A pressure switch is connected to the make-upliquor pump from the same pressure-sensing unitthat stops the cold blow pump. This safety deviceis necessary to avoid over-pressurizing thedigester and it is set at approximately 240 psig.

The digester operates at 165 psig, measured atthe bottom heating zone. The transfer of chips andliquor from a 15-18 psig steaming vessel pressureto the digester operating pressure needs to beaccomplished via a pressure lock system. This isaccomplished by the high pressure feeder, whichis similar to the low pressure feeder. The highpressure feeder has a tapered rotor with fourhelical type pockets which go from one side of therotor to the other and are set at an angle of 45� toeach other. The feeding of chips is continuous.Two liquor pumps are used to aid in filling anddischarging the high pressure feeder. They are asfollows:

1. Chip chute pump — Chips falling into the chipchute pool of liquor tend to float or be drawn to theside screen plates by the liquor overflowing to thelevel tank. In order to counteract this effect, thechip chute circulating pump is set to pull or suckthe chips downward into the rotating high pressurefeeder with a force greater than the sideways pullof the overflowing liquor. The discharge of thepump re-circulates the liquor to a point locatedabove the pool of liquor in the chip chute. Whensevere conditions of sawdust and fines areprevalent, an in-line grainer may be installedbetween the pump discharge and the entry abovethe pool of liquor. Its function is to take liquoraround the screen section, thus cutting down onthe sideways pull of overflowing liquor. Thepressure drop across the screens will reduce,allowing the natural wiping action of moving chipsto keep the screens from plugging. As the pumpwill deliver 2,000 - 2,500 GPM, closing the valvebetween the in-line drainer and the level tank will

allow the in-line drainer to clear itself by liquor flowto the chip chute.

2. Top circulation pump — The pump circulatesliquor to the top of the digester and back out. Itsaction is to flush the chips out of the feeder pocketonce it has rotated to the discharge point and tocarry the chips into the digester. The liquor is thenseparated from the chips via the top separatorscreen and returned to the suction of the pump toform a continuous loop.

When a pocket is in a vertical position, the chipsare fed with the help of the chip chute pump.When the pocket has rotated 90� to a horizontalposition, the chips are flushed out into the highpressure system with the liquor from the topcirculation pump. The whole system is arrangedso that there is always one pocket being filledwhile another is being emptied.

As mentioned previously, there is liquor leakagearound the rotor due to the pressure differentialbetween the operating digester pressures and thepressure in the chip chute and steaming vessel.This liquor leakage is an important feature of thehigh pressure feeder operation, as it providesconstant lubrication between the feeder plug andhousing, and washes sand and grit from this area.As the liquor is at a high temperature, it will boil inthe chip chute unless held under higher pressurethan the vapor pressure of the liquor. This isparticularly true following a shutdown when heatrises to the top of the digester from the cookingzones due to convection currents. The hot liquorboils rapidly, or flashes, at the high pressurefeeder and chip chute if the high pressure feederis started. Therefore, the high pressure feedermust never be started unless the top section ofthe digester is first cooled below 240� F byaddition of cold filtrate through the make-up liquorpump. The plug clearance is adjusted from time totime as it wears. Excessive wear of gap betweenthe plug and housing allows too much liquor topass to the chip chute and overloads the make-upliquor pump due to excess flow of liquor from thelevel tank to the make-up liquor pump. This cancause liquor to back up through the level tank intothe chip chute and even back up as far as the chipbin. The high pressure feeder serves to completethe transfer of the chips from 15 psig to digesterpressure without subjecting the chips to any harshmechanical action, which would damage the fiberand degrade the resulting pulp quality.

The top circulation line, a part of which is a holethrough the high pressure feeder, is a part of thedigester vessel itself during normal operation. Thismeans that it is under the influence of the digester

10B −6

Figure 10B-3. Cooking Flow Diagram

pressure control valve. Two large piston-operatedvalves are designed to isolate the top circulationline from the digester. These valves cannot beopened unless there is equal pressure on eachside of the valve. This prevents damage to pipingand valves by a sudden surge of liquor from thedigester flowing into the lines. A pressure switch ismounted on the return leg and is set to hold thevalves closed until the pressure in the line isalmost equal to that in the digester. Only then canthese valves be opened by hand switches on theinstrument panel. Chips only enter the digesterthrough the top separator. This unit consists of acylindrical screen with continuous vertical slotsthat separate the liquor from the chips so theliquor may be re-circulated through the highpressure feeder back to the top separator again. Aslow-moving vertical screw conveyor inside thescreen pushes the chips downward into thedigester, keeping the screen clear. The chips fallonto the top of the column of chips. This iscontinuous throughout the entire digester. In otherwords, there is a solid body of chips from the topto the bottom of the digester undergoing variousstages of treatment.

The sluicing liquor is extracted through the topseparator screen and is returned to the suction ofthe top circulation pump to be recycled with chipsto the top of the digester. Built into the top

separator is a level-indicating device, which allowsthe operator to have an indication of the level ofchips inside the digester. This level device is asmall paddle which turns with the top separatordown in the chip mass. The resistance of the chipsagainst the paddle is measured by a torqueindicator on the top of the top separator. Thismeasurement is transmitted to the control panel,which has a low level light (green), a normal(yellow) and a high (red). The digester is normallyrun at a yellow-red level indication. If the level getshigh enough so that the chips ride against thescrew conveyor, an increase in amperage of themotor will be seen on the control panel. When theload becomes severe, the top separator alarm willgo off, warning the operator to take correctivemeasures. The drive shaft of the top separator issealed by packing. Digester pressure compressesthe packing, forming a seal. When the pressure isradically changed, the packing may not properlyseal for some time.

Valve: FV-3A High pressure feederpurgeNote: In older systems, this is HV-3A or HV-35.

� Typical process conditions:

— Fluid = White liquor

10B −7

— T = 190�F— P = 240 psig – 280 psig— dP = 75 psi – 110 psi— Q= 50-125 gpm

� Typical valve selection:

— NPS 2 to NPS 3 valves with alloy 6scraper seats are utilized due to concernsover white liquor scaling. A SST V300 valvewith an alloy 6 HD seal and alloy 6 bearingsshould be used in this application.

Valve: TV-2 Flash steam to chip bintemperature control

This valve controls the flow of low pressure steamfrom the No. 2 flash tank to the steaming vesselwhere the wood chips are pre-steamed atatmospheric pressure.

� Typical process conditions:

— Fluid: Saturated steam— T = 220�F— P = 2.5 – 30 psid— dP = 1.0 psi— Q= 29000 – 50000 lbs/hr

� Typical valve selection:

— This is specified as a Fisher HPBV byKamyr. Valve size is in the range of NPS 10to NPS 18. Metal bearings and a metal seatare recommended by Kamyr. However,PEEK bearings and Teflon seats in astainless steel body and with a 17-4 splinedshaft are also appropriate.

Valve: PV-5A Steam vessel pressurerelief

This valve is used for steam pressure relief in thechip steaming vessel at the beginning of thepulping process.

� Typical process conditions:

— Fluid: Steam— T = 256 – 280�F— P = 18 – 30 psig— dP = 16 psi

� Typical valve selection:

— Kamyr specifies this valve as a NPS 8one way, tight, spring assisted full bore ballvalve due to relief valve sizing downstream.The V300 valve in stainless steel with alloy 6

trim would be a good alternative in thisapplication if the specification allows.

Valve: PV-5 Steaming vessel pressurereliefThis valve vents air and non-condensable gasesfrom the steaming vessel.

� Typical process conditions:

— Fluid: Steam and non-condensable gases— T = 312�F— P = 60 psig— dP = 17 - 40 psid— Q = 4300 lbs/hr

� Typical valve selection:

— In cases where entrained particles arefound in the flow, the recommended valve isa CV500 with hardened trim due to thepotential erosion. If no particles are found inthe flow, a standard Vee-Ball may be used.This is typically specified as a NPS 8 valve.

Valve: HV-5 Steaming vessel reliefsteam flowThis valve sends “clean” steam from the steamingvessel pressure relief to the steam condensers.

� Typical process conditions:

— Fluid: Steam— T = 255�F— P = 18 – 33 psig— dP = 16 psi— Q = 1200 lbs/hr

� Typical valve selection:

— This is a throttling application and Kamyrhas specified a V150 valve with attenuatorfor this application. Stainless steel bodymaterial together with a Nitronic 50 shaft, HDmetal seal, and PEEK bearings arerecommended. The valve is typically in theNPS 6 size range.

Valve: PV-2 Steaming vessel pressurecontrolThis valve controls the steam pressure from theNo. 1 flash tank and fresh make-up low pressuresteam.

� Typical process conditions:

— Fluid: Low pressure saturated steam— T = 320�F— P = 80 psig

10B −8

— dP = 42 psi

� Typical valve selection:

— In cases where entrained particles arefound in the flow, the recommended valve isa CV500 with hardened trim due to thepotential erosion. If no particles are found inthe flow, a standard Vee-Ball may be used.This is typically specified as a NPS 8 valve.

Valve: LV-6 Chip chute level control

This valve controls the liquor level in the chipchute. The recommended value for this applicationis a Vee-Ball, possibly with an attenuator. Carbonsteel body can be used, although stainless steelwould provide an added level of durability.

� Typical process conditions:

— Fluid: Black liquor— T = 235�F— P = 25 – 60 psig— dP = 3 psi – 7 psi— Q = 600 – 1700 gpm

� Typical valve selection:

— The Fisher CV500 will handle thisapplication with alloy 6 seal and in thereverse flow orientation. This is a scalingapplication which calls for the eccentric plugaction of the CV500. If the valve fails torespond to changing levels in the chip chute,there is a danger of liquor and chip spillage.

Valve: FV-3B White liquor pump tobottom circulation

These valves are used to control the amount ofwhite liquor which is added to the wood chips.They maintain the proper wood/liquor ratio. Asproduction increases, these valves open more tomaintain the proper ratio.

� Typical process conditions:

— Fluid: White liquor— T = 190�F— P = 260 psig – 290 psig— dP = 60 psi – 90 psi

� Typical valve selection:

— These are NPS 2 to NPS 3 valves withalloy 6 scraper seats due to concerns overwhite liquor scaling. A SST V300 valve withan alloy 6 HD seal and alloy 6 bearingsshould be used in this application.

Valve: FV-3C White liquor to washcirculation pump

� Typical process conditions:

— Fluid: White liquor— T = 190�F— P = 270 psig – 290 psig— dP = 75 psi – 90 psi

� Typical valve selection:

— NPS 2 to NPS 3 valves with alloy 6scraper seats due to concerns over whiteliquor scaling are used. A SST V300 with analloy 6 HD seal and alloy 6 bearings shouldbe used in this application.

Valve: FV-3D White liquor to MCcirculationThese valves are used to control the amount ofwhite liquor which is added to the wood chips.They maintain the proper wood/liquor ratio. Asproduction increases, these valves open more tomaintain the proper ratio.

� Typical process conditions:

— Fluid: White liquor— T = 190�F— P = 270 psig – 290 psig— dP = 70 psi – 90 psi— Q = 180 gpm

� Typical valve selection:

— NPS 2 to NPS 3 valves with alloy 6scraper seats due to concerns over whiteliquor scaling. A SST V300 valve with analloy 6 HD seal and alloy 6 bearings shouldbe used in this application.

Valve: FV-3E White liquor to washcirculation

� Typical valve selection:

— NPS 2 to NPS 3 valves with alloy 6scraper seats due to concerns over whiteliquor scaling are utilized. A SST V300 withan alloy 6 HD seal and alloy 6 bearingsshould be used in this application.

Valve: FV-3F White liquor make-up toimpregnation vesselThese valves are used to control the amount ofwhite liquor, which is added to the wood chips.They maintain the proper wood/liquor ratio. Asproduction increases, these valves open more tomaintain the proper ratio.

10B −9

� Typical process conditions:

— Fluid: White liquor— T = 190�F— P = 260 psig – 280 psig— dP = 35 psi – 60 psi— Q = 200 – 900 gpm

� Typical valve selection:

— NPS 2 to NPS 4 valves with alloy 6scraper seats due to concerns over whiteliquor scaling. A SST V300 valve with analloy 6 HD seal and alloy 6 bearings shouldbe used in this application.

Valve: LV-7 Digester level control valve

This valve controls liquor flow to the top of theimpregnation vessel. This is a critical and difficultapplication on the Kamyr digester. This valveneeds to be operable with up to 300 psi ofdifferential pressure on some digesters. Thisrequires a valve capable of tight shut-off and highpressure throttling. Cavitation is a commonproblem with this valve.

� Typical process conditions:

— Fluid: Black liquor— T = 235�F— P = 300 psig— dP = 45 psi – 90 psi— Q = 1000-3000 gpm

� Typical valve selection:

— The V500 valve or CV500 in stainlesssteel has proven to be a successful valve inthis application. The design of the Fishervalves provides higher flow velocities whichtends to reduce the problems with scalingthat are seen at this location. Both the V500valve and the CV500 should be supplied withalloy 6 seats and installed in reverse floworientation.— NPS 6 valve. This may be the mostcritical valve in the system other than blowline control. Capacity issues are a concern.

Valve: FV-61A Impregnation vesselupper slouse flow

This valve is found only on dual vessel digesterswhere it is used to add liquor to the upper dilutionzone to assist in the discharge of chips and liquorfrom the impregnation vessel.

� Typical process conditions:

— Fluid: Black liquor— T = 325�F— P = 250 psig – 290 psig— dP = 15 psi – 82 psi

� Typical valve selection:

— NPS 6 to NPS 8 valve with an alloy 6scraping seat due to concerns over theprecipitation of calcium carbonate causingplating or scaling buildup. A SST V300 valvewith an alloy 6 HD seal and alloy 6 bearingsshould be used in this application.

Valve: FV-61 Impregnation vessellower slouse flowThis valve is found only on dual vessel digesterswhere it is used to add liquor to the dilution zoneto assist in the discharge of chips and liquor fromthe impregnation vessel.

� Typical process conditions:

— Fluid: Black liquor— T = 345�F— P = 235 psig – 260 psig— dP = 25 psi – 52 psi

� Typical valve selection:

— NPS 4 to NPS 6 valve with an alloy 6scraping seat due to concerns over theprecipitation of calcium carbonate causingplating or scaling buildup. A SST V300 valvewith an alloy 6 HD seal and alloy 6 bearingsshould be used in this application.

Valve: PV-30 Impregnation vesselpressure reliefThis valve is used to relieve excess pressure inthe impregnation vessel by releasing excess liquorto the No. 2 flash tank.

� Typical process conditions:

— Fluid: Black liquor— T = 235�F— P = 180 psig— dP = 155 psi— Q = 2800gpm

� Typical valve selection:

— A CV500 (NPS 3) in stainless steel withalloy 6 trim is a good valve for thisapplication.

Valve: FV-60 Black liquor toimpregnation vessel (bottomcirculation flow)This valve is found only on dual vessel digestersand is used to add black liquor to the outlet of the

10B −10

impregnation vessel to raise the temperature ofchips and liquor going to the digester.

� Typical process conditions:

— Fluid: Black liquor— T = 345�F— P = 250 psig – 275 psig— dP = 38 psi – 75 psi

� Typical valve selection:

— NPS 6 to NPS 8 valve with an alloy 6scraping seat due to concerns over theprecipitation of calcium carbonate causingplating or scaling buildup. A SST V300 valvewith an alloy 6 HD seal and alloy 6 bearingsshould be used in this application.

Valve: TV-60A-C Bottom circulationtemperatureThese valves control steam to the bottomcirculation heaters which, in turn, control thetemperature of liquor leaving the heaters for thebottom zone of the impregnation vessel.

� Typical process conditions:

— Fluid: Steam— T = 400 – 525�F— P = 165 psig— dP = 60 psi— Q = 10000 – 40000 lbs/hr

� Typical valve selection:

— These are throttling valves, typically NPS6, for which Kamyr has specified V300valves with stainless steel bodies andstainless steel trim. Nitronic 50 shafts, PEEKbearings, and HD metal seals are also calledfor.

Valve: PV-10 Digester pressure reliefThis valve is used in emergency situations torelieve elevated digester pressure by releasingblack liquor from the top screens to the No. 2 flashtank. This valve requires tight shutoff andfail-closed spring return actuation.

� Typical process conditions:

— Fluid: Black liquor— T = 350�F— P = 180 psig— dP = 180 psi maximum— Q = 2600 gpm

� Typical valve selection:

— Kamyr has specified full bore ball valvesfor this application. However, a CV500 (NPS3 to NPS 6) in stainless steel with alloy 6 trimis an excellent valve for this application.

Valve: PV-11 Digester pressure controlThis is a critical valve in the digester process. Thisvalve is used to maintain pressure in the digesterand to distribute fresh cool water to the pulp. Thisprevents the overheating of the pulp, which wouldresult in the degradation of the pulp fibers. Thisvalve is typically interlocked with PV-10 to preventover-pressurization of the process.

� Typical process conditions:

— Fluid: Washer filtrate— T = 180�F— P = 250 psig— dP = 55 psi

� Typical valve selection:

— Kamyr has specified a Vee-Ball V300valve, NPS 4 to NPS 6 size, for thisapplication. The valve should have astainless steel body and ball, Nitronic 50shaft, PEEK bearings, and an HD metal seal.A piston actuator is typically used.

Valve: HV-54 Top circulation pressurecontrolThis valve is used primarily during start-up tosupply cooking liquor to pressurize the topcirculation lines prior to pumping.

� Typical process conditions:

— T = 170�F— P = 320 psig— dP = 45 psid— Q = 400 gpm

� Typical valve selection:

— The V300 valve is an appropriatealternate valve for this tag. This valve istypically in the NPS 3 size range.

Valve: HV-52 Top circulation isolationvalveThis valve serves to isolate the high pressurefeeder line from the impregnation vessel. It istypically installed on the main chip/liquor feed-line.

� Typical process conditions:

— T = 240�F— P = 250 psig

10B −11

— Q = 9000 gpm

� Typical valve selection:

— This valve is specified as a full bore ballvalve by Kamyr, and is referred to as theRO1 valve. It is normally open, and onlyclosed when the digester must be isolatedfrom the HP feeder. NPS 12 to NPS 16 insize with additional 1/8 inch thickness on RFmale flange.— This unique size makes this valve anon-ANSI flange thickness. Fisher does nothave an offering available for this valve.

Valve: HV-51 Top circulation isolationvalveThis is an isolation valve for liquor being sent backto the high pressure feeder from the impregnationvessel.

� Typical valve selection:

— This valve is specified to be a full boreball valve by Kamyr, and is referred to as theRO2. It too has the special “male flange”requirement as referenced in HV-52.

Valve: PDV-18 Digester outlet devicedifferential pressureThis valve adds liquor to the bottom of the digesterto assist in chip discharge and to help regulateconsistency.

� Typical process conditions:

— Fluid: Washer filtrate— T = 170�F— P = 238 psig— dP = 45 psi— Q = 100 – 1000 gpm

� Typical valve selection

— This is typically a NPS 4 valve for whichKamyr has specified a V300 valve withstainless steel body and ball. The valveshould be offered with a Nitronic 50 shaft,PEEK bearings, and an HD metal seal. Apiston actuator is normally specified.

ImpregnationThe chips at the top of the column now enter theliquor impregnation zone. In this zone, chips aresubjected to a complete soaking or penetration ofthe cooking liquor at a temperature ofapproximately 250�F. The impregnation stagelasts about 45 to 60 minutes at design tonnage. It

is important that thorough penetration takes placebefore the heating stage. If penetration isincomplete, chips with uncooked centers willresult.

The white liquor, or cooking liquor, is added to thedigester together with black liquor, which isrecovered from the chip chute overflow. It is alsopossible to put weak black liquor into the makeupliquor pump suction, but this is normally notrequired.

The chip chute overflow goes to the surge tank(level tank). There, it is controlled at a constantlevel, and then goes to the make-up liquor pumpwhere it is combined with the white liquor. Themake-up liquor pump transports the mixture to thetop of the digester and injects it at a point justbelow the top separator. All of this liquor returns tothe high pressure feeder by the top circulationreturn line, and is mixed with fresh chips to bereturned to the inside of the top separator andthen to flow down the digester. The amount ofwhite liquor added is based upon the productionrate and is in direct proportion to the chip feedrate.

Valve: TV-3A White liquor to make-upliquor line temperature

� Typical process conditions:

— Fluid: White liquor— T = 120� F— P = 55 psig— dP = 10 psi

� Typical valve selection:

— This valve has been specified by Kamyrto be a NPS 3 V150 valve with stainlesssteel body and 317 stainless chrome platedball, Nitronic 50 shaft, HD metal seal, PEEKbearings, and a fail-close actuator.

Valve: FV-4 Black liquor flowThis valve only sees infrequent use in the process.Its function is to add cool washer filtrate liquor tothe black liquor line coming from the level tank inthe event the black liquor and the temperature atthe top of the digester exceed specifiedtemperature limits.

� Typical process conditions:

— Fluid: Washer filtrate— T = 180�F— P = 30 – 60 psig— dP = 10 – 60 psid

10B −12

— Q = 500 gpm

� Typical valve selection:

— Kamyr has specified a V150 Vee-Ballvalve, NPS 2 – NPS 3 size, with stainlesssteel body and ball, Nitronic 50 shaft, PEEKbearings, and an HD metal seal. Typically, afail closed actuator is used.

Valve: LV-17 No. 2 Flash tank levelBlack liquor level in the No. 2 flash tank iscontrolled by LV-17, located between the No. 2flash tank outlet and the evaporators.

� Typical process conditions:

— Fluid: Black liquor— T = 220�F— P = 60 psig— dP = 10 – 20 psid— Q = 1000 – 2000 gpm

� Typical valve selection:

— NPS 8 to NPS 10 butterfly valve thatKamyr has specified as a stainless steelHPBV with stainless steel disc, alloy 6bearings, 17-4 shaft, and a NOVEX metalseal.

Valve: LV-16 No. 1 Flash tank levelBlack liquor level in the No. 1 flash tank iscontrolled by LV-16 located at the outlet of thetank. This valve flows into the No. 2 flash tank.

� Typical process conditions:

— Fluid: Foamy black liquor— T = 260�F— P = 20 – 50 psig— dP = 2 – 12 psid— Q = 500 – 1500 gpm

� Typical valve selection:

— This valve is specified by Kamyr as aHPBV with a stainless steel body and discand a NOVEX metal seal. alloy 6 bushingsand a 17-4 shaft are recommended. Thisvalve is in the NPS 8 to NPS 14 size range.

Valve: LV-81 Clean condensate flashtank level

� Typical process conditions:

— Fluid: Clean condensate— T = 312�F— P = 90 psig

— dP = 90 psi maximum

� Typical valve selection:

— This valve has been specified by Kamyrto be a NPS 2 to NPS 3 V150 valve withstainless steel body and trim, Nitronic 50shaft, HD metal seal, PEEK bearings, and afail-close actuator.

Valve: LV-91 Contaminated condensate flash tanklevel

� Typical process conditions:

— Fluid: Condensate— T = 212�F— P = 11 psig— dP = 10 psi

� Typical valve selection:

— This valve has been specified by Kamyrto be a NPS 2 to NPS 3 V150 valve withstainless steel body and trim, Nitronic 50shaft, HD metal seal, PEEK bearings, and afail-close actuator.

Valve: KV-24 Sand separator valveThis is also called the “pocket valve”. This is a fullbore ball valve with one of the normally open endsof the ball sealed. It is used to collect sand at thebottom of the sand separator. Occasionally, thevalve rotates 180� to dump the collected sand.

� Typical valve selection:

— Full bore valves that can rotate 145degrees are chosen. It is typically NPS 6 toNPS 8, and cycles every five to ten minutes.El-O-Matic� actuators with 180 degreerotation with an adjustment are usedcommonly. Due to high cycle life, this valveassembly needs to be inspected on a regularbasis.

Heating StageFrom the impregnation zone, the chip columncontinues to move down until the upper cookingzone is reached. Chips and liquor are pre-heatedto within 20�F of the actual cooking temperaturewhile the lower cooking zone controls the finalcooking temperature. This is accomplished bywithdrawing liquor from the digester through ascreen plate, circulating it through an indirectheater and returning the heated liquor downthrough a central distribution chamber. Then, it isdischarged at a point opposite and slightly abovethe screen plates. This allows the chips to receiveuniform heating as the liquor enters at the center

10B −13

and flows to the outer shell equally in alldirections.

Chips continue down the digester for a shortdistance into the lower cooking zone, where asecond heating process occurs. In this case, thechips are heated to the desired cookingtemperature. To accomplish this, liquor is againextracted radially through screen plates to asecond heater. It is returned through a pipe, whichis located inside the central distribution chamber,and is discharged at a point just above the lowercooking screen plates. A spare heater is providedand may be valved as either the upper or lowerheater so that the heat exchangers, whichperiodically become fouled by liquor scale, can beacid-cleaned while the digester is still in operation.Normally, the impregnation zone is subject toconsiderable scaling and the upper heater willrequire more cleaning than the lower. The time aheater may be run before it requires cleaning mustbe determined by operating experience on yourliquor, wood, and cooking conditions. Oncedetermined, heaters may be cleaned before theybecome fouled so severely that the tubes plug.

The function of all circulations is two-fold: first, tocarry heat and chemicals into the digester;second, to homogenize the conditions in thedigester cross-section. If the circulation is to workproperly, the screen plates must be kept clean.This will not be possible if the chip column standsstill and a large quantity of liquor is drawn throughthem. Therefore, the screen sections are built intwo sections, one above the other. These two setsoperate alternately, one set sucking while theother rests and is cleaned by the wiping action ofthe downward moving chip column. The change isaccomplished by digester switching valvesmounted in the circulation suction lines. The chipcolumn is not too dense at this point, and alternateautomatic switching time is fairly short (typically 90seconds).

Temperature recorders on the upper and lowercooking zone heaters show a cycle of 5� to 10�F.The temperature cycle is due to the temperaturedifferential across the heating zone screens andfollows the timing of the switching valves. Therecorded temperatures are averages for the entirescreen section, and provided the liquor circulationis adequate, the end temperature is in control. Theinlet and outlet of each heater should be within10�F of each other.

The temperature recorders serve as an indicationof the chip column movement. Interruptions ofmovement are reflected by temperature changes.

Failure of the chip column to move is reflected bythe coming together of the inlet and outlet heatertemperatures, whereas a sudden drop of the chipcolumn is shown by a sharp drop in the heaterinlet temperature.

Valve: KV-8A and B Trim liquorswitching valvesThese valves extract liquor through screenslocated in the upper part of the digester. Theextracted liquor is then sent to the bottomcirculation heaters. The KV tagged valves arerequired to fully stroke approximately every 90seconds. This causes a flow reversal through theextraction screens preventing the screens fromplugging with chips and fiber.

� Typical process conditions:

— Fluid: Black liquor— T = 325�F— P = 130 psig— dP = 130 psi— Q = 1500 gpm

� Typical valve selection:

— NPS 6 to NPS 8 size range. The DSVvalve is suitable for this application. This is amodified 8510 body with a strengthenedshaft and no seal. Used in conjunction withthe 1061 actuator with a quad seal option,this assembly is capable of a relatively longlife in this service.

Valve: KV-60A and B Bottomcirculation return switching valvesThese valves extract liquor from the digester (inthe upper region of the digester vessel) and sendliquor to the bottom circulation heaters. Thesevalves are only found on dual vessel digesters.The KV tagged valves are required to fully strokeapproximately every 90 seconds. This causes aflow reversal through the extraction screenspreventing the screens from plugging with chipsand fiber.

� Typical process conditions:

— Fluid: Black liquor— T = 325�F— P = 120 psig— dP = 120 psi

� Typical valve selection:

— NPS 12 to NPS 14 size range. The DSVvalve is suitable for this application. This is amodified 8510 body with a strengthened

10B −14

shaft and no seal. Used in conjunction withthe 1061 actuator with a quad seal option,this assembly is capable of a relatively longlife in this service.

Valve: KV-60C and D Bottomcirculation screen backflush valvesThese valves work in conjunction with tags KV-60A and B to ensure that the bottom circulationscreens remain free of chips. These valves areonly found on dual vessel digesters. The KVtagged valves are required to fully strokeapproximately every 90 seconds. This causes aflow reversal through the extraction screens,preventing the screens from plugging with chipsand fiber.

� Typical process conditions:

— Fluid: Black liquor— T = 325�F— P = 120 psig— dP = 120 psi

� Typical valve selection:

— NPS 6 to NPS 8 size range. The DSVvalve is suitable for this application. This is amodified 8510 body with a strengthenedshaft and no seal. Used in conjunction withthe 1061 actuator with a quad seal option,this assembly is capable of a relatively longlife in this service.

Valve: PV-16 PV-17 No. 1 Flash tankrelief steam pressure valvesThese valves are used to slightly pressurize theflash tanks in order to reduce foaming of the blackliquor in the tanks.

� Typical process conditions:

— Fluid: Steam— T = 220�F— P = 60 psig— dP = 5 psi— Q = 15000 – 75000 gpm for PV-16— Q = 20000 – 100000 gpm for PV-17

� Typical valve selection:

— Both of these valves have been specifiedby Kamyr as HPBV valves with PV-16 beingan NPS 8 or NPS 10 valve and PV-17 beingan NPS 16 to NPS 18 valve. Both valvesshould be supplied with a stainless steelbody and disc, 17-4 shaft, PEEK bearings,and a NOVEX metal seal.

Cooking ZoneBelow the heating zones, the chips enter thecooking zone at the full cooking temperature.Here, the actual cooking takes place. The activechemicals in the cooking liquor are sodiumhydroxide, NaOH, and sodium sulfide, Na2S.These chemicals react with the lignin in the woodchips, converting it into chemical compoundswhich dissolve in the alkaline cooking liquor. Thelignin, as it exists in wood, is somewhat like acementing material and holds the individual fiberstogether; however, when it is made soluble duringcooking, the fibers are set free and can beseparated into the fibrous mass called wood pulp.

The chemicals in the cooking liquor also react withthe pulp fibers themselves. This is not desirablebecause the fibers are required in their originalcondition. Therefore, cooking conditions are usedwhich result in the highest removal of lignin withthe least attack on the cellulose fibers.

The chips continue to fall and will increase slightlyin temperature until cooking is complete and it hasreached the next stage (hi-heat washing). Thecooking reaction is stopped by cooling the chipmass down to about 280� - 300�F. This is done byextracting the residual cooking liquor throughscreens on the side of the digester, and replacingit with cool liquor, which rises from the wash zone.The amount of cool liquor used is determined bythe quantity of wash liquor required to wash thepulp to a suitable low soda content. In order tocontrol the uniformity of temperature, a portion ofthe extracted cooking and washing liquor isreturned to the digester via the quench circulationpump through a pipe inside the central distributionchamber. This re-circulation lowers thetemperature of the entire chip mass and permitswashing to be carried out without furtherdelignification or over-cooking.

Valve: KV-19 A-D A-F Modified cookingextraction switching valvesThese valves extract liquor in the modified cookingzone of the digester located near the middle of thedigester vessel. The extracted liquor is then sentto the modified cooking heater. The KV taggedvalves are required to fully stroke approximatelyevery 90 seconds. This causes a flow reversalthrough the extraction screens preventing thescreens from plugging with chips and fiber.

� Typical process conditions:

— Fluid: Black liquor— T = 325�F

10B −15

— P = 170 psig 10— dP = 170 psi

� Typical valve selection:

— These are typically in the NPS 8 sizerange for KV-19 A, B, C, D and in the NPS 3size for KV-19 E and F. The DSV valve issuitable for this application. This is amodified 8510 body with a strengthenedshaft and no seal. Used in conjunction withthe 1061 actuator with a quad seal option,this assembly is very capable of a relativelylong life in this service.

Extraction and Hi-Heat WashingChips, having now passed through the cookingzone, reach the extraction screens. The column isconsiderably denser following cooking and has abetter wiping action so that a cycling system is notrequired at this zone. The section from theextraction screens down to the bottom of thedigester is the wash zone. Black liquor is extractedfrom two rows of screen plates. The upper screenextracts primarily the hotter spent liquor from thedownward flow of chips and the lower screenextracts cooler liquor flowing upward, orcountercurrent, from the washing zone.

The portion of extracted cooking liquor that is notreturned to the digester through the quenchcirculation pump goes to the No. 1 flash tank,which serves as the first stage of the digester heatrecovery system. The liquor leaves the digester atabout 180 psig and discharges to the flash tank at15-18 psig. The sudden pressure drop causes theliquor to boil rapidly, or flash, and form steam,which goes to the steaming vessel. The amount ofsteam produced is directly proportional to the hotliquor flow through the extraction line and reducesthe flow of the fresh make-up steam required atthe steaming vessel. The remaining liquor thenbleeds off to the No. 2 flash tank and is pumped tothe unoxidized weak black liquor storage tank. Asthe No. 2 flash tank is under atmosphericpressure, again flashing occurs and the flashsteam goes to a condenser.

Counter-current hi-heat washing is accomplishedby extracting a greater volume of liquor throughthe extraction screens than the liquor volumecoming down with the chips, causing an upflow ofwash filtrate. The wash filtrate is pumped in at thebottom of the digester where it fills the voidscreated by the increased extraction. It also flowsout through the blow line with the pulp acting asdilution liquor.

Wash filtrate is weak black liquor produced byextraction and shower displacement of residualliquor from the pulp. This filtrate is returned to thedigester, mixes with the stronger cooking liquors,and cools the pulp before it is discharged (coldblow). The ratio of the number of pounds ofexcess filtrate (upflow) added per minute to thenumber of pounds of O.D. pulp produced perminute is called “dilution factor”. For example, adilution factor of two would require a net upflow inthe wash zone of two pounds per minute of filtrateto each pound of A.D. pulp. The amount of filtrateis determined by the remaining soda content in thepulp and should be held to a minimum value. Morefiltrate produces less soda loss, but at some point,filtrate addition, which later has to be evaporatedby steam heat, is greater than the cost of soda. Itcan, therefore, be seen that a balance must bemade at the most efficient point.

Washing efficiency increases with increasedtemperature. Therefore, it is necessary to heat thewash liquor. Thus, another extraction screen islocated near the bottom of the digester whichdraws liquor out. This liquor is circulated via thewash pump through a steam heater and returnedthrough yet another chamber located inside thecentral distribution chamber. This heated liquordischarges at the lower wash screen and diffusesupward through the downflowing chips. Iteffectively washes the mass as it replacesstronger liquor, which has been extracted above.

The reason for injecting the cold liquor into thebottom and then extracting it to be heated forwashing is explained in the next section. The pulpcolumn has now reached the final or blowingstage.

Valve: HV-16 Digester washerextraction flowThis valve controls the flow of black liquor to theflash tank.

� Typical process conditions:

— Fluid: Black liquor— T = 325�F— P = 130 psig— dP = 45 psi – 60 psi— Q = 500 – 2500 gpm

� Typical valve selection:

— This application is typically specified as afull-bore ball valve. A CV500 in stainlesssteel will be a suitable valve for thisapplication due to its ability to handle scalingprocess conditions.

10B −16

Valve: KV-16A-D Digester extractionswitching valvesThese valves provide screened liquor extractionfrom the upper wash zone of the digester. This islocated near the middle of the digester vessel. Theextracted liquor is then sent to flash tank No. 1.The KV tagged valves are required to fully strokeapproximately every 90 seconds. This causes aflow reversal through the extraction screenspreventing the screens from plugging with chipsand fiber.

� Typical process conditions:

— Fluid: Black liquor— T = 325�F— P = 10 psig— dP = 150 psi— Q = 1500 gpm

� Typical valve selection:

— NPS 6 to NPS 8 size range. The DSVvalve is suitable for this application. This is amodified 8510 body with a strengthenedshaft and no seal. Used in conjunction withthe 1061 actuator with a quad seal option,this assembly is capable of a relatively longlife in this service.

Valve: TV-9H Modified cookingcirculation temperature valveThis valve controls steam to the Modified cookingheater which, in turn, controls the temperature ofliquor going to the upper wash zone.

� Typical process conditions:

— Fluid: Steam— T = 379�F— P = 165 psig— dP = 80 psi

� Typical valve selection:

— This is a throttling valve, typically NPS 4to NPS 6, for which Kamyr has specifiedV300 valves with stainless steel bodies andstainless steel trim. Nitronic 50 shafts, PEEKbearings, and HD metal seals are also calledfor.

Valve: HV-20 Washer heater circulationflowThis valve controls the flow of wash liquor from thebottom of the digester to the wash circulationheater of the main digester.

� Typical process conditions:

— Fluid: Black liquor— T = 255�F— P = 250 psig— dP = 20 psi – 60 psi

� Typical valve selection:

— This is typically not a scaling application.A V300 valve or a CV500 will be suitable.

Valve: FV-14 Inner counterwash flowThis valve controls the flow of washer filtrate tothe pulp washing section of the digester vesseland is typically a NPS 4 valve.

� Typical process conditions:

— Fluid: Washer filtrate— T = 170�F— P = 235 – 350 psig— dP = 5 – 130 psid— Q = 100 – 1000 gpm

� Typical valve selection:

— Kamyr has specified a V300 valve withstainless steel body and ball. The valveshould be offered with a Nitronic 50 shaft,PEEK bearings, and an HD metal seal. Apiston actuator is normally specified.

Valve: TV-20 Wash circulationtemperature controlThis valve controls the flow of low pressure steamto the wash circulation heater for the digester.

� Typical process conditions:

— Fluid: Steam— T = 379�F— P = 165 psig— dP = 55 psi

� Typical valve selection:

— This is a throttling valve, typically NPS 6to NPS 8, for which Kamyr has specifiedV300 valves with stainless steel bodies andstainless steel trim. Nitronic 50 shafts, PEEKbearings, and HD metal seals are also calledfor.

Valve: KV-20A-D Wash extractionswitching valvesThese valves extract wash water from screenslocated near the bottom of the digester vessel.The extracted wash water is then sent to the washheater.

10B −17

� Typical process conditions:

— Fluid: Black liquor— T = 260�F— P = 185 – 195 psig— dP = 185 – 195 psi— Q = 1500 gpm

� Typical valve selection:

— These are typically in the NPS 3 to NPS 8range. The DSV valve is suitable for thisapplication. This is a modified 8510 bodywith a strengthened shaft and no seal. Usedin conjunction with the 1061 actuator with aquad seal option, this assembly is capable ofa relatively long life in this service.

BlowingAt high temperatures, particularly over 220�F, themechanical action of the chips resulting from theviolent expansion during the blow is harmful to thefibers. As mentioned before, washing efficiencyincreases with increased temperature; however,damage to fibers will be reduced by blowing at alower temperature. In order to overcome any lossin pulp strength, the pulp mass is cooled to about190�F by the weak black liquor filtrate. This liquoris pumped via the cold blow pump into the digesterat two locations: a portion through a screen platearound the digester shell near the bottom andanother portion through four nozzles located underthe paddles of the outlet device. The liquorentering through the screens regulates thedigester pressure, whereas the flow under theoutlet device is fixed by the operator and providesthe required dilution before the pulp is actuallyblown from the digester.

Summarizing, we find that the chips undergo threebasic temperature changes that take place fromthe extraction zone to the blowing of pulp from thedigester:

First, the chips are cooled or quenched 40 to 50�Fat the extraction zone. This drop occurs where theup-flowing wash liquor meets the down-flowingresidual cooking liquor and both are drawn off.Most of the residual liquor goes to the No. 1 flashtank for heat and chemical recovery. Theremainder is re-circulated via the quenchcirculation pump before being recovered at theflash liquor tank. Re-circulation provides a moreuniform temperature control and is important toavoid over-cooking the pulp.

The chips are then cooled (30�F) graduallythroughout the wash zone until a second sharp

drop occurs in the blowing dilution zone. Here thechips are cooled rapidly again by about 60�Fbefore blowing to avoid mechanical damage to theexploding chip fibers. The blow line temperaturemust be maintained at 210�F or less. The gradualtemperature drop over the wash zone is simply aresult of gradual heat exchange between theup-flowing wash liquor and the down-flowing chips.

The reverse procedure occurs with the cold filtrateliquor (160�F). After entering the digester, it isheated to about 190�F by heat transfer from thechips and is then extracted to be heated by thewash heater (260 - 280�F) before returning to thedigester wash zone. It gradually increases intemperature as it picks up heat from thedown-flowing chips and reaches the lowerextraction at about 290�F to provide the quench atthe end of the cook zone. The wash heater controlpoint is, therefore, set to arrive at the properquench temperature and it is affected by theproduction rate, cooking temperature, and dilutionfactor.

Chips, upon reaching the outlet device, are nowready to be discharged from the digester. A loadreading ammeter on the outlet device motorserves to indicate the consistency in the bottom ofthe digester. It has a variable speed drive and thespeed at which it runs directly affects theconsistency; the faster it runs, the higher theconsistency and vice-versa. The reason for this isthat the arms of the outlet device are designed toscrape the chips from the bottom of the columnand carry them into the discharge port. Throughexperience, the operator learns the best speedsfor each production rate. It can also be seen fromthe above that the outlet device is used in the finecontrol of the digester chip level.

The chips then enter a 12-inch blow line and goesto a small cigar shaped vessel called the blow unit.The blow unit is equipped with an agitator and twoexit lines to the blow tank. The purpose of theagitator is to act as a consistency indicator for thedigester operator by a torque or power-sensingdevice. As the consistency goes up or down, thepower required to turn the agitator goes up ordown. The variation is recorded on a chart at thecontrol panel.

The chips and liquor flow out of the blow unit intoone of the blow lines, and the flow is measuredand recorded by a magnetic flowmeter. Cookedchips are at digester pressure up to the final ballvalve before the blow tank. As they pass through,they are subjected to a sharp pressure drop whichcauses the chips to explode and break up intoindividual fiber bundles which are a form of raw

10B −18

pulp. The pressure drop is from digester pressureto atmospheric in the blow tank.

As in the top circulation line, there are twoisolation valves against digester pressure. There isa large isolation valve between the digesterpressure and the blow unit. If the blow unit isempty and the valve is opened under full digesterpressure, the sudden surge severely damages theblow unit and valves. The blow unit must,therefore, be filled with liquor and pressurizedbefore the large valve can be opened in a similarmanner to the top circulation lines. A pressureswitch prevents the valve from opening until theblow unit pressure is increased to at least 175psig.

The pulp flows to the blow tank, which is simply astorage tank with a predetermined retention time,before it goes to the next stage in the pulpinggroup, pulp washing and high density storage.

Valve: HV-87 Blow line pressurizationThis valve provides dilution and pressurization inthe blow line prior to the opening of the blow valve.

� Typical process conditions:

— Fluid: Washer filtrate— T = 170�F— P = 225 – 350 psig— dP = 10 – 50 psid— Q = 165 gpm

� Typical valve selection:

— It is typically a NPS 2 valve. Kamyr hasspecified a V300 valve with stainless steelbody and ball. The valve should be offeredwith a Nitronic 50 shaft, PEEK bearings, andan HD metal seal. A piston actuator isnormally specified.

Valve: FV-12A Blow line flow control� Typical process conditions:

— Fluid: High consistency pulp (10%)— T = 180�F— P = 170 psig— dP = 80 psi

� Typical valve selection:

— The requirement for full bore is based onthe potential for plugging as large depositscome out of the digester. This is a verydemanding and critical loop. The shape ofopening and flow area of the V300, by

design, has a lower potential for pluggingthan a full bore valve. Recommended valveconstruction includes: SST body with alloy 6insert, alloy 6 V-notch, alloy 6 taper key,alloy 6 silver plated bearings, alloy 6 HDseal, and PTFE ENVIRO-SEAL packing. Anoversized, piston actuator is also suggested.In addition, the valve should be sized tooperate at >60% opening.

Valve: FV-12B Blow line flow control� Typical process conditions:

— Fluid: High consistency pulp (10%)— T = 180�F— P = 170 psig— dP = 80 psi

� Typical valve selection:

— The requirement for full bore is based onthe potential for plugging as large depositscome out of the digester. This is a verydemanding and critical loop. The shape ofopening and flow area of the V300 valve, bydesign, has a lower potential for pluggingthan a full bore valve. Recommended valveconstruction includes: SST body with alloy 6insert, alloy 6 V-notch, alloy 6 taper key,alloy 6 silver plated bearings, alloy 6 HDseal, and PTFE ENVIRO-SEAL packing. Anoversized piston actuator is also suggested.In addition, the valve should be sized tooperate at >60% opening.

Valve: HV-81B Blow line isolation valveThis valve is an on/off valve that serves to isolatethe digester blow line.

� Typical valve selection:

— It is specified by Kamyr as a full bore ballvalve.

Valve: HV-90A and B Blow lineisolation valves

� Typical process conditions:

— Fluid: High consistency pulp (10%)— T = 180�F— P = 70 psig— dP = 4 psi

� Typical valve selection:

— These valves are generally the samevalves as the flow control valves with theexception that they are often one line sizelarger and used for isolation only. These

10B −19

valves are on/off valves but are usuallyequipped with positioners so that they can beused as backup flow control valves.

Control Valve SelectionThe following charts list Fisher valve selections fora typical Kamyr process. Control valve metallurgyhas usually been 316 SST except for some valveson the bleach plant. On the C/D, D1 and D2extractions and stock flow to these stages, thevalves are usually titanium and sometimes 317SST. More recently, the use of carbon steel valveson steam and some filtrate service has beenconsidered. Metal seated ball valves are usedwhere the service requires a scraper seat (HDdesign) such as white liquor and cookingcirculations on the digester. Metal seats are alsosued on throttling service that have high pressure

drops where the metal seat is more resistant toerosive wear at the resultant high velocity. Theheavy duty butterfly valve (Special 8500) is usedalmost exclusively for digester circulationswitching. The main features of this valve for thisservice are the stellite bearings, double packing,and extra heavy design shaft. There are no seatsin this valve so that it will not jam due to scalebuildup. It is also sometimes used on throttlingservice where tight shutoff is not required, butwhere there is a scaling tendency. The history ofvalve selection has not been determined solely bythe process requirement but in many cases bywhat valve technology was available at the time.The choice is sometimes limited due to Kamyr’srequirement for flanges on all except butterflyvalves.

10B −20

KAMYRTAG#

KAMYR CONTINUOUS DIGESTER FISHER CONTROL VALVEPRODUCT DESIGN

Application Description ControlFunction

V150 V200 V300 V500 CV500 8580TypicalValveSized

FV-3AHigh Pressure Feeder PurgeWhite liquor flow control to purgeHigh Pressure Feeder end bells

T P 2’’

FV-3B White Liquor to Bottom CirculationControls white liquor to Bottom Circulation

T P 2’’

FV-3C White Liquor to Bottom CirculationControls White Liquor to Bottom Circulation

T P 2’’

FV-3D White Liquor to Modified Cooking CirculationControls white liquor to Modified Circulation

T P 3’’

FV-3FWhite Liquor Flow to Make-Up Liquor Linethis is the white liquor flow to theImpregnation Vessel

T P 3’’

FV-4Black Liquor FlowWasher filtrate is sent to black liquor linefrom Level Tank to cool black liquor

T P 3’’

FV-12ABlow Line Flow (a)Controls flow from main Digester. 10-12% consistency

T P 6’’

FV-12B Blow Line Flow (B). See FV12A T P 6’’

FV-14 Inner − Counterwash FlowControls liquor flow to pulp washing section

T P 4’’

FV-60(1)

Bottom Circulation FlowBlack liquor added to outlet of ImpregnationVessel to raise temperature of chips and liquorto Digester

T P 8’’

FV-61(1)Impregnation Vessel Bottom Dilution (Lower)Adds liquor to dilution zone to assist dischargeof chips and liquor from Digester

T P 6’’

FV-61A(1) Impregnation Vessel Bottom Dilution (Upper)See FV61 except upper zone

T P 6’’

PDV-18Digester Outlet Device Differential PressureAdds liquor to assist in chip dischargefrom Digester and regulate consistency

T P 4’’

LV-6 Chip Chute LevelControls liquor level in Chip Chute

T P 8’

LV-7Level Tank LevelValve controls liquor level in tank.Full 300 psi drop across valve is possible

T S P 6’’

LV-16 No. 1 Flash Tank LevelControls level in No. 1 Flash tank

T P 30’’

LV-17 No. 2 Flash Tank LevelControls level in No. 2 Flash tank

T P 20’’

PV-5Steam Vessel Pressureprovides low pressure steam to SteamingVessel

T P 8’’

PV-5A Steaming Vessel Safety ReliefRelief valve for Steaming Vessel

T P 8’’

PV-10Digester Pressure Reliefthis valve relieves excess liquor fromtop screens to Flash Tank No. 2

T P 4’’

PV-11

Digester PressureThis valve regulates digester pressure andcontrols temperature using washer filtrate in thedischarge zone

T P 6’’

1. Dual vessel only.CODE: P = Primary selection, S = Secondary selection, T = Throttling, O/O = On/Off

10B −21

KAMYRTAG#

KAMYR CONTINUOUS DIGESTER FISHER CONTROL VALVEPRODUCT DESIGN

Application Description ControlFunction

V150 V200 V300 V500 CV500 8580TypicalValveSized

PV-16 No. 1 Flash Steam Pressurethis valve controls pressure in No. 1 Flash Tank

T P 10’’

PV-17 No. Flash Steam PressureThis valve controls pressure in No. 2 Flash Tank

T P 18’’

PV-30(1)Impregnation Vessel Pressure ReliefRelieves excess liquor from ImpregnationVessel to No. 2 Flash Tank

T P 4’’

TV-2Chip Bin TemperatureTakes steam from flash tank #2 to providesteam for atmospheric presteaming in Chip bin

T P 18’’

TV-2AChip Bin TemperatureProvides alternate source of steam from lowpressure steam line to Chip Bin

T P 8’’

TV-19H

Modified Cooking Circulation TemperatureValve controls steam to Modified CookingHeater which controls temperature of liquor toupper wash zone

T P 6’’

TV-20HWash Circulation TemperatureControls low pressure steam to WashCirculation Heater of the Digester

T P 6’’

TV-60A(1)

Bottom Circulation TemperatureValve controls steam to Bottom CirculationHeaters which control temperature of liquorleaving heaters to bottom zone of theImpregnation Vessel

T P 6’’

TV-60B(1) Bottom Circulation TemperatureSame as TV60A

T P 6’’

TV-60C(1) Bottom Circulation TemperatureSame as TV60A

T P 6’’

HV-5 Steaming Vessel ReliefRelief Valve, sends steam to Condenser

T P 6’’

HV-5ASteaming Vessel Relief Screen BlowbackThis valve is used to blow back fresh steam toclean relief screen

O/O S P 1.5’’

HV-8(1)Trim Liquor DownflowControls liquor extracted from upper screensand sends it to bottom Circulation Heater

T P 8’’

HV-16

Digester Extraction to No. 1 Flash Tankthis valve controls extraction flow fromextraction screens KV16A, B, C, D to FlashTank #1

T P 8’’

HV-19Modified Cooking Circulation FlowControl liquor flow to modified cooking zonefrom Modified Cooking Heater

T P 8’’

HV-20Wash Circulation Flow − Controls wash liquorfrom bottom of digester to the Wash CirculationHeater of the main Digester

T P 4’’

HV-51(2)Top Circulation Isolation − Isolation valve forliquor being sent back to High Pressure Feederfrom Impregnation Vessel. Special male flanges

O/O 14’’

HV-52(2)Top Circulation Isolation − this valve isolateshigh pressure feeder from the ImpregnationVessel. Installed on main chip/liquor feedline

O/O 14’’

1. Dual vessel only.CODE: P = Primary selection, S = Secondary selection, T = Throttling, O/O = On/Off

10B −22

KAMYRTAG#

KAMYR CONTINUOUS DIGESTER FISHER CONTROL VALVEPRODUCT DESIGN

Application Description ControlFunction

V150 V200 V300 V500 CV500 DSV 8580TypicalValveSized

HV-54Top Circulation PressurizationThis valve supplies cooking liquor to topcirculation line to pressurize it before pumping

T P 3’’

HV-62(1)(2)Bottom Circulation IsolationThis valve isolates Impregnation Vessel. It ison the discharge line feeding the digester

O/O 12’’

HV-65(1)Impregnation Vessel Cooling Liquor Flowfiltrate from blow line is added to theImpregnation Vessel to bottom dilution zone

T P 6’’

HV-81(2) Blow Line IsolationValves isolates digester blow line

O/O 10’’

HV-87Blow Line DilutionBlow line dilution and pressurization line priorto opening HV81

T P 2’’

HV-90A Blow Line Isolation (A)Valve isolates blow line A from blow line B

T P 8’’

HV-90B Blow Line Isolation (B)See HV90A

T P 8’’

HV-120(2) Sample Valve (Digester)This valve is used to monitor pulp quality

O/O 1-1/2’’

HV-120B(2) Sample Valve (Blow Line)See HV120A

O/O 1-1/2’’

QV-27 White Liquor to Sand SeparatorThis valve is a purge for the Sand Separator

O/O P 1’’

KV-8A(1)Trim Liquor Switching − This valve extractsliquor through screens to Bottom CirculationHeaters.

O/O P 6’’

KV-8B(1) Trim Liquor SwitchingSame as KV8A

O/O P 6’’

KV-16A

Digester Extraction SwitchingThese valves provide screened liquorextraction from upper wash zone to FlashTank No. 1

O/O P 8’’

KV-16B Digester Extraction SwitchingSee KV-16A

O/O P 8’’

KV-16C Digester Extraction SwitchingSee KV-16C

O/O P 8’’

KV-16D Digester Extraction SwitchingSee KV-16A

O/O P 8’’

KV-19A

Modified Cooking Extraction SwitchingThese valves extract liquor in the modifiedcooking zone and send it to the ModifiedCooking Heater

O/O P 8’’

KV-19B Modified Cooking Extraction SwitchingSee KV-19A

O/O P 8’’

KV-19C Modified Cooking Extraction SwitchingSee KV-19A

O/O P 8’’

KV-19D Modified Cooking Extraction SwitchingSee KV-19A

O/O P 8’’

KV-20A Wash Extraction Switching − This valveextracts wash water to the wash heater

O/O P 8’’

KV-20B Wash Extraction SwitchingSee KV20A

O/O P 8’’

KV-24(2) Sand Separator Dump ValveThis valve is known as the pocket valve

O/O

1. Dual vessel only.CODE: P = Primary selection, S = Secondary selection, T = Throttling, O/O = On/Off

10B −23

KAMYRTAG#

KAMYR CONTINUOUS DIGESTER FISHER CONTROL VALVEPRODUCT DESIGN

Application Description ControlFunction

V150 V200 V300 V500 CV500 DSV 8580TypicalValveSized

KV-60A(1)

Bottom Circulation Return SwitchingThese valves extract liquor from digester andsend it to bottom circulation heaters. *Digesterswitching valve

O/O P 14’’

KV-60B(1) Bottom Circulation Return SwitchingSee KV-60A

O/O P 14’’

KV-60C(1) Bottom Circulation Screen BackflushThis valve works with KV-60A

O/O P 6’’

KV-60D(1) Bottom Circulation Screen BackflushThis valve works with KV-61C

O/O P 6’’

LV-19M.C. Heater Condensate LevelThis valve controls level in Modified CookingHeater. *Not shown on schematic

T P 1.5’’

LV-20Wash Heater Condensate LevelThis valve controls Wash Heater Level. *Notshown on schematic

T P 2’’

LV-60AB.C. Heater “A” Condensate LevelThis valve controls level in Bottom CirculationHeater A. *Not shown

T P 1.5’’

LV-60B B.C. Heater “B” Condensate LevelSee LV60B except heater B

T P 1.5’’

LV-60C B.C. Heater “C” Condensate LevelSee LV60A except heater C

T P 1.5’’

LV-81 Condensate Flash Tank LevelNot shown on schematic

T P 3’’

CV-80A Condensate Conductivity to Tank*Not shown on schematic

O/O P 6’’

CV-80B Condensate Conductivity to Dump*Not shown on schematic

O/O P 6’’

CV-80C Water Conductivity to Dump*Not shown on schematic

O/O P 3’’

1. Dual vessel only.CODE: P = Primary selection, S = Secondary selection, T = Throttling, O/O = On/Off

10B −24

KAMYRTAG#

KAMYR CONTINUOUS DIGESTER FISHER CONTROL VALVEPRODUCT DESIGN

Application Description ControlFunction

V150 V200 V300 V500 CV500 8580 EZTypicalValveSized

LV-23 First Stage Backflush Tank Level O/O P 4’’

LV-24AFirst Stage Filtrate Tank (Bypass)This valve takes black liquor from FiltrateTank to weak black liquor storage

T P 6’’

LV-24BFirst Stage Filtrate Tank (Makeup)This valve takes wash water to add to weakblack liquor from Filtrate Tank

T P 6’’

LV-33 Second Stage Backflush Tank LevelThis valve controls level in Backflush Tank

O/O P 4’’

LV-34Second Stage Filtrate Tank Level MakeupThis valve controls level in second stageFiltrate Tank

T P 6’’

QV-22 First Stage Backflush O/O P 10’’

QV-32 Second Stage Backflush O/O P 10’’

PV-23 First Stage Backflush Tank PressureControls pressure in Backflush Tank No. 2

O/O P 1/2’’

PV-33 Second Stage Backflush Tank PressureControls pressure in Backflush Tank No. 2

O/O P 1/2’’

FV-27First Stage Wash FlowThis valve controls flow of first stage washin Diffusion Washer

T P 8’’

FV-28 Wash Water Flow for Float Out T P 2’’

FV-37Second Stage Wash FlowThis valve controls flow of filtrate to secondstage wash

T P 8’’

HV-21A(2) Blow Line IsolationValve isolates blow line from Diffusion Washer

O/O 16’’

HV-21B(2)Blow Line IsolationValve isolates blow line from highdensity storage

O/O 8’’

HV-22First Stage Extraction FlowThis valve controls flow of filtrate from firststage to Filtrate Tank

T P 8’’

HV-30 Second Stage Wash Isolation O/O P 8’’

HV-32Second Stage Extraction FlowThis valve controls filtrate flow from secondstage to Filtrate Tank

T P 10’’

TV-13Filtrate to Digester TemperatureThis valve control temperature ofwasher filtrate to Digester

T S P 6’’

1. Dual vessel only.TCODE: P = Primary selection, S = Secondary selection, T = Throttling, O/O = On/Off

www.Fisher.com

Chapter 11

Black Liquor Evaporator & Concentrator

The evaporator/concentrator system serves as abridge between the pulp mill and powerhouse.This is the first step in reclaiming spent cookingchemicals. The evaporator receives weak blackliquor from the pulp washers (or continuousdigester) and concentrates the solution byevaporating a large portion of the water content.The concentrated black liquor is then sent to thepowerhouse for combustion in the recovery boiler.A portion of the water content must be removed tomaintain safe and efficient combustion in therecovery boiler. Although various methods existfor this process, the primary purpose ofeconomical evaporation of water is a commongoal.

Multiple-Effect EvaporatorThe most common method, referred to as amultiple-effect evaporator, uses a series ofevaporator bodies to remove water from weakblack liquor. These evaporator bodies typicallyreceive weak black liquor at 12 - 15% solidsconcentration and evaporate a portion of the waterto raise the solids concentration to 50 - 60%.“Solids” refers to the organic wood constituentsand inorganic cooking chemicals. The product isreferred to as heavy or strong black liquor.

The primary advantage of multiple evaporatorbodies is steam economy. A series of evaporatorbodies makes it possible to remove 4 - 6 poundsof water per pound of motive steam used. This isaccomplished by connecting the bodies in seriesso the vapor generated from one evaporatorbecomes the steam supply for the next evaporatorin the series. Effects are numbered in order of thesteam flow. Weak black liquor starts at the lasteffect, thus moving in a counter-current flow tosteam. Although the number of effects may vary,

six, or a sextuple-effect, is most common. Capitalcost usually offsets any increase in steameconomy if more effects are used.

A typical sextuple-effect evaporator set is shownin figure 11-1. Each individual evaporator consistsof a heating element and a vapor head. Eachheating element consists of a tube bundle withupper and lower tube sheets. Liquor flows on theinside of the tubes and steam on the outside of thetubes. Some designs use plate-type heatingelements instead of tubes. In either design, thetransfer of heat causes the liquor to boil and thevapor that forms is carried to the next effect tocontinue the evaporation process. Condensateformed by condensing steam vapor is alsoremoved.

Motive steam is fed to the first effect and weakblack liquor feed split between the 5th and 6theffects. The black liquor increases in solidscontent and boiling temperature as it progressestoward the first effect. In order for boiling to takeplace, the pressure on the liquor side of the tubesmust be less than the pressure on the steam side.Thus, the pressure must be different in each effectand decrease from a high in the first effect of 30 -40 psig to a low in the sixth effect of 20 - 25inches of Mercury (Hg) vacuum. Maintaining thispartial vacuum is accomplished by piping vaporsfrom the sixth effect to a condenser and removingnon- condensible gases with an ejector or vacuumpump.

Evaporator TypesAs mentioned earlier, the most common type ofevaporator is the multiple-effect type. However,many variations of this concept are used. Thelargest installed base of evaporators is the risingfilm or long tube vertical (LTV) type (see figure11-1). This was the prevalent design until the late1970s. In this design, the liquor enters a cavity

11−2

E0894

Figure 11−1. Multiple-Effect Evaporator Six Effect / Rising Film / LTV

11−3

below the lower tube sheet. As it boils orpercolates, a thin film of liquor rises up the insideof the tube or plate. The liquor overflows the uppertube sheet and falls via a downcomer pipe to atransfer pump. The vapor exits via a centrifugalseparator and is piped to the next effect. Thisdesign provides high evaporation capacity at a lowcost, but is sensitive to scaling and plugging above50% solids.

Although the rising film design is predominant ininstalled base, the falling film design has becomeincreasingly popular since the early 1980’s and isnow the most common type of evaporator. Thisdesign looks like an upside-down rising filmevaporator (vapor dome at the bottom and thetube bundle extending upward). As its nameimplies, liquor is fed into the upper tube sheet areaand flows as a thin film down the inside of thetubes. The liquor collects in the lower dome and isdischarged from the evaporator body. Since theliquor flow is in the same direction as gravity andflowing in a thin film, higher heat transfercoefficients are realized. The higher coefficientsallow for lower temperature differentials (vapor vs.liquor) resulting in the ability to achieve highersolids concentration and less scaling than a LTVdesign. The main disadvantage is associated withthe high pumping cost of multiple-pass forcedcirculation employed on most falling film designs.

A number of alternative systems and variations tothe classical multiple effect system has emergedrecently in an effort to obtain higher solidsconcentrations, and reduce fouling of heatingsurfaces. Some of the variations to the classicalsequence involve changing the feed liquor inputlocation and using lower solids liquor to washsurfaces where higher solids liquor are normallymade. This extends the time between generalwashings or “boilouts”.

A recent alternative system involves combiningrising and falling film evaporator bodies in amultiple effect system. In this design, the first twoor three effects are falling film evaporators and thelast three or four effects are rising filmevaporators. This gives the advantage of pumpingenergy conservation on the “back end” wheresolids and scaling potential are lower and theresistance to fouling on the “front end” as solidsincrease.

Another system has emerged in recent yearsknown as mechanical vapor recompression(MVR). This system typically employs a singleevaporator body, a compressor, and heat

exchangers. This system reuses vapors by raisingthe temperature and pressure with a compressor.It is used mainly where steam supply isinadequate and electrical power is economical.

Auxiliary EquipmentVarious pieces of auxiliary equipment are requiredto support the operation of an evaporator set.Some of this equipment is briefly described below:

� Soap Skimming/Removal

Soap or tall oil soap is composed of fatty and resinacids found in wood products. During evaporation,the soap will not stay dissolved beyond 25 - 30%solids concentration. Failure to remove the soapresults in excessive foaming and a lower efficiencyfor the entire recovery cycle. Typically, liquorleaving the fourth effect is diverted to a skimmingtank where the soap is removed for processing.After the soap is removed, the liquor is transferredto the third effect to continue evaporation.

� Flash Tanks

Flash tanks are used to recover heat from flashingliquor or condensate to a lower pressure. Typicalflash tank locations are product liquor and cleansteam condensate from the first effect. The flashsteam is then used for process heating.

� Condenser and NCG Removal

As mentioned earlier, a condenser is used tomaintain a vacuum at the “back end” of theevaporator set. The condenser is connected to thevapor duct from the sixth effect. The condensedvapors from the sixth effect, referred to as foulcondensate, contains contaminants such as sulfurgases and black liquor organics. Thesecontaminants are removed by a steam strippingsystem since they create odor and pollutionproblems.

Non-condensible gases (NCG) such as hydrogensulfide, mercaptans, and carbon dioxide also tendto accumulate in the condenser. These gases areremoved with a steam or air fed ejector systemand sent to an incinerator. Failure to remove thesegases will limit an evaporator set by reducing theavailable vacuum and temperature differential.

� Foul Condensate Stripping

11−4

Figure 11-2. Falling Film Concentrator

E0895

Foul condensates are formed in both theevaporators and digesters. The primary reason forstripping foul condensate is pollution control. Acommon method of treatment involves feeding thecondensate to a stripping column or tower, whichis supplied with fresh steam. The steam tends toremove most of the contaminants and leavesclean condensate suitable for pulp washing. Thecontaminants are usually carried in a gaseousform to an incinerator.

ConcentratorsConcentrators are an extension of the evaporationof water from black liquor. As mentioned earlier,evaporators are typically limited to 50 - 60% solidsconcentration due to scale build-up of sodium saltson heating surfaces. Concentrators accept the 50- 60% solids from the evaporators and furtherconcentrate it to 65 - 80% solids. A typical fallingfilm concentrator is shown on figure 11-2.

� Direct Contact Concentrators

The first type of concentrator used in the pulp andpaper industry was a direct heated type. Eventhough it served the same basic purpose as

11−5

today’s concentrator, it was often referred as adirect contact evaporator. The two most widelyused types, the cyclone and cascade evaporators,utilized hot flue gas exiting the recovery boiler tofurther concentrate the black liquor. However, thedirect contact of flue gas and black liquor stripssulfur compounds from the liquor which results inair pollution and sulfur loss. Most mills haveeliminated direct contact evaporators in favor ofmore modern indirectly heated concentrationequipment.

� Forced Circulation Concentrators

To overcome the limitations of direct contactevaporators, indirectly heated concentrator bodiesare used for final liquor concentration. This type ofinstallation typically involves one or moreadditional effects ahead of the first effect in themultiple-effect evaporator set. The first effect isfed with a separate supply of steam andconcentrates the product liquor from the first effectof the multiple-effect evaporators.

The concentrator effects are generally used in aswitching arrangement such that one effect isconcentrating high solids black liquor while theother effect is concentrating lower solids liquor.This type of arrangement uses the lower solidsliquor to wash deposits left from concentratinghigh solids liquor.

� Falling Film Crystallizers

One of the most recent concentrator designs isreferred to as a falling film crystallizingconcentrator. As mentioned earlier, salt crystalstend to form once black liquor becomes saturatedat about 55% solids. In typical evaporators thesecrystals tend to stick to heat exchange surfacesand prohibit heat transfer. Crystallizers aredesigned to control crystal formation such thatnewly formed crystals will bond to crystals

contained in the recirculating liquor rather than theheating surfaces. This allows for extremely highsolids concentration (up to 80%) with reduced riskof fouling. The FFC concentrator can be used forall effects in an evaporator system wherecrystallization will occur.

� Other Designs

A couple of other designs may also be used asconcentrators. One type is a single concentratorbody with two or three separate sections. In thisdesign one section(s) is washed with weak liquorfrom the evaporators while the other section(s) areused to achieve final liquor concentration. Asecond type combines preheat, falling, and risingfilm sections in one body. These units typically useforced circulation and are complicated to operate.

Although figures 11-1 and 11-2 indicate many ofthe critical control valves, other general servicecontrol valves are required for the successfuloperation of an evaporator/concentrator set. Manyof the valves are on mill supply water, instrumentair, plant air, or low pressure steam heating lines.

Control Valve SelectionBlack liquor is a thick media which can be erosive,corrosive, or cause scaling problems in valves.Valves must be able to perform in both control andtight-shutoff applications, with specialconsideration give to valves where black liquor isthe thickest; typically before entering and exitingthe concentrator.

In applications where black liquor has lower solidscontent, the Control-Disk butterfly valve may beused. However, in most cases, the Vee-Ballsegmented ball valve is the primary valve ofchoice.

11−6

ValveTag #

EVAPORATORS/CONCENTRATORS FISHER CONTROLS VALVE PRODUCT DESIGN

Application Description ControlFunction

V150 V5008580/

Control-Disk

A81 E−BodyTypicalValveSize

LV-1 1st Effect Liquor Level T P S 6�

LV-2 2nd Effect Liquor Level T P S 6�

LV-3 3rd Effect Liquor Level T P S 6�

LV-4 4th Effect Liquor Level T P S 8�

LV-5 5th Effect Liquor Level T P S 8�

LV-6 6th Effect Liquor Level T P S 6�

LV-7 1st Effect Condensate Level T P S 4�

LV-8 2nd Effect Condensate Level T P S 4�

LV-9 3rd Effect Condensate Level T P S 6�

LV-10 4th Effect Condensate Level T P S 6�

LV-11 5th Effect Condensate Level T P S 8�

LV-12 6th Effect Condensate Level T P S 8�

LV-13 Clean Condensate Flash Tank Level T S P (ET) 4�

LV-14 Intermediate Product Liquor Flash Tank Level T P S 6�

LV-15 Soap Skimming Tank Level T P S 8�

LV-16 Foul Condensate Hotwell Level T S P (ED) 8�

LV-17 Concentrator Condensate Level T S P (ET) 4�

LV-18 Concentrator Liquor Level T P S 6�

LV-19 Product Liquor Flash Tank Level T P 6�

PV-1 Steam to 1st Effect T S P (EWD) 12�

PV-2 Flash Steam from Intermediate Product Flash Tank T P 8�

PV-3 Flash Steam from Clean Condensate Flash Tank T P 8�

PV-4 Steam to Concentrator T S P (EWD) 10�

PV-5 Vapor from Concentrator T P 14�

PV-6 Flash Steam from Product Liquor Tank T P S 8�

FV-1 Contaminated Condensate to Sewer O/O P 6�

FV-2 Soap to Processing T P S 6�

FV-3 Liquor Feed to 5th Effect T P 6�

FV-4 Liquor Feed to 6th Effect T P 6�

FV-5 Cooling Water to Condenser T P 18�

FV-6 Steam to NCG Ejector T S P (EZ) 2�

FV-7 NCG to Incinerator T P S 6�

FV-8 Feed Liquor to Concentrator T P 6�

FV-9 Contaminated Condensate to Sewer O/O P 6�P=Primary Valve ChoiseS=Secondary Valve ChiceT=Throttling ServiceO/O=On/Off Service

www.Fisher.com

Chapter 12

Kraft Recovery Boiler − Black Liquor System

The kraft recovery boiler is the heart of a complexseries of chemical processes referred to as thekraft recovery cycle. The two main functions of thekraft recovery boiler are to:

1. Reclaim digester cooking chemicals, sodiumand sulfur, in a suitable form for regeneration ofcooking liquors.

2. Provide efficient heat recovery and steamgeneration from combustion of organics in blackliquor fuel.

Since the kraft recovery cycle is far removed fromthe finished product of the paper mill, it sometimesdoes not receive the attention it deserves.However, efficient chemical and heat recoveryhave a critical impact in overall mill efficiency andprofitability.

Recovery Process OverviewAs mentioned earlier, the recovery boiler is amajor component of the kraft recovery cycle.However, other components perform importantfunctions in the cycle. For continuity, a briefdescription is given to indicate the role of therecovery boiler in the overall cycle.

The basic components of the kraft recovery cycleare:

Digester(s)

Wood chips are mixed with a solution of cookingchemicals called white liquor (sodium sulfide andsodium hydroxide). The contents are cookedunder pressure with steam to dissolve the glue-likelignin, which holds the wood fibers together. Aftercooking, the contents are blown into a holdingtank.

Washers

The holding tank contents, known as pulp orstock, are transferred to the washers where wateris used to wash residual cooking chemicals fromthe wood fibers. The wash water is sent to theevaporators and washed pulp to paper making.

Evaporators

The wash water, containing wood by-products andcooking chemicals, is transferred to theevaporators. This solution is commonly known asweak black liquor. Evaporators employ a series ofeffects or bodies using steam which evaporatesmuch of the weak liquor water content. The finalproduct is referred to as strong black liquor.

Recovery Boiler

Strong black liquor is transferred to the recoveryboiler for combustion. The combustion processburns the organics (for steam production) andtransforms the chemicals to a molten liquid knownas smelt. The smelt flows from the boiler into atank of water (or weak wash) and produces greenliquor.

Recausticizing Plant

The green liquor is transferred to a series ofcomponents which add lime to regenerate thecooking chemicals or white liquor. Clarified whiteliquor is then pumped to the digesters to begin thecooking process again.

Black Liquor PreparationAs stated in the overview, black liquor preparationfor the recovery boiler actually begins with theevaporators; however, other components (seefigure 12-1) presented in this guide also playimportant roles.

12−2

Two components commonly employed on older, orconventional boilers, are direct contactevaporators and black liquor oxidation systems.Direct contact evaporators may be the cascadeevaporator or cyclone evaporator type.

Black Liquor OxidationBlack liquor oxidation is the exposure of blackliquor to air (oxygen) to form more stable sulfidecompounds. This exposure prevents the releaseof hydrogen sulfide and mercaptans when theliquor is exposed to hot flue gas in the directcontact evaporators. Release of these gasesresults in sulfur loss and odor emission. Theoxidation process is commonly performed by ablower forcing compressed air into a liquor filledtank via a sparger ring.

Direct Contact EvaporatorsDirect contact evaporators are used for final liquorconcentration from 45-50% solids to 60-65%solids. They are so named because the boilerinduced draft fan pulls the hot flue gas into directcontact with the black liquor to evaporate waterprior to combustion.

The cyclone evaporator, commonly used withBabcock and Wilcox recovery boilers, consist of acylindrical vessel with an opening which allowsflue gas to enter tangentially. Black liquor issprayed into the swirling gas to allow mixing andevaporation of water.

The most common type of evaporator for the pasttwo decades has been the falling film or cascadeevaporator. This evaporator consists of a rotatingassembly of hot plates or tubes that are alternatelysubmerged and then exposed to hot flue gas.These tend to be more efficient and versatile.

Due to the high flue gas temperatures and thepresence of black liquor fuel, the potential for a fireexist in both types of evaporators. A commonmethod employed to combat this potential is theinjection of steam to displace oxygen and smothera fire.

Low Odor DesignAs mentioned earlier, both black liquor oxidationand direct contact evaporators are employed onolder or conventional type boilers. However, mostmodern designs referred to as low odor type use

indirectly heated concentrators (similar toevaporators) to raise the black liquor solidsconcentration to the 65-70% range. Similar to thefalling film evaporator, concentrators eliminate theneed for direct contact evaporators, which, in turn,eliminates the need for black liquor oxidation.Lower sulfur and odor emissions are a result ofthe flue gas and liquor having no direct contact.This design also requires additional boilereconomizer section(s) to absorb the flue gas heatwhich had been removed in the direct contactevaporators. The concentrator is more energyefficient and environmentally friendly than itspredecessors.

Precipitator AshFollowing preparation by concentrators or blackliquor oxidation, liquor is transferred to theelectrostatic precipitator. Ash, consisting primarilyof salt cake (Na2SO4 or sodium sulfate), iscollected from the flue gas and mixed with theliquor. In some older designs, referred to as wetbottom, the liquor fills the precipitator bottom andsalt cake falls directly into the pool. Since thispresents a potential fire hazard and a source forodor emissions, most newer precipitators have thedry bottom design. In this design, chains or screwsconvey the dry ash into a liquor filled sluice tank.

Salt Cake Mix TankBefore introduction to the recovery boiler furnace,the liquor is mixed with more ash and salt cake inthe blending tank or salt cake mix tank. Salt cakeaddition is required due to small chemical lossesoccurring in the recovery cycle. Salt cake istypically added from two sources. One source isvia fallout hoppers below the steam generatingand economizer sections of the boiler. A secondsource is purchased make-up. Direct steamheating of the tank is typically used to maintainliquor viscosity at suitable values for pumping.

In order to reduce emissions, some mills haveeliminated salt cake as a soda makeup chemicalbecause of its high sulfur content. Caustic soda(NaOH) and soda ash (Na2CO3) have beensubstituted since they are sulfur free chemicals.

Liquor DivertBlack liquor is typically at 65-70% solids before itis sent to the recovery furnace for combustion.Density, or percent solids, is usually measuredbetween the liquor heater and liquor guns via

12−3

magnetic flow meters. If solids drop below 60%,auxiliary fuel is added due to the potential of asmelt/water explosion or bed “blackout”. If solidsfall below 57-58%, the liquor is diverted from thefurnace to the mix tank until solids reach anacceptable level. Each individual liquor feed linetypically has a valve connected to emergencyshutdown interlocks. These valves snap close ona trip or shutdown signal.

Black Liquor HeatingA final preparation stage for the liquor involvesheating. Liquor is typically heated to 230°F-250°Fto impart the desired viscosity and burningcharacteristics before spraying into the furnace.Many older designs employ direct steam heaters,but most newer designs are using indirect steamheating. The indirect heating does not add watercontent to the liquor which is a safety andefficiency consideration. Recirculation systemsand steam desuperheating are often used withindirect heating.

Liquor Flow/Pressure ControlA common method used to control the flow orpressure of black liquor to the furnace isrecirculation to the mix tank. This allows thenozzle pumps to run at a constant speed and keepthe liquor moving to avoid potential plugging oftransport piping.

Auxiliary FuelBlack liquor is not used as a fuel to start-up arecovery boiler. Natural gas or fuel oil is typicallyused to bring the boiler up to a prescribedtemperature before black liquor fuel is introduced.This is done primarily as a safety considerationdue to the potential of a smelt-water explosion inthe lower furnace.

Kraft Recovery BoilerWhile recovery boiler design considerations varyamong each manufacturer, their basic two-foldpurpose of chemical recovery and steamproduction is common to all. Since the primaryfunction is chemical recovery, black liquor flow tothe furnace is a constant and steam production isa by-product. This requires the recovery boiler

steam outlet header to be piped to a commonheader with a power boiler steam outlet. Swingingsteam demands due to mill processes areaccommodated by manipulating fuel to increasesteam production of the power boiler.

Combustion AirAir required for combustion in the furnace isintroduced separately from the black liquor.Ambient air is forced into the boiler through an airheater via the forced draft fan(s). The air heatertypically uses steam coils to heat the incoming air.Most modern designs introduce air at three levels:primary, secondary, and tertiary. These variouslevels are used to ensure chemical reduction,complete combustion of organics, and propershape of the smelt bed. The primary air ports arelocated a few feet above the hearth and carry theresponsibility to provide as low a velocity aspractical while still supplying between 50-65% ofthe total air requirement. Secondary and tertiaryports establish higher velocities to ensurecomplete mixing and combustion of the unburnedgases. Combustion flue gas is pulled from thefurnace section to the convective section of theboiler via induced draft fan(s). This creates aslightly negative pressure inside the boiler. Thisaction prevents hot gases from leaving boileropenings and is commonly referred to as“balanced draft”.

Black Liquor CombustionBlack liquor is introduced into the recovery furnacevia nozzles or liquor guns. The guns produce aspray of coarse droplets exposed to hot flue gas.Depending on the manufacturer, the guns may bestationary or oscillating, and spray the liquor onthe walls or into the center of the furnace. Theflow of this black liquor to the guns is controlled bya valve. When selecting a control valve for thisapplication, it is crucial to select proper materialsdue to corrosion. As the organics burn and releaseheat to the flue gas, the remaining char, consistingof the sodium and sulfur cooking chemicals, fallsto the smelt bed on the furnace floor.

Sootblowers

Application Summary:The efficiency of a fossil-fuel boiler is highlydependent on the heat transfer effectiveness ofthe boiler tubes. These tubes are fairly delicate,and hot spots (due to soot buildup cannot) be

12−4

tolerated as a leak could result. A cleaningprocess for the boiler tubes is needed even whilethe boiler is in operation. This process is calledsoot blowing.

Process:When firing fuels such as coal, oil, biomass orother waste products fouling of the boiler tubesbecomes a concern. Deposits from thecombustion process can collect on the heatexchanging tubes reducing thermal efficiency andcan cause operational difficulties. In order to keepthe unit operating, an online cleaning method mustbe used. This is usually accomplished by usingwhat are called sootblowers.

Most sootblowing systems utilize either air orsteam. Widespread use of water has been limiteddue to the possibility of thermal shock on the tubebanks. Air or steam systems each have their ownadvantages, but one is not considered better thanthe other.

Air systems have much simpler pipingarrangements. This is due to the elimination ofcondensate drain piping. The number ofcompressors, compressor capacity and thesootblower flow requirement; however, limits thissystem.

Steam systems have an advantage in terms ofexpansion. The supply of steam (typicallyremoved after the primary superheater) is virtuallyunlimited, but leads to additional maintenanceconcerns related to the numerous valves required.Also, as stated above, the steam systems requireadditional piping to address the possibility ofcondensate in the steam lines.

As high pressure air or steam is required toremove the deposits from the boiler tubes, thecontrol valve must be able to withstand highpressures. Steam systems present a greaterchallenge due to the combination of high pressureand temperature. Because of the high inletpressure, downstream pressure and pipe size, thevalve must also withstand issues with noise andvibration. As the sootblowers operateintermittently, tight shutoff (class V) is required forvalve trim protection and when using steam,maintaining unit efficiency. These valves modulateover a wide range of flows and are required tomaintain downstream pressure.

Design Considerations and ServiceConditions:

� High pressure class rating due to thepressures and temperatures.

� Tight shutoff so valves don’t leak valuablesteam.

� Large pressure drops can create noise,vibration, and excessive wear to trim.

� Cyclic conditions as valve are operatednumerous times a day.

Typical Process Conditions:� P1 = 800 - 1400 psig

� P2 = 0 psig

� T = 400 - 800°F

12−5

Typical Specification:

ES flow-down for on/off sootblowers.

To conserve energy, mills have moved tothrottling sootblower valves. Use an ED.

Tight shutoff not needed as sootblower nozzleshave tight shutoff.

Model:

ES for on/off, flow down, quick opening cage,Class VED for throttling, =% cage, flow down, Class IIThese units provide high pressure andtemperature capability along with the tightshutoff required.Drilled-hole cage (flow-up)

Body: WCC

Trim: Quick opening cage or Whisper cage for noise attenuationCage: S31600 CoCr-A seat and guidePlug: 316 CoCr-ASeat: R30006 (alloy 6)Seat Ring Retainer: R30006 (alloy 6)Stem: Nitronic 50. Optional oversized stem or oversized VSC

Bonnet:: PTFE packing

Actuator: 667 spring-and-diaphragm or 585C piston

Positioner: DVC6010 c/w performance diagnostics

Fisher Engineered Specification:

The Fisher ES and ED cage-guided globe valvesprovide maximum stability and ruggedness in highpressure drop applications. The ES is offered withan unbalanced plug and the ED with its balancedplug to minimize actuator thrust requirements.Both designs have hardened trim for superiorerosion resistance. The plug and stem assemblyis reinforced through the use of high tensilematerials and oversized valve stem connection.

High noise levels are often present in highpressure drop, high flow steam applications.Noise attenuation can be achieved withengineered valve trim combined with an inlinediffuser for additional noise reduction whenrequired. Typically Whisper Trim� I and a Fisher6011 downstream diffuser are the most practicaland economical solution; however, there are otheroptions available that allow for full attenuation atthe valve such as the Fisher Whisper Trim III orWhisperFlo� trim. There are many factors toconsider such as capacity, turndown, line size,and overall economics when selecting anappropriate solution.

Chemical Reduction

Chemical reduction occurs via the char bed at the

floor of the recovery furnace. The molten liquidbed typically operates at temperature of 1600°F -2500°F. The introduction of air via the primary airports provides the oxygen required to burn thecarbon and reduce the sulfate to sulfide. Themolten smelt, consisting primarily of Na2CO3 andNa2S, flows by gravity to the dissolving tank. Thedissolving tank is filled with weak wash to cool thesmelt and produce a solution suitable for pumping.The solution formed, known as green liquor, isthen transferred to the recausticizing area foraddition of lime and regeneration of white liquor.

The dissolving tank is vented to the atmospherevia a tall stack. Since the sodium and sulfurcompounds in the smelt present a source of odoremission, the vented gas is often treated with ascrubber. Although the scrubbers vary in designand complexity, a solution of weak wash is oftenused as the scrubbing medium.

Control Valve Selection

Although table 12-1 indicates many of the criticalcontrol valves, other general service control valvesare required for the successful operation of arecovery boiler. Many of the valves are on millsupply water, instrument air, plant air, or lowpressure steam heating lines.

12−6

Table 12-1. Kraft Recovery Boiler / Black Liquor System Valve SelectionPROCESS FISHER VALVE PRODUCT DESIGN

ValveTag

Kraft Recovery Boiler / Black Liquor SystemV150/V300

V500 Control-Disk

A81 E-Body TypicalValve SizeApplication Description Control

Function

LV-1 BLOX Tank Level Control T P P 6�

LV-2 Sluice Tank Level Control T P P 6�

LV-3 DCE Level Control T P P 6�

FV-1 Black Liquor Emergency Divert O/O P S 6�

FV-2 Black Liquor Emergency Divert O/O P S 6�

FV-3 Black Liquor Recirculation Flow orPressure Control

T P 2�

FV-4 Auxiliary Fuel/Natural Gas T P 3�

FV-4 Auxiliary Fuel/Fuel Oil T P 2�

FV-5 Black Liquor Shutoff O/O P 6�

FV-6 Green Liquor from Dissolving Tank T S P 6�

FV-7 Weak Wash to Dissolving Tank T P 4�

FV-8 Weak Wash to Scrubber T P 2�

FV-9 Sootblower Steam O/O P S 4�

FV-10 Smothering Steam to Precipitator O/O S P 2�

FV-11 Smothering Steam to DCE O/O S P 2�

TV-1 Steam to Mix Tank T S P 2�

TV-2 Steam to Black Liquor Heater T S P 2�

TV-3 Steam to Air Preheater T S P 2�P=Primary Valve ChoiceS=Secondary Valve ChoiceT=Throttling ServiceO/O=On/Off Service

12−7

E08

93

Figure 12-1. Kraft Recovery Boiler Black Liquor System

12−8

www.Fisher.com

Chapter 13

Recausticizing and Lime Recovery

The recausticizing and lime recovery plant is thefinal step in the Kraft recovery process. It servesas a link between the Kraft recovery boiler and thedigester. The function of the recausticizing area isto convert the inorganic chemicals in the greenliquor from the recovery boiler dissolving tank towhite liquor for cooking chips in the digester. Thisprocess consumes lime and produces lime mud.The purpose of lime recovery is to convert the limemud back into lime for the recausticizing process.Proper control of the recovery and reclaim of thecooking chemicals is essential in the economicsuccess of a Kraft recovery mill.

RecausticizingAs mentioned earlier, the recausticizing processinvolves reclaiming the cooking chemicalscontained in the green liquor and converting orregenerating them to produce white liquor. Thewhite liquor is then used to cook wood chips in thedigester. A typical flow sheet is shown in figure13-1.

Dissolving Tank

Green liquor is produced in the recovery boilerdissolving tank by mixing weak wash and smelt.Smelt, consisting primarily of Na2CO3 and Na2S,is produced by burning black liquor in the recoveryboiler furnace. Weak wash, which is basicallywater, is a product of lime mud washing. Inaddition to the chemical components, the greenliquor contains impurities known as dregs whichconsist of unburned carbon and inorganicimpurities such as calcium and iron.

Valve: FV-9 Weak wash to dissolvingtank

� Typical process conditions:

— Fluid: Secondary condensate

— T = 176°F— P = 70 - 75 psig

— ΔP = 40 - 45 psig

� Typical valve selection:

— NPS 6 valves with alloy 6 scraper seatsdue to concerns with scaling. A SSTvalve with an alloy 6 HD seal and alloy 6bearings should be used in thisapplication. Depending upon processconditions, the Control-Disk could act asa great alternative.

Green Liquor ClarifierThe dregs, which cause the green color, areimpurities that must be removed from the greenliquor. These fine particles are removed bypumping the green liquor to a sedimentation tankor clarifier. Since the density of the dregs isgreater than the green liquor, settling of the dregsby gravity occurs. A slow moving rake pulls thematerial to a discharge cone in the center of theclarifier where it is concentrated and removed.The clarified liquor exits via overflow piping to astorage tank.

Dregs WashingBefore disposing of the dregs, they must bewashed to recover any residual cookingchemicals. This is important from an economicaland environmental aspect. Most mills employ apre-coat filter to wash the dregs. The systemconsists of a rotating cylindrical filter in a vat. Lime

13−2

mud is admitted to precoat the outside of the filter.A vacuum is maintained on the inside of the filterwith a vacuum pump. The dregs slurry is pulledthrough the lime mud and filter media. A knifeblade removes the lime mud as it becomessaturated with dregs. The lime mud/dregs arehauled to a landfill and the clear filtrate is recycledto the green liquor clarifier.

Valve: LV-1 Dregs Precoat Filter Level� Typical process conditions:

— Fluid: Lime mud

— T = 212°F— P = 55 - 65 psig

� Typical valve selection:

— These are typically NPS 2 or NPS 3 SSTvalves with solid VTC (ceramic) internalsand alloy 6 bearings.

Valve: FV-3 Dregs Slurry Underflowfrom Green Liquor ClarifierPlease reference the lime mud underflowinformation below as this application closelyresembles its process.

SlakerThe slaker is the heart of the recausticizingoperation. At this point, clarified green liquor ismixed with lime to produce white liquor. Thereburned lime (CaO) from the lime kiln andmakeup lime are added and react with the water inthe green liquor to form calcium hydroxide(Ca(OH)2). This reacts with Na2CO3 in the greenliquor to form sodium hydroxide (NaOH) or causticand precipitate calcium carbonate (CaCO3) or limemud. A retention time of approximately fifteenminutes is allowed in the slaker. Recausticizingefficiency is improved by steam heating theincoming green liquor to near boiling and addingan agitator to the slaker.

Valve: FV-5 Green Liquor to Slaker� Typical process conditions:

— Fluid: Green liquor

— T = 212°F— P = 80 - 85 psig

— ΔP = 5 - 10 psig

� Typical valve selection:

— NPS 6 valves with alloy 6 scraper seatsdue to concerns with green liquor scaling.A SST valve with an alloy 6 HD seal andalloy 6 bearings should be used in thisapplication.

CausticizersThe retention time in the slaker is not enough toallow a complete reaction between the lime andgreen liquor. Causticizers consist of a series oftwo or more agitated tanks having a total retentiontime of 1-1/2 - 3 hours. The white liquor slurryusually flows by gravity from the slaker andthrough the causticizers.

White Liquor Clarifier and Lime MudWasherThe white liquor clarifier is essentially the same asthe green liquor clarifier described earlier. Thewhite liquor slurry is pumped in from the lastcausticizer and the lime mud solids (CaCO3) settleto the bottom of the clarifier due to densitydifferential. The lime mud, at 35-40% suspendedsolids, is raked to a center discharge cone whereit is concentrated and removed. The clarified whiteliquor containing sodium hydroxide and sodiumsulfide overflows and is pumped to the digester forcooking wood chips.

The lime mud underflow from the white liquorclarifier must be washed to recover residualcooking chemicals. The lime mud washer is verysimilar to a white liquor clarifier with the possibleexception of multiple compartments. The feed tothe washer is from a mix tank, which acceptsfiltrate streams from the lime mud filter and limekiln scrubber in addition to the lime mudunderflow. The washed lime mud is removed at45-50% suspended solids and sent to storage.The overflow, referred to as weak wash, is sent toweak wash storage and primarily used in therecovery boiler dissolving tank.

Valves: FV-6/FV-8/FV-16/FV-19 LimeMud Underflow

� Design Considerations and ServiceConditions

— Lime mud is extremely erosive anddifficult to handle due to fine particulateand high solids concentration.

— The underflow valve is throttled to controlmud density which directly impacts the

13−3

operation and efficiency of the lime kiln.An accurate and reliable control valve isrequired for optimum performance.

— The underflow valve must beappropriately sized to ensure the mudlevel in the tank never reaches the filtersocks. The filter cannot operate properlyif this occurs.

� Typical Process Conditions:

— Fluid: Lime mud

— P1 = 20 - 50 psi

— P2 = 0 - 5 psi

— T = ∼175 °F— Q = 100 - 300 USGPM

— SG = 1.36

� Typical Valve Selection:

— NPS 3 - NPS 6 V500 ANSI 150/Reverseflow/Trim #4 for erosive service

— Plug: VTC Ceramic on alloy 6 Hub

— Seat: Solid VTC Ceramic

— Shaft: Oversized, 17-4PH Stainless Steel

— Retainer: Solid alloy 6 with Ceramic bore

— Bearings: Sealed alloy 6 construction

— Packing: PTFE

— Actuator: 2052, fail closed

— Positioner: FIELDVUE DVC6200 withPerformance Diagnostics

White Liquor and Lime Mud PressureFiltersAlthough white liquor clarifiers and lime mudwashers are predominate in installed base, thetrend in recent years has been to substitutefiltration equipment. This is possible because ofthe relatively large size of lime mud particles. Atypical filtration system is shown in figure 13-2.

The white liquor pressure filter performs the samefunction as the white liquor clarifier. The vessel isdivided into two compartments by a tube sheet.The tube sheet supports a number of perforatedtube filter elements. Each perforated tube iscovered with a polypropylene filter sock. The whiteliquor slurry must pass through the filters to reachthe upper compartment. As lime mud builds up onthe filters backflushing is required to restorenormal operation. Backflushing is accomplished byrecirculating the feed white liquor back to thecausticizer. This allows the level to drop and the

air cushion at the head of the vessel to expandand force the liquor into the socks. This knocksthe lime mud from the sock filters. The lowerportion of the pressure filter acts as a settling zonefor the lime mud following backflushing. Clarifiedwhite liquor is removed from the uppercompartment above the filter elements.

The lime mud pressure filter performs the samefunction as the lime mud washer and the principleof operation is the same as the white liquorpressure filter. The lime mud is removed from thebottom of the unit and the filtrate known as weakwash is removed from the upper compartmentabove the filter elements.

Valve: FV-17 Lime Mud to Filter� Typical process conditions:

— Fluid: Lime mud

— T = 176°F— P = 50 - 60 psig

� Typical valve selection:

— These valves can range from NPS 4 toNPS 10. A SST body with 316 CRPLball, alloy 6 hard faced seat and alloy 6bearings should be used in thisapplication.

Valve: FV-18 Lime Mud Recirculationto Lime Mud Mixer

� Typical process conditions:

— Fluid: Lime mud

— T = 176°F— P = 50 - 60 psig

� Typical valve selection:

— NPS 2 or NPS 3 SST valves with solidVTC (ceramic) internals and alloy 6bearings.

Lime Mud FilterLime mud from storage at 45-50% suspendedsolids is pumped to a precoat filter for dewatering.Dilution water is added at the intake of the transferpump to dilute the solution to 25% suspendedsolids. The lime mud slurry is pumped to a vatcontaining a rotating filter. The lime solids build toa sufficient thickness on the filter and aredewatered by means of a vacuum maintained onthe inside of the filter with a vacuum pump. A

13−4

fresh water spray may also be used for furtherwashing of cooking chemicals from the lime mud.The dewatered lime mud is removed with ascraper blade and the filtrate sent to a mix tankfeeding the lime mud washer (or pressure filter).The operation is very similar to the dregs precoatfilter.

Valve: FV-10 Dilution water for limemud transfer

� Typical Process Conditions:

— Fluid: Clean condensate

— T = 176°F— P = 30 - 40 psig

— ΔP = 5 - 10 psig

� Typical valve selection:

— NPS 2 carbon steel valves with 316CRPL ball and TCM plus seat to achieveclass VI shutoff. Depending upon processconditions, the Control-Disk could serveas a great alternative.

Valve: FV-11 Lime Mud to Filter� Typical process conditions:

— Fluid: Clean condensate

— T = 176°F— P = 50 - 60 psig

— ΔP = 5 - 10 psig

� Typical valve selection:

— These valves can range from NPS 4 toNPS 10. A SST body with 316 CRPL ball,alloy 6 hard faced seat and alloy 6bearings should be used in thisapplication.

Lime RecoveryLime recovery, lime reburning, or calcining areterms commonly used to describe this portion ofthe chemical recovery cycle. As mentioned earlier,the lime recovery area accepts the lime mud(CaCO3) from the lime mud filter and converts it tolime (CaO) for use in the slaker. This solves anyproblems associated with lime mud disposal andalso has a significant economic impact byreducing the need to purchase lime. With lime

recovery, purchased lime is only required to makeup system losses.

The conversion of the lime mud to lime is usuallyaccomplished in a rotary lime kiln. A rotary kiln is alarge steel tube lined with refractory bricks. Thecylinder is mounted on an incline, supported onrollers, and rotated at a slow speed with anelectric motor/gear reducer set.

The lime kiln accepts the lime mud from the limemud filter at 60-70% solids. It is conveyed fromthe upper to lower end by the rotation of the kiln. Aburner, utilizing oil or gas, is installed at the lowerend. The heat of the flame evaporates theremaining moisture and yields lime and carbondioxide from the lime mud. This process alsocauses the lime mud powder to agglomerate intopellets which can be handled.

The lime product is conveyed to a storage silo foruse in the slaker. A scrubber is also used toalleviate the dusting and pollution problemassociated with the exiting flue gas.

Valve: FV-12 Natural gas to lime kilnburner

� Typical process conditions:

— Fluid: Fuel gas

— T = 86°F— P = 50 - 60 psig

— ΔP = 1 - 3 psig

� Typical valve selection:

— NPS 4 carbon steel valves with 316CRPL ball and TCM plus seat to achieveclass VI shutoff. Depending upon processconditions, the Control-Disk could serveas a great alternative.

Valve: FV-13 Fuel Oil to lime kilnburner

� Typical process conditions:

— Fluid: Fuel Oil

— T = 260°F— P = 210 - 220 psig

— ΔP = 20 - 25 psig

� Typical valve selection:

— These valves are usually NPS 2 carbonsteel valves with SST internals.

13−5

C08

14 /

IL

Figure 13-1. Recausticizing and Lime Recovery

13−6

Figure 13- 2. White Liquor and Lime Mud Pressure Filters

C0813 / IL

13−7

Control Valve Selection

PROCESS FISHER VALVE PRODUCT DESIGN

ValveTag #

Recausticizing and Lime RecoveryV150/V300

V500 Control-Disk

A81 E−BodyTypicalValveSizeApplication Description Control

Function

FV-1 Steam to Green Liquor Heater T S P 1�

FV-2 Green Liquor from Dissolving Tank T S P 6�

FV-3 Dregs Slurry Underflow from Green Liquor Clarifier T P 2�

FV-4 Lime Mud to Dregs Filter O/O P 1-1/2�

FV-5 Clarified Green Liquor to Slaker T P 6�

FV-6 Lime Mud Slurry Underflow from White Liquor Clarifier T P 3�

FV-7 Clarified White Liquor to Digester T P S 6�

FV-8 Lime Mud Washer Underflow T P 3�

FV-9 Weak Wash to Dissolving Tank T P S 6�

FV-10 Dilution Water for Lime Mud Transfer T S P 2�

FV-11 Lime Mud to Filter T P 4�

FV-12 Natural Gas to Lime Kiln Burner T P S 4�

FV-13 Fuel Oil to Lime Kiln Burner T S P 2�

FV-14 White Liquor Recirculation to Causticizer O/O P 18�

FV-15 White Liquor Feed to Pressure Filter O/O P 18�

FV-16 Lime Mud Slurry Underflow from White Liquor PressureFilter

T P 6�

FV-17 Lime Mud Feed to Pressure Filter O/O P 18�

FV-18 Lime Mud Recirculation to Lime Mud Mixer O/O P 18�

FV-19 Lime Mud Pressure Filter Underflow T P 6�

LV-1 Dregs Precoat Filter Level T S P 2�

LV-2 Causticizer Level T P 6�

LV-3 Lime Mud Filter Level T S P 4�

LV-4 Lime Kiln Scrubber Level T P 2�P=Primary Valve ChoiceS=Secondary Valve ChoiceT=Throttling ServiceO/O=On/Off Service

13−8

www.Fisher.com

Chapter 14

Bleaching and Brightening

Bleaching and BrighteningIf the kraft pulp being produced is going to bebleached, then pulping is allowed to proceed until90% or more of the lignin originally found in thewood is removed; however, the small amount oflignin that is left gives unbleached pulp itscharacteristic light brown color. Bleaching is theway to remove the residual lignin while causingminimal damage to the fibers and produce whitepulp.

The differences between bleaching andbrightening are as follows:

Bleaching — This process removes the lignin andis used to increase the brightness of chemicalpulps.

Brightening — This process converts chemicalgroups in lignin to forms that do not darken pulp,thereby making it whiter. This process is used formechanical or chemi-mechanical pulps that stillcontain vast amounts of lignin.

The Process

Oxygen DelignificationThis process is typically found midway betweenpulping and bleaching. Oxygen can be used insodium hydroxide (NaOH) solution under pressureto delignify (i.e. remove lignin for the wood)unbleached pulp (see figure 14-1). Up to one-half

Figure 14-1. Oxygen Delignification Diagram

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

14−2

of the remaining lignin can be removed; furtherdelignification would cause excessive cellulosedegradation. Lignin removal in oxygendelignification significantly reduces the amountand cost of the bleaching which follows, andreduces the load on effluent treatment facilitiesbecause the filtrate from the post-oxygen washersgoes back to the brown stock washers and thechemical recovery system.

Oxygen delignification is typically done at amedium to high consistency. Although goodresults are obtained from both high andlow-consistency systems, medium-consistency isfavored because of its lower capital costs andinherently safe operation.

In a medium-consistency system, pulp comingfrom the brown stock washer at about 10-14%consistency is delignified. It is then preheated in alow pressure steam mixer and pumped throughone or more medium-consistency gas mixers to anupflow pressurized reaction tower. Steam andoxygen are added upstream of the consistencymixer or added directly to the pulp slurry. Themost recent mills have two consecutive stages inorder to improve the chemical efficiency of thetreatment.

BleachingPulps that have or have not been delignified arebleached in a continuous sequence of processstages, typically three, four, or five. The chemistrychanges in each stage and the pulp is washedbetween stages.

The common bleaching chemicals andnomenclature are:

Chlorination (C): Reaction with elemental chlorinein an acidic medium (Cl)

Alkaline Extraction (E): Dissolution of reactionproducts with sodium hydroxide (NaOH) - ref

Chlorine Dioxide (D): Reaction with chlorinedioxide in acidic medium (ClO2) - ref

Oxygen (O): Reaction with molecular oxygen athigh pressure in alkaline medium (O2) - ref

Hypochlorite (H): Reaction with hypochlorite inalkaline medium (ClO−) - ref

Peroxide (P): Reaction with peroxide in alkalinemedium - ref

Ozone (Z): Reaction with ozone in acidic medium(O3) - ref

For many years, chlorination was always the firststage of bleaching. However, since the 1990’s, ithas largely been replaced by chlorine dioxide toavoid the formation of dioxins. This is due to theenvironmental push for all pulp and paper mills tobe ECF, or elemental chlorine free. A few papermills are TCF, or totally chlorine free.

Conventional BleachingThe equipment is the most common aspect of thestages. This includes: a steam mixer to heat thepulp suspension with direct steam, a pump totransport the pulp, a chemical mixer to combinethe pulp with the aqueous bleaching agent, aretention tower to allow time for the bleaching tooccur, and a washer to separate the spentbleaching solution from the pulp.

Referring back to the common bleachingnomenclature, a typical bleach plant has aDEOPDED sequence, or a low-consistencychlorine dioxide first stage with one or twochemical mixers, an upflow tower, and a rotaryvacuum drum filter for pulp washing. The chlorinedioxide comes to the mixer at a solutionconcentration of about 10 grams per liter in coldwater. The pulp suspension is around 3.5%consistency and has been heated to about 140�F.After mixing, the pulp and chlorine dioxide go tothe retention tower to react for about 45 minutes.The pulp is then washed afterward.

Each bleaching stage has its own set of processconditions. The amount of bleaching agent,consistency of the pulp, pH, temperature, and timemay all vary in each stage. Pulp consistency isaround 3-4% in the chlorine dioxide stage andabout 10% in all subsequent stages. Thetemperature is the lowest in the first stage at140�F, and between 140-176�F, in the otherstages. How much bleaching agent requireddepends on which chemical, which stage in thesequence, and what kind of pulp. With chlorinedioxide, progressively less is used as you go alongthe sequence (figure 14-2).

The objective of bleaching is to remove theresidual lignin from the unbleached pulp. Chlorinedioxide is the preferred bleaching reagentworldwide. It is selective in dissolving residuallignin without degrading the cellulose andhemicelluloses. In a bleaching sequence, chlorinedioxide stages are always interspersed withalkaline extraction stages (see figure 14-2). This

14−3

Figure 14-2. Conventional Bleaching ProcessDrawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

alternating pattern of acidic and alkaline stageshelps to break down the increasingly smalleramounts of residual lignin, ultimately dissolving themajority of the lignin so it can be washed out ofthe pulp.

Pulp brightness only increases modestly in oxygendelignification and does not increase uniformlyacross the bleaching sequence. This is due to theaction that each chemical has on the lignin.Brightness increases substantially in the first andsecond chlorine dioxide stages and modestlyagain in the final chlorine dioxide stage (see figure14-2). The alkaline extraction stages do notchemically whiten the pulp, they actually darken.The alkaline stages are there to dissolve andremove the lignin which has already been brokendown by the chlorine dioxide. Kraft market pulpsare normally bleached to a final brightness of 90%or higher; however, the final brightness is basedsolely on customer’s standards and needs for themarketplace.

The filtrate flow in a bleach plant is counter-current, or opposite that of the pulp flow. In thiscase, there are actually two filtrate flows (seefigure 14-3). Filtrate from the final chlorine dioxidestage washer is used as shower water on the thirdchlorine dioxide stage washer, and its filtrate isused on the first chlorine dioxide stage washer.Some of the acidic filtrate from the first chlorinedioxide stage washer is used to dilute and controlthe consistency of the pulp effluent treatment.Chlorine dioxide is always used with other chlorinedioxide stages and not mixed with the alkalinestages.

The filtrate flows of the two alkaline stages areconnected in a similar fashion (see figure 14-4).Filtrate flow from the second extraction stage isused as shower water on the first extraction stageand the filtrate from the first extraction stagewasher goes to the effluent treatment. Unlike theyellowish acidic filtrate stream, the effluent fromthis process is brown due to the lignin.

FiberlineThe fiberline refers to the equipment andprocesses that the pulp travels through as it isbeing processed from chips to final bleached pulp(figure 14-1). Kraft pulp’s strength declines along afiberline. This loss is due to the degradation of thecellulose fibers during the pulping and bleachingprocess. Chip quality, appropriate pulpingequipment, and good operating practices are keyaspects to produce strong kraft pulps.

Other quality aspects of bleaching includes pulpcleanliness, i.e. no dark particles caused by bark,pitch, small stones, and a minimal amount ofextractives such as resins, which cause problemsin the papermaking process. Fiber bundles, knownas shives, have to be removed and reconvertedinto good pulp fibers. Metal ions must be removedvia chelation, or the binding of ions similar tocalcium being removed by a water softener.

Totally Chlorine FreeAs previously mentioned, a few paper mills areTCF. This process eliminates organo-chloridecompounds from bleach plant filtrates and makesit possible to recycle these filtrates back to

14−4

Figure 14-3. Chlorine Dioxide (ClO2) Filtrate FlowDrawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

Figure 14-4. Alkaline Filtrate Flow

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

chemical recovery plants. TCF sequences usevarious combinations of oxygen-alkali chemistry,hydrogen peroxide, and ozone. This process tendsto be more expensive, have lower brightness, andis susceptible to strength loss problems.

Mechanical PulpingThis process dissolves most of the lignin in theoriginal wood. High-yield mechanical pulps cannotbe bleached the same way because they containtoo much lignin. Bleached or brightenedmechanical pulps will never receive the highbrightness like that of a chemical pulp. The

14−5

brightness effect is even reversible when the finalproduct is exposed to sunlight.

Mechanical pulping bleaching is done with sodiumhydrosulfite (Na2S2O4) and hydrogen peroxide(H2O2). A two-stage alkaline sequence can beused to raise the brightness of chemi-thermomechanical pulp to 85% or higher. Bothhydrosulfite and peroxide attack the chemicalgroups that can cause darkening of the paper.Unlike the case in chemical pulp bleaching, ligninis not removed in brightening.

Caustic (NaOH) Valve ApplicationsSodium hydroxide (NaOH) solution is a causticchemical used to break down the lignin that bindscellulosic fibers. This is a highly used chemical butrequires precise control so accurate addition to thewood chips is provided. Poor control can lead toeconomic loss both in NaOH solution and woodchip degradation.

Chips are fed via a screwfeeder into the top of adigester where it is mixed with cooking liquorwhere it is then cooked to a schedule. In modernKraft mills, the lignin is removed by the action ofsodium hydroxide and sodium sulfide under heatand pressure. This solution is known as whiteliquor. As the chips are cooked the lignin and othercomponents are dissolved, and the cellulose fibersare released as pulp.

Design Considerations� Material choice highly temperature

dependent

� Low flows require low flow trims

� Tight shutoff

Typical Specification

Body:

Fisher V150 in CG8M (317SST)CW2M (Hastelloy-C) premium selection

Trim:

Ball: CG8M chrome coated (Microscratch,Micronotch, or Macronotch)Seal: Alloy 6 HD (Alloy 255HD alsoacceptable)Shaft: Nitronic 50

Bearings:Alloy 6 (PEEK if temperature allows)

Packing:

PTFE

Actuator:

Spring-and-diaphragm

Positioner:

FIELDVUE DVC6200 PD level

Chlorine Dioxide ApplicationsDue to environmental restrictions many mills aregoing to elementally chlorine free (ECF) bleachingpractices. This has pushed the mills of today to anew chemical compound to bleach and brightentheir pulp rather than the traditional pure,elemental chlorine. Chlorine dioxide (ClO2) hasbeen the choice by the majority of mills today. Thisis because the compound minimizes degradationto the cellulose fibers while still achieving higherfinal brightness to the pulp. However, it is anexpensive chemical to generate and highlycorrosive, so proper care must be given to choosethe correct solution.

Depending on the type of furnish created at eachmill, pulp from the digester can head toward thebleaching section of the mill. Bleaching is done byremoving the lignin whereas brightening pulpschanges the chemical groups in lignin to formsthat do no darken pulp. Chlorine dioxide is rapidlybecoming an industry standard as a bleachingagent because of its high selectivity in destroyinglignin without degrading the cellulose fibers, thuspreserving pulp strength while still providing astable brightness.

After the oxygen delignification stage (typicallyfound in modern mills), the medium consistencystock (10-14% bone dry) heads to theconventional bleaching sequence which can varyfrom four to six separate stages depending on theend-user’s brightness requirements. A standardmill would utilize a DEOPDED sequence, or analternating sequence of chlorine dioxide (D) andalkaline extraction stages (E) and an oxygen (O)and peroxide (P) brightening stage.

One will always see chlorine dioxide stagesinterspersed with alkaline extraction stages. Thisalternating pattern of acidic and alkaline stageshelps to break down the increasingly smalleramount of residual lignin. But this combination ofchlorine dioxide and alkaline mixture also makes

14−6

this a very corrosive solution and potentiallyerosive due to the percentage of stock content.

Design Considerations� Highly corrosive chemical (not the same as

chlorine applications)

� Erosive depending on stock consistency andvelocity through valve

� Accurate control needed due to cost forchlorine dioxide

� Material choice dependent on chemicalconcentrations; high concentrations must betitanium. Hastelloy C (CW2M) can be used.

Typical Specification

Model:

V150 or V300 (V150S with titanium and ceramic trim can alsobe used. No liner required.)

Body:

Titanium C3 (R50550) or Hastelloy C (CW2M)

Trim:

Ball: Titanium C3 (R50550) or Hastelloy C(CW2M)Seal: TCM PlusShaft: Titanium Grade 5 or Hastelloy C(N10276) *Pin and Taper Key should also be Titanium orHastelloy C

Bearings:

Titanium C3 (R50550) or Hastelloy C (N10276)lined with PTFE

Packing Box:

ENVIRO-SEAL PTFE

Actuator:

Spring-and-diaphragm

Positioner:

FIELDVUE DVC6200 digital valve controllerwith Performance Diagnostics (PD)

The following diagrams are typical valve layoutsfor the various bleaching stages.

Chlorine Dioxide (D) Stage

ValveTag #

ApplicationDescription

ControlFunction Vee-Ball�

High PressureButterfly Valve

(HPBV)V150E

1 Medium Consistency (MC) Control T S P

2 Pulp/Chemical Mixing O/O P S (8580)

3 Filtrate Valve O/O P (8580)

4 Bleaching Agent Valve T P

5 Bleached Pulp Valve T P

6 Washed Pulp Valve T P

7 Filtrate Valve O/O P (8580)

8 Filtrate Valve O/O P (8580)CODE:P = Primary selection, S = Secondary selection, T = Throttling, O/O = On/Off

Alkaline Extraction (E), Hypochlorite (H), Peroxide (P) and Ozone (Z) StagesValveTag #

ApplicationDescription

ControlFunction

Vee-Ball HPBV V150E

1 MC Control T S P

2 Pulp/Chemical Mixing O/O P S (8580)

3 Filtrate Valve O/O P (8580)

4 Bleaching Agent Valve T P

5 Bleached Pulp Valve T P

6 Washed Pulp Valve T P

7 Filtrate Valve O/O P (8580)CODE:P = Primary selection, S = Secondary selection, T = Throttling, O/O = On/Off

14−7

Oxygen (O) StageValveTag #

ApplicationDescription

ControlFunction

Vee-Ball HPBV V150E

1 MC Control T S P

2 Bleaching Agent Valve T P

3 Pulp/Chemical Mixing O/O P S (8580)

4 Control Valve T P

5 Pump Valve O/O P (8580)

6 Control Valve T P

7 Discharge Tank Valve T P

8 Washed Pulp Valve T P

9 Filtrate Valve O/O P (8580)CODE:P = Primary selection, S = Secondary selection, T = Throttling, O/O = On/Off

Figure 14-5. Alkaline Extraction (E), Hypochlorite (H), Peroxide (P) and Ozone (Z) Stages

Figure 14-6. Chlorine Dioxide (D) Stage

14−8

Figure 14-7. Oxygen (O) Stage

www.Fisher.com

Chapter 15

Stock Preparation

Stock preparation is the start of the papermakingprocess and is the controlling factor over finalpaper quality and how well the paper machineruns. To be more specific, this process preparesthe fibers for the paper machine. In order for thisto occur, the fibers must be blended and theconsistency⎯or the percentage of fibers in thewater⎯controlled. Any contaminants must beremoved from the slurry and fibers mechanicallyabraded or refined so they will bond well in thepapermaking process to form a clean sheet.

The stock flows through the various preparationsteps where additives including sizing agents,fillers, starch, retention and drainage agents, anddyes are added to the fiber furnish. The stock isthen further diluted to the final consistency so theslurry can be pumped to the headbox and on tothe paper machine wire.

The stock preparation system can be broken downinto two main areas:

1. Thick stock system (figure 15-1): The initialpart of the stock prep system where fibers arescreened, refined, and blended to prepare theslurry for each grade of paper to be made. Thisprocess has a consistency of 3% to 5% solids.

2. Thin stock system: This system cleans,screens, and dilutes to papermaking consistency.This process has a consistency of 0.4% to 1.0%solids.

Thick Stock ProcessThis process begins with in-coming bales ofpurchased pulp, secondary fiber that might befound in the high density storage chests, or fibersfrom the pulp mill. The pulp is broken down and

Figure 15-1. Thick Stock SystemDrawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

15−2

Figure 15-2. Pulper

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

diluted to a slurry in order to break the fibers apartbefore they are pumped toward the papermachine.

PulpersPulpers, also known as repulpers or slashers, helpto break down the bales into individual fibers(figure 15-2). The bales of pulp or waste paper arefed to the pulper, either by a forklift truck or by aconveyor. In some mills, different fibercomponents are re-pulped separately and blendedtogether later in stock prep. Water is added to thepulper, the pulp bales are added, and theremaining water is added to bring the pulp to theright consistency; typically around 4 to 5% solidsfor low consistency pulpers or up to 18% solids forhigh-consistency pulpers.

Pulping can be done by one of two types ofpulpers:

� Batch pulper: Typically, this process iscompleted in a single vessel.

� Continuous pulper: Supplemental in-linetreatment is commonly used following the pulperto ensure complete dispersion.

The agitator in the bottom, or side, of the pulperprovides the repulping action. Steam is oftenadded along with sodium hydroxide or caustic toraise the pH of the slurry. Dyes and fillers can alsobe added at this point.

Figure 15-3. Pulper Dump

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

Depending on the type of pulp, a batch pulpermight take 30 minutes or more to break up thefibers. Once all the fibers are individually brokenapart, the pulp slurry can be pumped into thepulper dump or storage chest (figure 15-3). Oncethe broken apart pulp is pumped away, a newbatch of bales is re-filled into the pulper.

A continuous pulper is similar to a batch pulperexcept the bales of pulp are continually added andthe slurry is continually removed through anextraction plate under the rotor. The extractionplate has holes of 3/8 to 5/8-inch.

RefiningThis process helps the individual pulp fibers tobond together by employing both mechanical andhydraulic forces to alter the fiber characteristics.This is done by imposing shear stress on thefibers through rolling, twisting, and other tensionalactions occurring in a refiner. This process can beperformed by two different types of continuousrefiners.

1. Disc refiner: The most common type ofrefiner, this unit has a rotating disc and a set ofstationary plates, typically with a plate on eachside of the disc (figure 15-4). These discs are setclosely together so only a small passage betweenthe bars exists. Stock can flow in a parallelarrangement (duo-flow) or a series arrangement(mono-flow). Fibers pass between the movingbars where they are mechanically abraded. Thewater can then enter the walls of the fiber andcause it to swell. This process also helps to breakoff the extra small pieces of fiber, called fines.

15−3

Figure 15-4. Disc RefinerDrawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

2. Jordan (or Conical) refiner: In this type, therotating plug (rotor) and its housing (stator) arefitted with metal bars oriented lengthwise. Thefibers flow parallel to the bars. The position of theplug determines the clearance of the bars andcontrols the amount of work done on the fibers fora constant stock throughout.

It has been shown that having a disc refiner ismore advantageous to a mill. They have lowerno-load energy consumption, which lowers energycosts. They have greater versatility of their refinerplate designs and, because of these designs, cantake on higher stock consistencies. They are morecompact than the Jordan and are a lower capitalinvestment.

Blend ChestAll the furnish components including hardwoodpulp, softwood pulp, and broke are mixed togetherin the blend chest. All components must be wellblended before the stock is diluted down topapermaking consistency.

Stock ScreeningIn some mills, this process follows the blend chest.High consistency pulp of around 4% is run throughfine slotted screens to remove hard debris thatcould cause defects in the sheet of paper (figure15-5). Arranged in multiple stages, the screensgenerally have a protection screen followed by twostages of fine slotted screens using 6 to 10thousandth of an inch, although this size variesbased on the furnish or grade of paper (figure15-6). The first two stages run at a highconsistency, but the final tailing screen runs at a

Figure 15-5. Fine Slotted Screens

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

Figure 15-6. Stock Screening Process

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

somewhat lower consistency, around 1.0 to 1.5%,in order to make a better split between good fiberand debris.

Machine ChestThis is the last chest in the thick stock part ofstock prep. By this point, all the fibers have beenblended and the consistency controlled. It isimportant that the consistency, anywhere from 3%to 4%, is controlled when going forward in thesystem as this stock controls the basis weight ofthe paper. It is from here the stock will enter thethin stock portion of stock prep.

15−4

Figure 15-7. Thin Stock SystemDrawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

Figure 15-8. Fan PumpDrawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

Thin Stock ProcessThis section will describe the thin stock processesof cleaning, screening, and diluting topapermaking consistency (figure 15-7). The thickslurry in the machine chest is pumped to thesuction-side of a fan pump and is diluted to 1% orless with white water drained from the sheetforming process into a stuffbox (figure 15-8).

StuffboxThe stuffbox is located high in the air, near thewet-end of the paper machine, and ensures thebasis weight valve has enough, and constantpressure so the stock flow can be accuratelycontrolled to the paper machine (figure 15-9).Stuffboxes are frequently replaced with variablespeed pumps that supply the basis weight valvedirectly.

Basis Weight ValveThe stock from the stuffbox flows by gravitythrough a pipe leading down into the basis weight

Figure 15-9. Stuffbox

Drawing is fromTAPPI’s Making Pulp and Paper Seriesand is used with permission.

valve (figure 15-10). It influences moisture, caliper,brightness, opacity, draws, strength, machinestability, and product uniformity. If it is not withinacceptable limits, machine performance andproduct quality will suffer. Most paper machinesuse a precision basis weight valve to control theamount of stock going to the fan pump. Basisweight is the key variable in paper quality control.

Typical Specification� Fisher Vee-Ball V150 with either a 2052

actuator or SKF actuator.

� FIELDVUE DVC6200 digital valve controllerwith Performance Diagnostics

White WaterThe water used to dilute thick stock in preparationfor forming the paper sheet comes from the

15−5

Figure 15-10. Fisher Vee-Ball and SKF Actuator Basis Weight Valve

Figure 15-11. Disc Saveall

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

forming section of the paper machine and iscontinually recycled back to the headbox. Thewater, called white water due to its cloudyappearance, is collected in a silo under the papermachine. White water has fiber, fillers, and othervaluable chemicals, so it must be collected andrecycled. Most of the valuable components in thewhite water are eventually retained by the sheetforming process.

SaveallThe excess white water is sent to this devicewhere the fiber and fillers are removed from thewhite water. A disc saveall is made up ofscreen-covered segments (figure 15-11). Avacuum is applied after the segments enter thewhite water and the fibers collect in a mat on theoutside of the screen. The water that filtersthrough is called clear water and can be used for

Figure 15-12. Cleaner

Drawing is from TAPPI’s MakingPulp and Paper Series and isused with permission.

diluting water on consistency control or otheruses.

The collected fibers are washed off the segmentsas they leave the white water pond. The recoveredfiber is diluted and blended with the other fibersources in the blend or machine chest.

CleanersThe fibers, after being diluted by white water in thesilo, then proceed to the cleaners (figure 15-12).Each cleaner is only capable of handling a smallportion of the total flow so many are needed forcleaning. The stock enters the cleaner tangentiallycausing the flow to form a vortex. The spinningaction causes the heavy particles to be thrownoutside the vortex. These heavy contaminantsmove down the inner wall of the cleaner and arerejected through the bottom discharge hole of thecleaner. Lighter fibers stay near the center andexit through the top.

Some of the good fiber is rejected through thebottom. In order to reclaim this good fiber, therejects will flow from the primary to the secondarycleaners (figure 15-13). The rejects enter asecondary cleaner to reclaim the good fibers.There can also be tertiary, a fourth and fifthcleaner as well if needed (figure 15-14). Each

15−6

Figure 15-13. Primary Cleaners

Drawing is from TAPPI’s Making Pulp and Paper Series and is used with permission.

Figure 15-14. Secondary Cleaners

Drawing is from TAPPI’s Making Pulp and Paper Series and is used with permission.

successive bank of cleaners contains fewerindividual units than the previous bank.

Deaeration ChamberDepending on the stock, small air bubbles canform around the fibers. This slows drainage duringthe sheet forming process and can cause pinholesin the sheet. In order to remove these bubbles, adeaeration chamber is used (figure 15-15).

The chamber or tank is connected to a vacuumpump so the stock in the tank is also undervacuum. This vacuum allows the stock to boileven though the stock temperature is below100°C. This boiling action helps to release air inthe stock.

ScreensAlmost all paper machines have a screen beforethe headbox to remove contaminants from thefurnish. These screens have a basket of eitherholes or slots that allow fibers to pass through thatcollects shives, pieces of plastic, or fiber flakes.

These slots are generally 10 thousandths of aninch wide and are more efficient at removing smalldebris than holes.

These screens also have a way to backflush theholes to prevent plugging. Generally, this is donewith a rotor and hydrofoils. The foil passes overthe hole and produces a low-pressure pulsefollowed by a high vacuum pulse. These pressurepulses keep the openings in the basket frombinding.

Broke HandlingBroke is better known as internal waste papergenerated by the paper mill. This might be fromthe wet-end from the forming or press sections, orthe dry end from the dryer section, reel, winder, orother finishers. Broke contains good fiber andchemicals that should not be lost.

Broke is generally captured in an under themachine repulper or broke pulper. The brokeeither drops down into the pulper along a chute oris conveyed or blown into the repulper. Wet-endbroke is easy to break up into the individual fibers,but fully dried broke requires more aggressiveagitation. Sometimes, broke is sent through a highdensity centrifugal cleaner to remove large heavyparticles. This then goes through a deflaker, whichmechanically breaks up the underfibered flakes.

Compact Stock ProcessRecently, some paper makers have moved to amore simplified stock prep. Rather than relying onlarge volumes to reduce variation, there is morereliance on modern process control to make thenecessary adjustments to correct for variations(figure 15-16). For instance, for thick stockblending and feeding, instead of having two cheststo blend the various components and then feedthe stock to the fan pump, those functions arehandled by a small mixing tank (figure 15-17). Asfor deaeration, one can replace the large silo anddeaeration chamber with a small centrifugaldeaeration pump (figure 15-18). This new systemhas quicker response time and less total volume.

Brown Stock Rejects ValveProcess impurities have a negative affect on endproduct quality. These impurities may damageprocess equipment and cause runnabilityproblems. As such, all solids contaminants have tobe removed from pulp. Some contaminants can be

15−7

Figure 15-15. Deaeration Chamber

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

Figure 15-16. Compact Stock Prep SystemDrawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

separated from pulp on screens, whereas otherssimilar or smaller than fibers may be removed byother methods (figure 15-19).

Pulp from cooking always contains someunwanted solid material. Some of the chips maynot have been fiberized properly, and some of thefibrous material may not be completely in the formof individual fibers. The main purpose of the pulpscreening process is to separate harmfulimpurities from pulp with minimal fiber loss, and atan acceptable cost level. Bark, sand, shives, androcks are typically found within the cooked chipsand must be removed. There are multiple ways toseparate out the impurities. This can be donemechanically by screen plates where separation isbased on particle size, whereas gravimetric orcentrifugal force field is needed for weight-basedparticle separation. However, pressure screeningin multiple stages is the preferred method forremoval of impurities such as sand and shives

from the pulp. These operate by separating thefeed pulp into either acceptable (impurity free) orrejects (impurity rich). The acceptable pulp passesthrough the device, and the rejected pulp isremoved for further processing. This isaccomplished in stages since most pressurescreens cannot sufficiently concentrate theimpurities in only one screening stage due tothickening of the reject on the screen itself. Thepurpose of the rejects stages is to concentrate theimpurities in the reject stream, and to return thegood fibers to the main process line.

Design Considerations� Tight shutoff not required.

� Hardened materials to protect valve bodyfrom wear.

� Ball valve to shear through impurities.

15−8

Figure 15-19. Centrifugal Deaeration Chamber

Figure 15-17. Compact Stock Mixing Tank

Drawing is from TAPPI’sMaking Pulp and Paper Se-ries and is used with per-mission.

Figure 15-18. Centrifugal Deaeration Chamber

Drawing is from TAPPI’s Making Pulp and Paper Series and is used with permission.

Typical Specification� Fisher Slurry Vee-Ball V150S with high

chrome iron trim.

Medium Consistency (MC) PumpValveThe medium consistency centrifugal pump (MCpump) is used for continuously pumping pulp stockup to 18% bone dry (BD) consistency and can belocated in numerous areas of the mill. Inprocesses where the flow control valve isresponsible for controlling the pumps headpressure, special care should be given to the valveselection.

In some cases, concerns of pulp stock flowbehavior and buildup are addressed by specifyingan expanded outlet valve. Many MC pumpmanufacturer’s require expanded down streampiping.

Design Considerations:� Expanded Outlet

� Ball valve to shear through impurities

� Precise control

Typical Specification:� Fisher Vee-Ball V150E

www.Fisher.com

Chapter 16

Wet-End Chemistry

In this chapter, we will be discussing materials,other than fibers, that are added to the slurry offibers before paper is formed. It is important tokeep in mind that there are two types of additives.

1. Functional additives—These additions aretreatments necessary to meet the particular needsof an end-customer.

2. Process additives—These additions modifythe properties of the paper. They can be used in amultitude of different fashions.

SizingThe purpose of sizing is to enable paper productsto resist penetration by fluids. This is critical forprinting operations. If appropriate sizing is nottaken into account, the ink will diffuse into thesheet and cause severe quality problems. Sizingcan be achieved by using wet-end additives or byapplying a coating to the surface of the dried paper.

The traditional wet-end sizing agent is a modifiedrosin, better known as “rosin size.” This additivecan actually make paper repel water under acidicpapermaking conditions. Rosin size comes fromsoftwoods as a byproduct during the Kraft pulpingprocess. To make rosin work as a sizing agent,papermakers also add aluminum sulfate(Al2(SO4)3), better known as “papermaker’s alum.”This combination is an effective way to makepaper resist water and other fluids. The process isalso known as “acid sizing” as this combinationworks well in acidic aqueous environments.

Internal StrengthMany natural and synthetic polymeric substancescan be added to stock at the wet-end to improvethe physical properties of the dry paper sheet.They are to reinforce the fiber-to-fiber bondsthereby improving tensile strength, reduce “fuzz”or lint on the paper surface, and can reduce therate of water penetration.

Traditional internal strength additives are naturaland modified starches and gums. Starches arepolymers of glucose whereas gums are polymersof mannose and galactose. However, the trend isnow toward the use of synthetic polymers aslatexes and polyacrylamides used in combinationwith starches and gums. These new productshave now met a wider range of specificrequirements for greater paper strength withdifferent degrees of stiffness and stretch.

Wet-Strength ResinsThe purpose of wet-strength resins is to tie fibersand fines together with additional bonds that arenot taken apart by water. Wet-strength paper isdefined as paper that retains more than 15% of itstensile strength when wet.

The most common wet-strength agents areureaformaldehyde, melamine-formaldehyde, andpolyamide resins, and are water soluble. Theselong-chain polymers can be used on paper for juicecontainers or other liquid containers so the fibersremain strong even after getting wet. However,because these agents are water soluble, they mustbe fixed onto fibers with the help of retention fillers.

16−2

Figure 16-1. Stock Approach SystemDrawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

FillersFor most copy paper, around 15 - 30% of thepapers are fillers; the majority of which is clay andPCC — precipitated calcium carbonate. Theseadditions are generally made to lower the overallcost of materials and can be used for brightness,opacity, or even the smoothness of the paper.However, fillers do not bond together in the sameway cellulose fibers do, so they reduce thestrength of the paper.

This limits the amount one can put in the sheet.Another difficulty is trying to get the filler to remainwith the fiber during the sheet forming process. Itis not uncommon to lose fillers while on themoving wire or forming fabric. Additives are usedto increase the retention of the fillers so they arenot lost during this process.

All of these wet-end additives are either soluble inwater or are small enough to fit through the smallopenings of the forming fabric. To keep theseadditives from falling out of the process, retentionaids are added. These retention aids insure thatthe fillers attach to the fibers. These are typicallyadded just before the headbox or before theheadbox screen (figure 16-1 ). They are addedlate in the process because excess agitation couldbreak up the polymer chains.

The size and shape of mineral additives cangreatly affect the properties of the paper. For

instance, most grades of paper have a thicknessspecification. Papermakers want to make thepaper as thick as possible all while using the leastamount of fiber. The more flat the microscopicfiller, the more dense the paper. However, manytimes these fillers are so small the filler getsbetween the fibers and makes the papersmoother.

The common papermaking fillers are clay, calciumcarbonate (CaCO3), talc, and titanium dioxide(TiO2). Clay is the most popular since it is cheap,stable, and generally provides good performance.Calcium carbonate is a better opacifier than clayand has higher brightness. Titanium dioxide is thebrightest and most effective opacifier, however, ishigh cost. Talc is used as a “soft” filler that helpsto give paper a soft, silky feel to the product.

Finally, due to the amount of water being pumpedthrough the system, foam can be formed. This cancreate spots or pinholes in the paper. To alleviatethis problem, defoamers are added to help thebubbles coalesce into bigger bubbles. Theselarger bubbles will then rise to the surface andbreak. The water and warm temperatures can alsocreate a wonderful environment for bacteria andfungal slimes. These are called “bugs” in the papermill. These bugs can lead to holes and spots inthe paper and frequent sheet breaks but can becontrolled with the addition of biocides to thepaper machine “white water” system.

16−3

Titanium Dioxide (TiO2) ApplicationsTitanium dioxide is used as a paper additive toincrease brightness and opacity. It is a fine whitepowder that is added at a low flow rate as a slurryto pulp stock. The process is very erosive andrequires fine control and tight shutoff.

Typical Specification

� Fisher Vee-Ball V150 Micro-Scratch,Micro-Notch, Macro-Notch with Ceramic Trim

� V500, Reverse Flow with ceramic trim

16−4

www.Fisher.com

Chapter 17

Paper Machine

Wet EndThe paper machine is essentially a series ofprocesses all tied together (figure 17-1). Theseprocesses are designed to take fibers in a diluteslurry of water and produce a dried web of paper.The paper machine is described in two parts: thewet end, which is the forming section, and the dryend, which includes the pressing and dryingoperations.

Fourdrinier Single-Ply ProcessThe following are the steps taken to form thefibers onto the wet end of the paper machine innew and more modern pulp and paper mills:

Tapered Manifold

The stock heading toward the headbox is comingfrom the headbox pressure screen (figure 17-2).This stock is to be spread uniformly over the entirewidth of the paper machine where some machinesare up to 400 inches wide. This is accomplishedby using a tapered manifold (figure 17-3).

The tapered manifold starts large at the inlet endand tapers down over the length of the device.This allows the pressure to remain the same eventhough stock is being diverted through numeroustubes to the headbox. At the end of the manifold isa recirculation line that allows one to balance themanifold for different flow rates.

Figure 17-1. Fourdrinier Paper Machine

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

17−2

Figure 17-2.Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

Figure 17-3. Multitube Tapered Manifold

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

HeadboxFrom the manifold, stock enters the headboxthrough a series of tubes. The headbox hasseveral purposes. It needs to eliminate theturbulence coming out of the tubes from themanifold, it has to break up the fiber flocs, and itmust ensure the amount of stock coming out ofthe slice is uniform all the way across the width ofthe machine. Below are the major types ofheadboxes:

� Rectifier Roll (Air-padded) — This type ofheadbox has a number of hollow rolls within thestock stream, inside the headbox. The rolls areperforated with approximately one inch diameterholes throughout the surface of the roll (figure

Figure 17-4. Rectifier Roll (or Air-Padded) Headbox

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

17-4). Typically, there is one roll at the entrance ofthe box, a second roll in the main pond area, anda third roll near the discharge of the headbox. Thestock fills the box almost to the top of the rectifierroll so a cushion of air remains above the stock.This cushion of air helps to lessen pressurevariations, thus lessening changes in the basisweight of the paper being produced. The rollsrotate slowly to help eliminate any large-scale

17−3

Figure 17-5. Hydraulic Headbox

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

turbulence inside the headbox and to break up thefiber flocs so fibers are well distributed. This typeis good for slower paper machines and specialtymachines with a wide range of flow requirements.

� Hydraulic — These devices are completelyfilled with stock. Because no air cushion existsmany hydraulic headboxes use a pressureattenuator at the headbox manifold to reduce anypressure variations. Since hydraulic headboxeshave no rectifier rolls, they are designed todeflocculate the stock by changes in velocity asthe stock passes through tube bundles or acrossflat sheets. The discharge velocity from the slicedepends directly on the feeding pump pressure.This type of headbox can be found on most newpaper machines (figure 17-5).

� Dilution Control — This type uses dilutionor consistency control across the width of theheadbox to correct errors in basis weight (figure17-6). By opening a particular dilution valve, thecomputer will add dilute white water to the insideof the headbox in precisely the right point to dilutea heavy streak. If the sheet is too light, thecomputer removed some of the dilute water byclosing the valve. There can be as many as onehundred or more dilution valves located along the

Figure 17-6. Dilution Control Headbox

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

back of the headbox, all of which are individuallycontrolled.

Each type of headbox contains a headbox slice.The slice is a full-width orifice or nozzle with acompletely adjustable opening to give the desiredflowrate. The jet of stock emerging from a typicalheadbox slice contracts in thickness and deflectsdownward as a result of slice geometry. The jetthickness, together with the jet velocity,determines the volumetric discharge rate from theheadbox.

Every slice has a top lip and an apron (bottom lip),both constructed of suitable alloy materials toresist corrosion. The top lip is adjustable up ordown as a unit (main slice) and also in local areasby the use of individual micro-adjusters. Thesesmall adjustment rods are attached to the slice lipand, with manual adjustments or the help of acomputer system, can improve the distribution ofthe stock and allow for a more uniform basisweight across the width of the paper machine. Therectifier and hydraulic headboxes utilize adjustableslices in the same way.

Forming WireThe stock leaving the headbox slice, typicallybetween 0.5 - 1% solids, is deposited onto asynthetic forming fabric (or wire). Waterimmediately begins to drain through the forming

17−4

fabric producing a mat of fiber on the fabricsurface. The jet velocity at which the stock isdeposited onto the fabric is very important. Thisprocess controls fiber alignment and affects thestrength properties in the direction of web travel.This wire will continuously travel the length of thewet-end of the fourdrinier machine providing timefor sufficient water removal.

Forming BoardThis is the first static element under the wire usedto remove additional water. This element supportsthe wire at the point of jet impingement. Thisdevice is needed to prevent wrinkles in the formingfabric. This is accomplished by correctly spacingthe blades of the forming board at the correctangle so jet delivery can be optimized for bestsheet formation. This element also serves toretard initial drainage so fines and fillers are notwashed through the sheet.

Foil UnitsFollowing the forming board, water is drained fromthe sheet over foil units. These foil units have ablade with a high slope or angle towards the rearof the blade. This creates a small vacuum whichpulls more water through the mat. The foils alsocreate turbulence to help break up any celluloseflocs that are beginning to form. The higher the foilangle, the greater the vacuum created thus moreturbulence and water drainage.

FlatboxesThe sheet on the forming fabric is still very wet asthe action of the foils is not enough to remove anymore water. A flatbox, or vacuum box, is a narrowbox positioned under the forming fabric, acrossthe width on the paper machine and is connectedto large vacuum pumps that provide a differentialpressure or vacuum that is needed for furtherremoval of water. These vacuums are capable ofincreasing the sheet to around 15% solids.

Couch RollThe final device used to remove water on theforming section, the couch roll, consists of ahollow outer shell that rotates with the wire and astationary inner vacuum box. The vacuum box isconnected to a large vacuum pump. The holes inthe shell allow the vacuum inside to remove waterfrom the sheet. This removes water to make theconsistency 20 - 25% solids. The overall goal ofthis final section is to get the sheet as dry as

possible to improve the strength of the sheetbefore it is peeled, or couched, from the wire andsent for further de-watering to the press section ofthe paper machine.

Fourdrinier Multi-Ply ProcessThis process is exactly the same as the single-plyprocess except multiple layers or plies areeventually combined into a single sheet. This canbe accomplished in a multitude of ways:

� Stratified Headbox — This device iscapable of depositing two or three different layersof stock on the fourdrinier wire at the same time(figure 17-7).

� Secondary Headbox — This device issimply a common paper machine with a secondheadbox to put a secondary layer on top thepreexisting layer.

The above two options still have an issue; alldrainage occurs through the bottom layer thus,causing some mixing of the fibers. To alleviate thisproblem, papermakers have devised a few otheroptions.

� Multiple Fourdriniers — These machinesactually have multiple fourdrinier machines runningon top of one another. The sheets all wind into acentral area where the sheets are pressedtogether to make a multi-layer paper.

� Cylinder Former — This device consists ofa vat and a large diameter wire-covered cylinder(figure 17-8). The sheet is formed on the wire asthe cylinder rotates through the dilute stock slurry.The water is then drained into the cylinder. Thewet sheet is then consolidated and couched offonto a wet felt. This is most effective in themulti-ply process as each cylinder lays down anindividual ply of paper, which are each individuallycouched to the previous ply, thus, making aheavyweight paperboard.

Twin Wire FormersThis device now holds the speed and productionrecord for the majority of paper grades. Thistechnology is becoming more popular than thefourdrinier machine because:

� Water can drain from both sides of the sheetrather than one, leading to the top and bottom ofthe sheet being more alike.

17−5

Figure 17-7. Stratified HeadboxDrawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

Figure 17-8. Cylinder Former

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

� Water can be drained in a much shorterdistance.

� The technology is much faster than afourdrinier machine.

These twin-wire formers can be broken up into twotypes.

Figure 17-9. Gap-Wire Former

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

� Gap Formers — These devices inject theheadbox jet between two converging wires (figure17-9). Many gap formers have a large forming rollwhere the majority of the sheet drainage occurs.There are also several high vacuum boxes and asuction couch roll where the two wires eventuallyseparate and the sheet is taken into the presssection.

Gap formers hold the speed records for single-plygrades of paper such as newsprint or copypapers. This technology is also becomingcompetitive in the multi-ply paperboard market.

17−6

Figure 17-10. Top-Wire FormerDrawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

� Top-Wire or Hybrid Formers — This devicesits on top of a conventional fourdrinier single-wireformer and deposits a jet from the headbox ontothe single wire (figure 17-10). The wire passesover an expanse called the “free drainage zone”where the initial drainage is downward. Thesecond wire then covers the top of the sheetallowing dewatering to occur in both directions.Multiple configurations can be found with vacuumshoes and blades throughout the forming area tohelp the dewatering process.

What makes this device more advantageous to aGap Former is that a top-wire unit can be readilyfitted to an existing fourdrinier to improve sheetquality and allow for machine speedup.

Dry EndAfter the forming section of the paper machine,the sheet is still approximately 20% solids. Nowthat the forming section cannot take or vacuumout any additional water, we must mechanicallypress out the water.

Before discussing the press section, one mustunderstand the nip. Most paper machines have atleast two nips, and can have as many as five. Thenip is a process of removing water by mechanicalmeans by passing the wet paper sheet, almostalways with a felt, between two rotating pressrolls. Nips generally follow these steps:

1. Compression of the sheet and felt between tworolls begins. Air flows out of both structures untilthe sheet is saturated.

2. Now that the sheet is saturated, hydraulicpressure within the sheet structure causes waterto move from the paper into the felt. Once the felt

becomes saturated, water moves out of the felt.This phase brings the paper to its maximumhydraulic pressure.

3. The nip expands until the hydraulic pressure inthe paper is zero, corresponding to the point ofmaximum paper dryness.

4. In this phase, both paper and felt expand andthe paper becomes unsaturated. Although anegative pressure is created in both structures, anumber of factors cause water to return from thefelt to the paper; also known as “rewetting.”

The top roll is mechanically loaded to create thedesired pressure within the nip. The higher thepressure applied at each nip, the more effectivethe water removal. However, too much pressure atthe nip will take the felt and sheet beyond thepoint of saturation. This condition is called“crushing” and significantly weakens the sheetstrength in the nip. Operating a nip at the point ofcrushing will cause the sheet to break. At fastermachine speeds, higher pressure will have adiminished effect because of the brief residencetime in the nip.

PressThe primary objective of the press section is toremove water from the sheet and consolidate thepaper web. This section can also provide otherproduct requirements such as providing surfacesmoothness, reducing bulk, and promoting higherwet web strength.

The oldest style of presses is a straight throughpress arrangement (figure 17-11). For the moremodern arrangements, the most widely usedpresses can be described in full below.

� Roll — This press consists of twolarge-diameter rolls, loading arms to supplypressure, and a press felt (figure 17-12). As the

17−7

Figure 17-11. Straight-Through PressDrawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

Figure 17-12. Roll Press

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

sheet enters the ingoing nip between the two rolls,pressure builds and the sheet is compressed. Aspressure continues to rise, the water in the sheetis forced out of the sheet into the press felt. Thispressure can be anywhere from 500 - 2,000pounds per square inch where the higher thepressure, the dryer the sheet will become.

However, press rolls normally have rubber coversthat essentially smash together. The “footprint”that is created on the rolls, better known as the nipwidth, can be larger or smaller depending on howhard we press the rolls together. Past themid-point of the nip, the pressure begins todecrease and some of the water is actually suckedback into the sheet or re-wetted. How dry thepress can get paper is also determined by the niptime. Generally, the roll press can increase sheetconsistency to 38 - 32% solids.

� Shoe — Rather than a roll press, somepaper machines use a shoe press (figure 17-13).

Figure 17-13. Shoe Press

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

This type of device has a stationary roll that isactually a hydraulically loaded stationary shoewhich is concave shaped in order to fit the otherroll. The shoe is covered by a rotatingpolyurethane blanket lubricated with oil toeliminate any type of friction between the shoeand blanket. This creates a high pressure and longtime in the nip thus allowing for better drying(figure 17-14). This drying can get the sheet toapproximately 50% solids.

� Fabric Press — For this type, amultiple-weave, non-compressible fabric beltpasses through the nip between the felt and therubber-covered roll to provide void volume toreceive the water. The water is removed from the

17−8

Figure 17-15. Modern Straight-Through PressDrawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

Figure 17-14. Shoe Press Nip

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

fabric by passing over a suction box on the returnrun.

� Extended Nip — This type of press featuresa wide nip to give the sheet a long dwell time athigh pressure. When used as the last nip, thispress provides not only a much drier sheet, butalso a stronger sheet due to improvedconsolidation of the web structure.

Mechanically pressing the water from the sheet iseight times cheaper than trying to dry the sheet.Because less drying would be needed, faster

machine speeds and production can also beachieved.

One of the most important things on the press isthe press felts. In the past, these were wovenwoolen blankets. Now, these are commonlycomposted of a woven synthetic base fabric andfiber matt, attached by a sewing punchingprocess. These must be strong enough towithstand the compression of the rolls while stillproviding void volume for the water that isremoved from the sheet in the press nip.

Press ArrangementsBelow are the typical press arrangements that cantypically be seen in pulp and paper mills.

� Straight-Through — The oldest andsimplest of the press arrangements, this type canstill be found on paper and board machines. Eachpress within this design has a smooth top roll anda bottom felted roll so only the top surface of thesheet received smooth roll contact. Later, aninverse second press allowed the bottom of thesheet to be in contact with the smooth roll.

� Modern Straight-Through — This unit hastwo, double felted presses, which is common innew paper machines (figure 17-15). In this design,the sheet is fully supported off the couch, betweenthe individual presses, and into the dryer section.Depending on the weight of paper, the more shoepresses are necessary for water removal.

17−9

Figure 17-16. Two-Tier Drying System

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

Drying

After pressing, the sheet is conveyed through thedryer section where the residual water is removedby evaporation. This is due to the cellulose fibersbeing hydrophilic and wanting to hold onto water.At this stage, the wet web is approximately 40 -45% solids.

Most paper is dried on cast iron or steel dryingcylinders, each cylinder being 60 - 72 inches indiameter, that are fed with steam from the boilerwith pressures between 15 - 150 psi, dependingon the type of paper. The wet web is held tightlyagainst the cylinders by a synthetic, permeablefabric called the dryer felt. The evaporated wateris carried away by ventilation. The final result ofthe dryer section is paper with 5 - 8% moisture.

Most paper machines have three to fiveindependently-felted dryer sections, each withtheir own speed control to maintain sheet tensionbetween sections and adjust for any sheetshrinkage. The two-tier configuration is the mostcommon arrangement for dryers. The sheetpasses from dryer to dryer where it is tightlypressed against the dryer cylinders by the dryerfabric (figure 17-16). The paper passesunsupported between each of the dryers. Oncethe remaining water in the sheet rises to its boilingtemperature, water is converted into steam. Thissteam is collected in a containment hood toremove the water vapor. On the inside of the dryercylinders, the steam is condensed back into water,or condensate. This conversion of steam intowater supplies the majority of the energy that driesthe paper.

Figure 17-17. Single-Tier Dryer Section

Drawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

At high speeds, the unsupported paper betweenthe dryers can flutter and occasionally break.Because of this, many modern paper machinesare going with a single-tier or “Uni-run”arrangement where the dryer fabric is in constantcontact with the sheet (figure 17-17). This allowsfor the machine speed to increase.

The condensate that collects within the dryercylinders must constantly be removed. This is thejob of the siphons.

Steam and Condensate SystemThe heat energy for paper drying comes fromsteam as it condenses inside the dryer cylinders.This is known as latent heat. Steam alwayscondenses at the saturation temperature, asdefined by the pressure in the system. This isimportant when trying to have uniform dryingacross the machine. The condensate that forms inthe dryer cylinders is removed by a speciallydesigned pipe assembly called a siphon. Onslower machines, the condensate collects in apuddle at the bottom of the cylinder. Forhigh-speed machines, a true rimming conditioncan be reached where the condensate covers theentire inside surface due to centrifugal force.

In siphoning, differential pressure pushes thecondensate through the siphons (figure 17-18).The siphons carry the condensate from the dryerto a separator that collects the condensate andrecycles any steam that has been blown through.The condensate is reused by sending it back tothe mill’s boiler feedwater system (figure 17-19).

Hood VentilationIt is important to realize that a ventilation hoodexists over the entire dryer section beginning fromthe press section up to calendaring. Depending on

17−10

Figure 17-18. Steam Drum and SiphonDrawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

Figure 17-19. Condensate ProcessDrawing is from TAPPI’s Making Pulp andPaper Series and is used with permission.

17−11

the type of hood arrangement, 7 - 20 pounds of airare utilized for each pound of water evaporated.To prevent drips, buildup and corrosion within thehood, the volume and temperature of exhaust airmust be sufficient to avoid localized condensation.

Modern generation hoods, also called “high-dewpoint hoods”, are well sealed and insulated.Diffusion air is totally eliminated, and the amountof fresh makeup air is sharply reduced byoperating at high temperature with high recycle.

Size PressSizing solution is commonly applied within atwo-roll nip which gives the name size press. Thisdevice is found about three quarters of the waydown the dryer section where starches and othermaterials can be applied at this station in order toimprove sheet surface, internal strength,smoothness, and resistance to water penetration.The objective is to flood the entering nip withsizing solution so the paper absorbs some of thesolution.

The basic mechanisms that incorporate starchsolutions into the sheet is the sheet’s ability toabsorb the sizing solution and the amount ofsolution film passing through the nip. Sheetabsorption is greatly affected by sheet moisture ashigher sheet moisture promotes absorption.However, the level is typically controlled to 4 - 5percent or less to ensure the sizing agent is keptnearer to the surface.

Modern paper machine speeds can createturbulence due to changing nip pressures. Thisaffects the size press’ ability to evenly distributethe solution. To overcome this issue, most sizepresses have larger diameter rolls to keep solutionturbulence more manageable. Others have begunto use a metering size press, which applies thestarch to the surface of the size press rolls bymetering the amount applied to the sheet with ablade or rotating rod.

Today, the size press can also be used for coatingapplications; including pigmented coatings andother specialized surface applications. Somemanufacturers are beginning to include pigment inthe starch application.

CalendaringThis process helps to smooth and flatten thesheet, better known as the thickness or calipervariation of the sheet. This is done by

compressing the higher areas in the sheet morethan the lower. Surface smoothness improves sothe paper prints better. Sheet density is increasedso the paper becomes thinner and denser. Makingthe sheet denser also makes the paper less stiff.

Calendaring changes the surface and interiorproperties of the sheet by passing the webthrough one or more two-roll nips where the rollersmay or may not be of equal hardness. By usingextreme pressures, the objective is to press thepaper against the smooth surface with sufficientforce by using one of the calendaring types below.

� Hard-Nip — Both rolls are made of eitheriron or steel. Although this type smooths the sheetby calendaring to a uniform thickness via flatteningthe higher areas in the sheet, it can create areasthat are more dense as well. These densityvariations are due to basis weight variations andcan turn into variations in surface properties

� Soft-Nip — For this type, the loading roll isstill hard but the opposing roll has a soft polymericcover, usually some type of polyurethane. Sincethe side contacting the metal roll receives a muchbetter finish than the side contacting the resilientroll, it is necessary to have two nips for equalfinish.

Calendaring at high temperatures is desirablebecause the paper becomes more pliable and canbe calendared at lower pressure.

ReelAfter drying and calendaring, the paper productmust be collected in a convenient form forsubsequent process off-machine. This is typicallydone by a drum reel which collects the product toa specified diameter.

Most reels are motor-driven under sufficient loadto ensure adequate tension on the sheet from thecalendars. The web wraps around the reel drumand feeds into the nip formed between the drumand the collection reel, which is held by thesecondary arms. While the reel builds up, anempty spool is positioned on the primary arms.

To transfer from a full roll to empty spool, theparent roll needs to be removed. An empty spoolis held in the primary arms and is brought up tospeed before contacting the paper on the reeldrum. The paper is transferred to the new spooland the full parent reel is released from the reeldrum. Once the parent roll is removed, the primaryarms move the new roll down to the rails. Thesecondary arms are brought forward to take the

17−12

spool and the primary arms return to their originalposition.

WinderThe purpose of the winder is to cut and wind thefull-width, large diameter paper reels into suitablesized rolls. These rolls may then be wrapped andsent directly to the customer or may be processedthrough subsequent coating, calendaring, orsheeting operations. During winding, the twoedges of the reel are trimmed off and conveyedback to the dry-end pulper, or broke pulper.

The full-width machine reel is transferred from thereel stand to the unwind stand by an overheadhoist. From the unwind stand, the paper isthreaded through the web-tensioning rolls, theadjustable slitters, adjustable spreader bar andonto fiber or plastic cores. Typically, a steel shaftis inserted through the cores to provide a locking

arrangement. However, some of the newerwinders operate “shaftless” by providing aretaining surface on one side to preventcross-machine wandering.

The winder drive must be capable of speeds oftwo and a half to three times faster than the papermachine in order to have time to change rolls,change reels, repair breaks, remove defectivepaper, set up the slitting arrangement, and adjustthe spreader bar.

Roll Finishing

The steps in roll finishing are scaling, wrapping,crimping, heading, and labeling. Each of these isdone manually at one time. Today, most wrappingoperations are carried out semi-automatically, andthe labeling function is handled by a dataprocessing print unit.

www.Fisher.com

Chapter 18

Boilers — Water/Steam Cycle

The efficient production of steam and electricity isan important function in the overall process of pulpand paper production. These items are basic rawmaterials required in large quantities for themanufacturing of pulp and paper.

In years past, the cost of operating the powerhousemay not have been a priority; however, withincreasing prices of fuel and purchased electricityaffecting bottom line profits, the industry hasevolved to a more efficient and conservationconscious powerhouse. One key item in a moreefficient powerhouse involves utilization of areliable control system. The final element for manyof these controls is the control valve.

Although design will vary from mill to mill, ageneric water/steam cycle is shown in figure 18-1.This process is commonly referred to ascogeneration due to the simultaneous use of fuelenergy for both process steam requirements andelectrical power generation via steam driventurbines. Figure 18-2 details the upper, orconvective, section of a boiler and indicates valvesrequired for process control.

Condensate Return SystemWater pumped to the boilers for production ofsteam is composed of condensate returned fromprocess and demineralized make-up water. Thismixture is commonly referred to as boilerfeedwater (BFW).

The condensate returned from process isdemineralized water, which was used to producesteam in the boiler and has condensed after givingup vaporization energy to process. In a typical mill,about 40-50% of the condensate is returned to thepowerhouse. Most of this comes from indirect

heating such as paper machine dryers and variousheat exchangers. A large portion will also bereturned from the condenser if a condensing turbineis used. Losses occur in direct heating or cleaningapplications such as pulp cooking and sootblowers.Other losses occur in the transport system(pumps, valves, tanks, and piping) and condensatecontaminated by leakage (mixing with process).

Serious operational problems may result ifcontaminated condensate (with black liquor, whiteliquor, etc.) enters the boilers. To prevent this, acondensate dump system is used. A conductivityelement is used to sense contaminants and sendsa signal to valves which dump the condensate tosewer until the problem is corrected. Automationof this system can save the time lost to a manualoperated system.

Demineralized or deionized (DI) water involvestreatment with ion exchange resins to removehardness (minerals) and silica, which woulddeposit on boiler tubes. This process involveslarge equipment and operational expense. Mostmills are equipped to supply only a portion of thefeedwater required with demineralized water.Thus, the return of as much clean condensate aspossible is critical from an economic andoperational standpoint.

Condensate is usually brought to a singlecollection tank from various process users and theturbine condenser. It is then sent through acondensate polisher to remove any scale depositspicked up in the return system. From the collectiontank the condensate is pumped to the deaerator(DA) heater. In some mills an indirect feedwaterheater is used before the DA to raise temperatureto near saturation. Low pressure (150 psig) steamis used for heating and the resulting condensate isreturned to a lower pressure reservoir such as thecondenser hotwell or DA.

18−2

Boiler Feedwater SystemThe BFW system begins at the DA and ends atthe inlet to the economizer. The main componentsare the DA, the boiler feed pump, and the highpressure feedwater heaters. The main purpose ofthe BFW system is to condition the feedwater forentry into the boiler. The DA removes unwantedoxygen from the feedwater, which, in turn,prevents corrosion in the entire piping system. Theboiler feed pump raises the pressure and the highpressure feedwater heaters raise the temperatureof the feedwater. The critical valves within theboiler feedwater system are the boiler feedwaterrecirculation, feedwater startup, and feedwaterregulator valves.

Feedwater Recirculation Valve

In order to protect the feed pump, there must be arecirculation system. The boiler feed pumprecirculation valve takes feedwater from the boilerfeed pump and recirculates it to the DA. It is thereto protect the pump from cavitation and excesstemperature rise. There are three basic methodsof providing feed pump recirculation. Two oldermethods are continuous recirculation and on/offrecirculation. The current method is modulatingrecirculation. This provides minimum recirculationflow to protect the pump and optimize efficiency. Itrequires a high technology recirculation valve. Therecirculation valve typically experiences cavitationand if not properly taken into account with valveselection, cavitation damage will result. Becauseof the cavitation, tight shutoff is required. Anyfluids leaking past the valve will cavitate andcause damage to the seat. A leaking recirculationvalve can cause decreased unit capacity, reducedefficiency and repeated maintenance and repaircost. Plugging can occur if feedwater is not clean.A common issue with all feedwater applications iscorrosion due to materials chosen. Amine orhydrazine treated feedwater is corrosive to thecobalt binding in alloy 6. If the feedwater istreated, use of this material should be avoided.

Design Considerations:

� Cavitation

� Tight shutoff (Class V)

� Typical process conditions are 800-1200 psigand 200-400 °F

Typical Specifications:� easy-e, HP, EH, or Cavitrol IV trim

� Cavitrol III Trim

� HTS1 option with improved pressure balanceseal

� FIELDVUE digital valve controller with lowtravel cutoff

Optional:

� NotchFlo or Dirty Service Trim

� Protected inside seat

Feedwater Startup and RegulatingValvesThe feedwater startup valve is used to initially fillthe boiler. Depending upon the design, this can bethrough the main feedwater pumps or thecondensate pumps. The valve transitionsoperation to the feedwater regulator valve, orvariable speed drive, once drum pressure hasbeen built up. During drum fill operation, the boileris under minimal pressure. This causes the entirepressure drop to be taken across the feedwaterstartup valve. Because of this, the formation ofcavitation becomes a concern. Sizing of thestartup valve must be done in combination withthe feedwater regulator valve. This is to ensurethat the feedwater regulator valve does notexperience any service conditions that lead todamaging cavitation. The most common split isthat 80% capacity in the startup valve is equal to20% capacity in the regulator valve. Once thetransition to the regulator valve has begun, thestartup valve closes. Improper use is one of themain issues surrounding two valve feedwatersystems. For example, the startup valve is notbeing used at all and the regulator valve is beingused to perform both functions. This can be amajor problem if the boiler feedwater regulatorwas not sized or selected to perform bothfunctions. There can also be an issue if switchingbetween the startup and the regulator valve ishappening too quickly. Because of the cavitationconcerns and taking the full pressure drop, thestartup valve should utilize some form ofanti-cavitation trim. Typically, in process plants,since the pressures are not as high as powerplants, Cavitrol III trim is selected. 440C trim isrecommended for the case of treated feedwater.For cases where one valve is performing thestartup and regulator duties, characterized Cavitrol

18−3

trim can be designed to handle the cavitatingconditions at startup and then standard equalpercentage or linear characteristic for steady-stateconditions to maximize capacity. Another commonissue in both the startup and regulator valves is tosee them operated below the minimum operatingpoint. This can cause “gear-toothing” damage onthe plug.

Design Considerations:� Cavitation

� Tight shutoff (Class V)

� Typical process conditions are 800-1200 psigand 200-400 °F

Typical Specifications:� easy-e, HP, or EH

� Cavitrol III Trim

� HTS1 option with improved pressure balanceseal

� FIELDVUE digital valve controller with lowtravel cutoff

Optional:� NotchFlo or DST

� Protected inside seat

Steam GenerationThe number and types of boilers used for steamproduction varies considerably from mill to mill.Figure 18-1 indicates a simple system consistingof one power boiler and one recovery boilerdischarging into a common high pressuresuperheated steam header. For this system, therecovery boiler is base loaded at a constant flowof black liquor fuel with steam flow and pressureallowed to fluctuate. The steam header pressure(typically 1000-1500 psig) is controlled by varyingthe fuel input to the power boiler. Power boiler fuelis typically base loaded with bark or hog fuel andsupplemented with coal, oil, or gas.

Figure 18-2 provides an enlarged view of theupper convective section of a boiler. BFW enters

the economizer at 800-1200 psig and 200-400°Fbefore flowing to the steam drum of the generatingsection. As mentioned earlier, demineralized wateris used in boilers due to high operating pressuresand temperatures. Even so, as saturated steamleaves the steam drum, trace amounts of solidsare left behind. These solids must be removed viacontinuous bleed or blowdown of a small amountof water to prevent accumulation. The mud drumis also a low point for solids to settle and hasprovision for intermittent blowdown to preventaccumulation.

Saturated steam leaving the steam drum passesthrough the superheater section for further heatingand moisture evaporation. Most superheatersconsist of a primary and secondary section.Attemperation or desuperheating is used betweenthe sections to control final temperature andprevent overheating of tubes. The source of watermust be of demineralized quality to preventaccumulation of deposits on the inside of thetubes. A common source is boiler feedwater fromthe discharge of the boiler feedwater pump.

A vent is indicated on the superheated steamoutlet before the high pressure steam header. Thisvent may serve multiple purposes. One use is toclear the superheater of any moisture duringstart-up. This is to assure no water droplets reachthe steam turbine. A second function is pressurerelief in case an alarm indicates a build-up ofpressure. A final function may involve setting thevalve to open on high pressure just before thespring operated safety valve would lift. Due to highflow and pressure drop creating excessive noise,the valve is often used in series with a diffuserand/or silencer. Also shown in figure 18-2 is avalve for controlling the flow of steam to thesootblowers.

Sootblower ValveWhen firing fuels such as coal, oil, biomass, orother waste products, fouling of the boiler tubesbecomes a concern. Deposits from thecombustion process can collect on the heatexchanging tubes reducing thermal efficiency andcan cause operational difficulties. In order to keepthe unit operating, an online cleaning method mustbe used. This is usually accomplished by usingwhat are called sootblowers. Sootblowers utilizeflowing media such as water, air, or steam toremove deposits from boiler tubes. Widespreaduse of water has been limited due to the possibilityof thermal shock on the tube banks so steam isthe most common media. There are severaldifferent types of sootblowers used. Wall blowers

18−4

are used for furnace walls and have a very shortlance with a nozzle at the tip. The lance rotates asit moves into the furnace and cleans the depositsfrom the wall in a circular pattern. Retractablesootblowers are used in high flue gas temperaturezones. These operate the same as the wallblower, but the lance is inserted into the boiler toclean the internal tubes and can be partially orfully retractable. Partially retractable sootblowersare used where sootblower materials canwithstand the flue gas temperature.

Design Considerations:� Noise and vibration

� Tight shutoff (Class V)

� High cycling operation

� Typical service conditions are 800-1200 psigat 300-500 °F

Typical Specifications:� easy-e, HP, or EH

� Whisper Trim

� FIELDVUE digital valve controller with lowtravel cutoff

Optional:� Oversized stem/VSC and/or welded stem

connection

Steam Turbine GeneratorsThe majority of steam from the high pressureheader is used by large power generating steamturbines. Most mills use a backpressure turbine(s)(discharges to a lower pressure process header)and at least one condensing turbine (discharges toa condenser). Extraction steam from the turbinesis used to supply the medium and low pressureprocess headers. These headers typically operateat 400-600 psig and 60-150 psig respectively.Pressure reducing valves are also used betweenheaders to balance demand vs. extraction or toprovide process steam during a turbine outage. Ifdesuperheating is required, a steam conditioningvalve is recommended for this service.

While electrical power produced by the turbines(typically 30-70 MW) is important to milloperations, supplying the process with steam is ofprimary concern. Most mills are connected to alocal utility and purchase the balance of electricalpower required. By nature, the back pressureturbine provides more than double the utilization ofavailable fuel energy as the condensing turbine.The majority of steam discharged from the backpressure turbine is utilized by process, while thelatent heat in the steam of the condensing turbineexhaust is wasted in the condenser.

Main Steam PRV and TurbineBypassControl of steam pressures and temperatures arelikely the most critical applications in a pulp andpaper mill. Steam is used for wood chippreparation, process heating, pulp and paperdrying, boiler cleaning, energy production, and inmany other applications. Without steam, a pulpand paper mill cannot operate. To accommodate avariety of steam pressure requirements most sitesutilize three headers; high (1000-2000 psig),medium (500 psig), and low (100 psig) pressure.The power and recovery boilers supply highpressure, high temperature steam to the highpressure header. Much of the high pressure steamundergoes a pressure reduction and is directed tothe medium and low pressure headers. Whendemand for low pressure steam is high, themedium pressure header also supplies steam tothe low pressure header.

Main Steam PRVPressure reduction between headers can beachieved through the use of a pressure reducingvalve (PRV) or a steam turbine (also called turbogenerator). Main steam PRVs are often used tobridge the high (1000-2000 psig), medium (500psig), and low (100 psig) pressure headers. EachPRV can perform only a single pressure reduction,so multiple PRVs are required.

Design Considerations:� High pressure and temperature

� Noise and vibration

� Tight shutoff (Class V)

18−5

Typical Specifications:

� easy-e, EW, or HP

� Whisper Trim

� FIELDVUE digital valve controller with lowtravel cutoff

Optional:

� Inline diffuser (if extra noise attenuation isneeded)

Turbine BypassSteam turbines generate electricity throughpressure reduction and are becoming increasinglypopular. Each turbine can have multiple take-offpoints so one unit can simultaneously feed themedium and low pressure headers. In order tominimize unplanned downtime, a bypass valve isinstalled in parallel with the turbine to ensurepressure reduction occurs even when the turbineis offline. Most of the steam produced in papermills is not at the required conditions for allapplications. Thus, some degree of steamconditioning is warranted in either control ofpressure and/or temperature to protectdownstream equipment. Steam conditioningvalves represent state-of-the-art control of steampressure and temperature by integrally combiningboth functions within one control element unit.These valves address the need for better controlof steam conditions brought on by increasedenergy costs and more rigorous plant operation.These valves also provide better temperaturecontrol, improved noise abatement, and requirefewer piping and installation restrictions than theequivalent desuperheater and PRV.

Design Considerations:

� High pressure and temperature

� Noise and vibration

� Large turndown

� Tight shutoff (Class V)

� High cycling operation

� Stroking speed

Typical Specifications:� TBX

� Whisper Trim

� Bore-Seal

� FIELDVUE digital valve controller with lowtravel cutoff

Optional:� Separate PRV (easy-e, HP or EH) with a

Desuperheater

� WhisperFlo Trim

� 2625 booster(s)

Condensing and Cooling SystemEven though it decreases cycle efficiency, acondenser is essential to provide a “cushion” orlocation to dump steam when a portion of theprocess is down and still benefit from electricalpower production. The condenser is a shell andtube heat exchanger which operates at a vacuum.Cooling water passes through the tubes andcondenses the steam on the outside of the tubes.The cooling water passes through a closedsystem back to a cooling tower where the heat isdischarged to the atmosphere. Due to seasonaltemperature variations, all cells of a cooling towerare not always in use. Butterfly valves are used toisolate cells or even bypass the cooling tower.

Condensed steam accumulates at the bottom ofthe condenser in the hotwell. The condensate isthen pumped to the condensate collection tank tobegin the cycle again. Since the condensate isnear saturation, a minimum level must bemaintained to prevent pump cavitation and aminimum flow is required to prevent overheating.

Condensate Recirculation ValveThe condensate recirculation valve is similar to thefeed pump recirculation valve in that it alsoprotects the pump from cavitation. Inlet pressureand temperature differ from the feedwater system.The dissimilarities from the feedwater systeminclude the inlet pressure and temperature. Inletsizing often indicates that flashing is occurring,however, experience shows this is always a

18−6

cavitating application. The end user needs toensure that there is not a sparger or diffuserdownstream emitting back pressure on the valve.This will cause cavitation rather than flashing.Tight shutoff is needed on this application becauseit prevents loss of condenser vacuum, loss ofcondensate pressure and flow to the deaerator,and saves money in terms of wasted pumphorsepower.

Design Considerations:

� Cavitation

� Tight shutoff (Class V)

� Typical service conditions are 300-500 psi at100-150 °F

Typical Specifications:

� easy-e, HP, EH or Cavitrol IV trim

� Cavitrol III Trim

� HTS1 option with Improved PressureBalance Seal

� FIELDVUE digital valve controller with lowtravel cutoff

Optional:

� NotchFlo or Dirty Service Trim

� Protected Inside Seat

Control Valve Selection

PROCESS FISHER VALVE PRODUCT DESIGN

ValveTag #

Water/Steam Power CycleV150/V300

V500 Control-Disk

E-Body EH HPSteamCondi-tioning

TypicalValveSizeApplication Description Control

Function

LV-1 M.P. Heater Drain T S P 2�

LV-2 L.P. Heater Drain T S P 2�

LV-3 Condensate Collection Tank Level T P P 4�

LV-4 Demineralized Make-up Water T P S 4�

FV-1 BFW Regulator T S P 6�

FV-2 BFW Regulator Bypass T S P 2�

FV-3 BFW Pump Recirculation T S P 2�

FV-4 Deaerator Heating Steam T S P 12�

FV-5 Condensate to Deaerator T P S 4�

FV-6 Contaminated Condensate Dump O/O P S 4�

FV-7 Condenser Hotwell Recirculation T P S 2�

FV-8 Condenser/Cooling Tower Water O/O P 18�

FV-9 Sootblower Steam O/O P S 4�

FV-10 High Pressure Steam Vent T S P 4�

FV-11 Superheater Attemperation Water T S P 1�

FV-12 Continuous Blowdown T P (HPA) 1�

FV-13 Intermittent Blowdown O/O P (HPA) 1�

PRV-1 High/Medium Steam PressureReducing

T P S (TBX) 8�

PRV-2 Medium/Low Steam PressureReducing

T P S (TBX) 8�

P=Primary Valve ChoiceS=Secondary Valve ChoiceT=Throttling ServiceO/O=On/Off Service

18−7

E13

85

Figure 18-1. Water/Steam Cycle Diagram

18−8

Figure 18-2. Power or Recovery Boiler Upper Convective Section

E1386

D103540X012 Printed in USA / 3M / 01-11 / IL

Emerson Process ManagementMarshalltown, Iowa 50158 USASorocaba, 18087 BrazilChatham, Kent ME4 4QZ UKDubai, United Arab EmiratesSingapore 128461 Singaporewww.EmersonProcess.com/Fisher

The contents of this publication are presented for informational purposes only, and while every effort hasbeen made to ensure their accuracy they are not to be construed as warranties or guarantees, expressor implied, regarding the products or services described herein or their use or applicability. All sales aregoverned by our terms and conditions, which are available upon request. We reserve the right to modifyor improve the designs or specifications of such products at any time without notice. Neither Emerson,Emerson Process Management, nor any of their affiliated entities assumes responsibility for the selec-tion, use, or maintenance of any product. Responsibility for proper selection, use, and maintenance ofany product remains solely with the purchaser and end-user.