ON THE RESEARCH AREAS OF APPROACH FLOW SYSTEMS

Project F004

Report I

to the

MEMBER COMPANIES OF THE INSTITUTE OF PAPER SCIENCE AND TECHNOLOGY

March 1996
 INSTITUTE OF PAPER SCIENCE AND TECHNOLOGY

Atlanta, Georgia

ON THE RESEARCH AREAS OF APPROACH FLOW SYSTEMS

Project F004

Report 1

A Progress Report

to the

MEMBER COMPANIES OF THE INSTITUTE OF PAPER SCIENCE AND TECHNOLOGY

By

Wang, X.

March 1996
ProjectF004 Report1

Executive Summary

Institute of Paper Science and Technology

In this report, the results from the informal survey on the research areas of ap-

proach pipe systems are summarized. Current guidelines from TAPPI Technical In-

formation Sheet (TIS), Valmet, Voith, and Beloit are also reviewed and discussed.

Based on the survey results and the available information of the past research done in

this area, after having completed many telephone interviews, field trips in the three

major papermaking machine suppliers, and several mill visits, let alone dozens of

papers in this field, it is concluded worthwhile to establish a corresponding funded

project on approach pipe systems at IPST. The objectives of this project are to solve

some immediate problems relating to approach pipe systems and to obtain deeper fun-

damental understanding that can be used in the future designs. Specifically, based on

the state-of-art computational procedures and the development of concepts in fluidics

and hydraulics, computer simulations of different approach flow components and their

interactions within the whole system will be developed and analyzed. Some softwares

for analyzing and optimizing the approach flow systems according to different mill

conditions are to be delivered through this research project.

The project will provide general information for all of our membership compa-

nies and focus on the fundamental research of our paper industries facing the coming

information era. Although it is not the intention to associate this project with any

particular companies or interests, some specific research of particular interest to in-

dividual companies can be arranged separately.

With projections for the future, the commencement of this project will help to

establish an expertise center at IPST in the approach flow system areas and eventually

for the whole wet end papermaking processes.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
Project
F004 Report
1

Acknowledgments

First, I would like to thank Professor David Orloff for suggesting this interesting field
to me.

I am very grateful to Dr. David White and his papermaking PAC members,
especially Mr. Richard Reese, Mr. Louis M. Vance, and Mr. Karl Christianson, who
serve as the approach pipe system subcommittee.

It is my pleasure to have the privilege to work at IPST with so many wonderful

faculty, research staff, and students. Part of the inspiration of getting evolved in this
area is due to Professor Cyrus Aidun's recent success in applying numerical procedures
to the coating and headbox analyses for our paper industries.

My utmost gratitude is due to Mr. Seppo Laaksonen, Dr. Jay A. Shands, Mr.

Michael MacGregor, and the engineers working with them; their tremendous support
and generosity during the preparation of this proposal mean the most to me.

IPST ConfidentialInformation - Not for Public Disclosure

(ForIPSTMemberCompany'sInternalUseOnly)
Project
F004 Report
1

Contents

Titlepage I

Abstract 2

Acknowledgments 3

Contents 4

List of Figures 7

List of Tables 8

I Introduction 9
1.1 Overview
.................................. 9

1.2 Project Objective ............................. 11

1.3 Project Outline .............................. 12

2 Evaluation of Industry Needs 13

2.1 Flow-Induced Vibration ......................... 13
2.2 Fluid FlowControl ............................ 14

2.3 DeaerationSystems ............................ 14

2.4 Other Areas ................................ 14

3 Current Guidelines 21

IPST ConfidentialInformation - Not for Public Disclosure

(ForIPSTMemberCompany'sInternalUseOnly)
ProjectF004 5 Report
1

3.1 StuffBox ................................. 21

3.1.1 Valmet's Recommendation .................... 21

3.1.2 TIS' Recommendation ...................... 22

3.1.3 Voith's Recommendation ..................... 23

3.1.4 Beloit's Recommendation .................... 24

3.2 Wire Pits and Silos ............................ 25

3.2.1 Beloit's Recommendation .................... 26

3.2.2 Voith's Recommendation ..................... 28

3.2.3 Valmet's Recommendation .................... 29

3.3 Fan Pump .......... -. ...................... 30

3.3.1 Voith's Recommendation ..................... 30

3.3.2 Beloit's Recommendation .................... 31

3.3.3 TIS' Recommendation ...................... 31

3.4 Piping Arrangement ........................... 32

3.4.1 General Recommendations .................... 33

3.4.2 TIS' Recommendation ...................... 33

3.4.3 Voith's Recommendation ..................... 34

3.5 Flow-Induced Vibration and Variations ................. 35

3.6 Location of Valves ............................ 37

3.7 Elbow ................................... 38

3.7.1 Voith's Recommendation .......... ........... 38

3.7.2 TIS' Recommendation .... .................. 38

3.8 Discussions ................................ 39

4 Research Approaches 41

4.1 Engineering Approach ........................... 41

4.2 Displacement/Pressure (u/p) Formulation ............... 43

4.3 Pressure Pulsation Attenuator ...................... 45

4.4 Pipeline Simulation ............................ 46

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Intemal Use Only)
ProjectF004 6 Report1

4.5 SiloDesign ................................ 51

4.6 Vortex Forming .............................. 51

4.7 Application of Fluidics and Hydraulics ................. 52

4.8 Unsteady Flow in Piping System .................... 52

5 Research Planning 56

6 Conclusions 58

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
Project
F004 Report
1

List of Figures

3-1 Stuffbox arrangement........................... 23

3-2 Mixing pipe arrangemen t ......................... 32

4-1 Pressure pulsation attenuator mathematical model ........... 46

4-2 Analogy to electrical devices ........................ 48

4-3 Silo design ........... ...................... 49

4-4 An overview of the computer simulation of the approach flow system

(eventually the whole wet end processes.) ................ 50
4-5 Development of vortices .......................... 51

4-6 Prandtl Diffuser .............................. 53

4-7 Tesla's Diode ................................ 53

4-8 The velocity profile is changed by varying the control input ...... 54
4-9 Turbulence amplifier ............................ 54

IPST ConfidentialInformation - Not for PublicDisclosure

(ForIPSTMemberCompany'sInternalUseOnly)
ProjectF004 Report1

List of Tables

2-1 Sample _1 survey results ......................... 15

2-2 Sample _2 survey results ......................... 15

2-3 Sample _3 survey results ......................... 16

2-4 Sample _4 survey results ......... . ............... 16

2-5 Sample _5 survey results ......................... 17

2-6 Sample _6 survey results ......................... 17

2-7 Sample _7 survey results ......................... 18

2-8 Sample _8 survey results ......................... 18

2-9 Sample _9 survey results ......................... 19

2-10 Sample _10 survey results ......................... 19

2-11 Sample _11 survey results ......................... 20

5-1 Tasks and time line............................ 57

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
Project
F004 Report
1

Chapter I

Introduction

1.1 Overview

The approach pipe (or flow) system is comprised of numerous individual components
between the stock preparation section and the headbox [35] [38] [23]. These com-

ponents include fan pumps, piping, deaeration system, attenuator, pressure screens,
cleaners, control valves and white water recycling systems. The effects of each com-

ponent in the approach pipe system will directly influence the performance of paper
machines.

As the development of more high production rate forming devices and short dis-

tance forming units (such as hydraulic headboxes and twin-wire forming sections)
[17], the pressure pulsation tolerance and consistency fluctuation within the stock

(mixture of wood fibers, chemical additives, and water) are getting more strict. In
fact, the good performance of stock approach pipe systems is essential to the smooth
operation of the entire paper machines with higher and higher machine speeds.

From a design standpoint, the stock approach system is one of the most critical

areas of the paper machine and is often called "plumbers' nightmare." The current
design guidelines are normally used to avoid the following difficulties:

· non-uniform fiber delivery;

IPST ConfidentialInformation - Not for PublicDisclosure

(ForIPSTMemberCompany'sInternalUse Only)
ProjectF004 10 Report1

· unstable flow, pulsations and surges;

· formation of stack lumps from stagnant zones, trapped air, rough surfaces; and

· vibration.

Although the arrangement, design, and operation of stock approach equipment will

vary greatly from mill to mill and machine to machine, and the available guidelines

and research information are often found to be inadequate and even confusing [36] [20]

[37] [21] [14], there is no tool available to analyse the approach flow system as a whole

under the circumstance of component changes, consistency and stock variations [11]

[12] [39]. Therefore, to provide engineers the general design guidelines, more feasible

analysis tools for both the specific components within approach flow systems and the

whole system design are imperative.

It is the intention of this project to identify the fundamental research areas of

approach pipe systems [32] and to provide the state-of-art solution techniques. The

research results and procedures shall be widely applicable to both paper machine

suppliers and mills.

In order to establish a robust and worthwhile project in this area, the needs

and interests of the IPST membership companies were investigated, and based on

the available published research results, the following three main research areas are
identified:

· flow-induced vibration, including pressure pulsation and dynamic analysis

of approach pipe systems [2] [3] [34];

· fluid flow control, including chemical additive (such as retention aids) mixing

[25] [15], uniform consistency, stock and white water mixing; and

· deaeration systems, including cavitation, flow separation, and deculator sys-

tem [11] [26].

IPST Confidential Information - Not for'Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 11 Report1

The other areas related to approach pipe systems are surface finish [24] and pipe

system design. The surface finish problem is mainly an issue of metallurgy and
corrosion analysis; therefore it is not considered as one of our main research focuses.

The comments from papermaking PAC members were gathered through an infor-
mal telephone survey.

1.2 Project Objective

In engineering sciences, much effort has gone into the application of computer simu-

lations. With the development of more powerful computers and more effective com-

putational procedures, the traditional hydraulics and fiuidics are merging with the

concepts of computational mechanics [28]. We will, through this research project,

develop computer simulation models for the field engineers to use in accordance with

the specific mill and machine arrangement. The computer surrogates will contain'

· the mixing simulation within the pipe for fibres, water, and chemical additives;

· the computer simulation of individual components within the approach pipe

system;

· the nonlinear dynamics computer simulation of the hydraulic design of approach

flow systems; and

· the pressure pulsation simulation.

The specific objectives of this project are:

· to optimize the pressure pulsation attenuator for all ranges of frequencies, es-

pecially the lower range;

· to better understand the white water silo design;

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 12 Report1

· to better understand the mixing between stock and white water before the fan

pump;

· to better understand short- and long-term consistency variations;

· to optimize the pipe system design including the location of valves, elbow, and
pumps; and

· to develop a nonlinear dynamic model including the disturbance or perturbation

dampening mechanism.

1.3 Project Outline

Based on the evaluation of our membership industry needs, we will include three steps
in this project.

· Step 1' To identify the problems of individual components within approach

pipe systems and to form corresponding computer simulation models;

· Step 2' To incorporate all approach pipe system components into a whole

computer model and to develop an algorithm of computer system design simu-

lations; and

· Step 3' To implement the developed software in any of the test approach pipe

systems within our membership companies and to improve the current TAPPI

TIS guideline (if necessary, individual company's guideline as well, depending

· on the separate arrangement.)

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 Report1

Chapter 2'

Evaluation of Industry Needs

Stimulated by some membership industries' research needs [32], a telephone survey

was conducted on a potential project in the areas of approach pipe systems.

The following topics are discussed' fan pump, flow stability, flow-induced vibra-

tion, piping, valves, recirculation, cleaners, pressure screens, chemical additive mixing,
consistency control and design issues. However, discussions are focused on the afore-

mentioned three main issues; i.e. flow-induced vibration, fluid flow control and

deaeration systems.

The papermaking PAC members were asked how important these areas are, and

what aspects of approac h pipe systems are of most interest to them. In addition, they

were asked to comment on the other areas related to approach pipe systems.

2.1 Flow-Induced Vibration

Ten out of 11 interviewees agree that this is an important issue. It is reflected in their

comments that low frequency pulsation is more difficult to attenuate. In general,

the vibration of the approach pipe system will be introduced to the headbox and

eventually causes MD and CD base weight variations. The problem should be partic-

ularly considered for hydraulic headboxes. The sources of pulsation could be induced

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 14 Report
1

by insu_cient pipe supporting systems, flow-induced vibration, defective rotors, and

acoustic resonance with the flow.

2.2 Fluid Flow Control

Almost all papermaking PAC members I talked to consider this area to be interest-

ing. It was pointed out that the uniform consistency needs to be well understood.

The flow pattern which influences the stock consistency and fioccuation conditions,
thin stock, thick stock, and chemical additive mixing are of most interest to these

industry representatives. They feel that the existing guidelines are inadequate and
more hydrodynamic parameters are needed in practice.

2.3 Deaeration Systems

Six interviewees think that this is not an important issue in their perspective, com-
pared with five others on the other side. However, this problem is dependent on the

paper grades that individual mill and machine produce. For light weight grade paper
machines, it is still an issue of concern. Some PAC members mention the need for

a deeper understanding of cavitation and flow separation problems within different

pipes. The silo and wire pit are also of some interest. Some mills found that whether

or not to have deculator systems did not make much difference, and in order to save
room space, these deculators were withdrawn.

2.4 Other Areas

In addition to the three main topics, some papermaking PAC members commented

on the system design problem. The concern is about the placement of various com-
ponents, including the system design for new machines in old mills.

IPST ConfidentialInformation - Not for PublicDisclosure

(For IPST Member Company's Internal Use Only)
Project F004 15 Report 1

Area of study in How important is What aspects are of Other comments

approach pipe systems this area of study? interest to your industry?

Less tolerance for MD and

Flow-induced vibration Important CD base weight variation as
the machine speed increases

Machine chest, blend chest,

Flow control Important pressure screen, fan pumps, Guideline inadequate
and valves

Flow consistency, all kinds Pipe elbow

Deaeration system Important of mixing problems, more cavitation needs
hydrodynamic parameters to be investigated

Other areas Systems design is Placement of various

an important area components

Table 2-1: Sample _1 survey results.

Area of study in How important is What aspects are of Other comments

approach pipe systems this area of study? interest to your industry?

Pipe support and headbox Press section dynamic

Flow-induced vibration Important vibration causes MD _ CD behavior may
base weight variations affect the headbox

Retention aids mixing, silo

Flow control Important water level and pressure
variation

Deaeration system Important Fan pump, multisection

pressure drops

Other areas

Table 2-2: Sample _2 survey results.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
Project F004 16 Report 1

Area of study in How important is What aspects are of Other comments

approach pipe systems this area of study? interest to your industry?

Flow-induced vibration Important Influence on the headbox

Good way of consistency

Flow control Important control, zero variation

Deaeration system Important

System design, not Air deculator has

Other areas enough space for limited effects
various components

Table 2-3: Sample _3 survey results.

Area of study in How important is What aspects are of Other comments

approach pipe systems this area of study? interest to your industry?

Flow-induced vibration Important

Thin stock, thick stock

Flow control Important mixing, fan pump

Deculator is good
Deaeration system Not so important for lightweight
print

Other areas

Table 2-4: Sample #4 survey results.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
Project F004 17 Report 1

Area of study in How important is What aspects are of Other comments

approach pipe systems this area of study? interest to your industry?

Flow-induced vibration Not so important

Retention aids mixing,

Flow control Important uniform consistency, thin
stock, thick stock mixing

Depending on
Deaeration system Not so important different mills

Other areas

Table 2-5: Sample #5 survey results.

Area of study in How important is What aspects are of Other comments

approach pipe systems this area of study? interest to your industry?

Low frequency vibration High frequency

Flow-induced vibration Important affects the headbox and vibration is easy
is difficult to deal with to fix

Swirling affects additive &

Flow control Important stock mixing, consistency, Poor guideline
CD base weight variation

Deaeration system Not so important

Other areas

Table 2-6: Sample #6 survey results.

PST Confidential Information - Not for Public Disclosure

(For PST Member Company’s Internal Use Only)
Project F004 18 Report1

Area of study in How important is What aspects are of Other comments

approach pipe systems this area of study? interest to your industry?

More and more

Flow-induced vibration Important interests

Machine chest, consistency

Flow control Important control, and faster silo
water

Deaeration system Not so important Pipe design

Other areas

Table 2-7: Sample _7 survey results.

Area of study in How important is What aspects are of Other comments

approach pipe systems this area of study? interest to your industry?

Pressure pulsation and

Flow-induced Vibration Important recirculation line

Thick stock _ white water

Flow control Important mixing, uniform mixture,
and where to put additives

Deculator system, Silo

Deaeration system Important and wire pit

Other areas

Table 2-8: Sample _8 survey results.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
Project F004 19 Report 1

Area of study in How important is What aspects are of Other comments

approach pipe systems this area of study? interest to your industry?

Flow-induced vibration Critical Fan pump pulsation

Short circulation system

Flow control Critical on paper machines

Deaeration system Critical

The design of
Other areas Fan pump design entire system
is of interest

Table 2-9: Sample _9 survey results.

Area of study in How important is What aspects are of Other comments

approach pipe systems this area of study? interest to your industry?

Too much pressure drop,

Flow-induced vibration Important MD base weight variation

Mixing velocity, stock

Flow control Important white water mixing,
headbox circulation line

Upset fan pump, too much

Deaeration system Important pressure drop, cavitation,
diffuser, MD variation

Practical side
Other areas Fan pump cracks of research

Table 2-10: Sample _10 survey results.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
Project F004 20 Report 1

Area of study in How important is What aspects are of Other comments

approach pipe systems this area of study? interest to your industry?

Flow-induced vibration Important

Thick stock and thin

Flow control Important stock mixing

Deaeration system Not so important Pipe design

Hope to know
Otherareas moreabouthow
We are doing

Table 2-11: Sample _11 survey results.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 Report1

Chapter 3

Current Guidelines

The design guidelines of approach flow systems are results of accumulated field expe-

riences with installations, combined with basic hydraulic principles. These guidelines

are generally recommendations for good system designs and should not be miscon-

strued as absolute design parameters. Therefore, for each individual system, due to

the mill condition, specific recommendations are necessary.

Recommendations of TAPPI TIS, Valmet, Voith, and Beloit's guidelines on some

approach pipe components' design are extracted in this chapter [36] [20] [37] [21] [14].

In the discussion section, we compare these design recommendations, and rationalize

why these guidelines are not adequate, how some of them conflict, and what the

future work will be to make the design process more practical and mill-specific.

3.1 Stuff Box

To ensure a stable heavy stock supply to the headbox approach pipe systems, some

design guidelines suggest the proper use of a stuff box (shown in Fig. 3-1.)

3.1.1 Valmet's Recommendation

In Valmet's engineering instructions [21], the following two criteria are proposed:

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal. Use Only)
ProjectF004 22 Report
1

1. General

The thick stock is fed into the fan pump suction either with stuff box or pump.
The stock (controlled about 3% consistency) is fed through basis weight valve and
quick release, or only through quick release valve, to the fan pump suction side.
'2. Stuff box construction

The stuff box shall be placed at least 4.5 m (14.8 ft) above the white water off
machine silo level to ensure sufficient inlet flow speed and pressure of the thick stock.

The connection of the thick stock inlet pipe and feed pipe to the stuff box should
be made with a conical pipe to calm the thick stock flow in connection with the stuff
box.

In regard to the Optimum flow conditions of the fiber suspension, the stock inlet

connection is situated in one end of the stuff box bottom, discharging connection in
the center compartment and overflow connection in the far end.

The overflow amount of thick stock in the stuff box can be best adjusted with
the adjustable-speed pump between the machine chest and stuff box, and thus avoid
too much overflow from the stuff box back to the machine chest. Suitable overflow

amount is 10- 20% of the inlet flow. The overflow shall take place laminarily.

The basis weight valve is to be dimensioned for its best possible operational range
and placed at least 1.5m (4.9 ft) below the wire pit liquid surface to avoid risk of

cavitation. (Dimensioning instructions for stuff box is referred to in ref. [21])

3.1.2 TIS' Recommendation

In the TAPPI 1995-1996 Technical Information Sheets (TIS), the following guidelines
are listed:

· Uniformity of basis weight in the machine direction can be no better than the

uniformity of fiber supplied by the thick stock pipe. Slurry consistency, volume,
and stability must be controlled with precision.

IPST ConfidentialInformation - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 23 Report1

Free surface

Overflow _ Inlet
To the pump suction

Figure 3-1' Stuff box arrangement.

· Monitor flow with stuff box or magnetic flow meter.

· Locate stuff box as close as possible to the pump suction.

· Figure 3-1 shows the stuff box arrangement.

3.1.3 Voith's Recommendation

If possible, the thick stock to the inlet of the fan pump should be controlled by means

of a constant level (stuff) box [36].

· This box must be designed to avoid dead corners or uneven flow characteristics,

while allowing a free surface to provide free air an opportunity to be released.

· When the stuff box is positioned very near the machine chest, the constant

overflow can be returned directly to the machine chest.

IPST Confidential Information- Not for Public Disclosure

(For IPST Member Company's IntemaI Use Only)
ProjectF004 24 Report1

· Use of a constant overflow provides a slight cushion against minor upsets in the

thick stock system, and helps avoid foam or dirt build up on top of the stock.

· In order to provide sufficient dP for the basis weight control valve, the stuff box

level should be 4m (13 ft) above the silo/wire pit level from which the fan pump
draws its water.

· The control valve can be either a motor operated V-port ball valve or a special

inclined regulating gate valve (with a narrow gate).

· Regardless of the design of the valve, it must be located below the silo/wire pit

level, to avoid cavitation in the line. For best response time, the valve should

be located as close to the fan pump as feasible.

3.1.4 Beloit's Recommendation

A stable system begins with a stable flow of heavy stock to that system. Proper use of

a stuff box is the best way to insure this stability. By feeding the bottom of one end,

discharging to the machine through a vertical dropleg from the center compartment

and overflowing the far end, the box accomplishes five things'

· It eliminates much of the free air in the stock.

· It isolates the refining system from the headbox flow loop, minimizing the effects

of pulsations from the preceding system. (Valves, Refiners, Pumps, etc.)

· It allows proper back pressuring of refiners, independent of head and flow re-

quirements to the machine.

· It provides a constant head reference for the basis weight valve.

· It allows relatively constant thick stock flow to the refiners by bypassing the
excess back to the machine chest.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
 Project FO04 25 Report 1

Correct stuff box operation can be achieved with the following guidelines [14]:

l All interior surfaces in contact with stock or additives must be adequately cor-
rosion resistant. All interior welds and surfaces must pass cotton tests.
,
0 Eight seconds deaeration time are required from points 11
X to
77 11
Y. 77

l Liquid level is to be 180 to 240 in (4500 to 6000 mm) above silo liquid level.

l The minimum submerged depth in feet should equal the maximum velocity in
fps at “E” or 48 in (1200 mm), whichever is greater.

Notice that the basis weight valve in the machine supply dropleg must be located
at a low level to insure submergence and reduce valve cavitation. This prevents
erratic pressure pulsations which accompany cascading flow, as well as promotes line
cleanliness. Always overflow the stuff box in the direction of flow to eliminate possible
foam build up. The stuff box should be located as close as possible to the silo to
minimize the length of piping to the fan pump suction.

3.2 Wire Pits and Silos

The other necessary supply to the headbox approach pipe system is the white water
from wire pits and silos, which are the major factors in system stability. Besides
being a collector for former drainage, they provide a stable suction head for the fan
pump and act as a deaeration device.
In order to maintain stock consistency in the machine chest, no return flow or
make-up flow should be fed into this chest without having its consistency controlled.
The consistency control, as well as any brush refiners, should be positioned before
the machine chest. This will allow the large volume of the chest to act as a buffer
against consistency variations, as it mixes and equalizes the stock.

IPST Confidential Information - Not for Public Disclosure

(For PST Member Company’s Internal Use Only)
ProjectF004 26 Report1

This requires an adequate mixing action within the machine chest to avoid sepa-

ration of stock and water. Inefficient or faulty consistency control after the machine

chest will have a direct effect on MD weight variations as well as CD profiles.

The use of properly designed wire pits and silos will provide deaeration and stabil-

ity. Two basic types of silos are the under- and off-machine silo. The design criterion

is to remove free air, to provide a constant fan pump head without vortexing, to pro-

vide constant back pressure, and to have a receptacle for introduction and deaeration

of make-up waters.

3.2.1 Beloit's Recommendation

Beloit is concerned with the design of wire pits and silos primarily because of their

utility in providing deaeration and stability. A considerable amount of air (2- 4% by

volume) is entrained in the white water in passing through the forming wires. This

is unavoidable. Most of this air can be removed, however, with proper wire pit and

silo design techniques [14].

The two basic types of silos are the under-machine type, where the pit is located

directly below the former and contained within the sill beams, and the off-machine

silo [14], located on the drive side of the former with connecting flow channels to the

saveall troughs. The basic requirements are the same for either, and are independent

of the former type for proper operation.

The under-machine type is normally used to save space on conventional Four-

drinier machines. Downspouts discharge the water drained by the table elements into

the channel at the breast roll end of the pit. The flow is directed to the couch end at

slow average velocities (1 to 1.25 feet per second, or 0.3 to 0.4 mps) where the flow

reverses and is returned to the silo located at the breast roll end. The relatively long

period of open channel flow is desirable to release air and dissipate turbulence (up to

60 seconds if possible).

The off-machine silo is often used when clays and other fillers are present and with

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 27 Report1

most twin-wire arrangements. Channels are often shortened and some deaeration time

sacrificed to prevent settlement of filler. A 30-sec dwell time is still minimum with this

design for flow stability. An inherent feature of the off-machine silo is the automatic

separation of table drainage and shower water.

The following design principles are offered as guidelines:

· The minimum dwell time for deaeration in the flow channel is 30 sec at the

required velocities [14]. (60 sec is a desirable target to achieve maximum deaer-

ation if filler separation is not a factor.)

· Liquid level in channel should be 6 to 24 in below bottom of former saveall to

prevent possible flooding of the saveall pans.

· The bottom floor should slope downward as shown, to maintain self-drainage

at shut down.

· Typical slopes, shown upwards to prevent air accumulation and subsequent

slime build up.

· Adjustable weir to permit level control at least 6 in (150 mm) below to 12 in

(300 mm) above design weir setup dimension.

· A minimum continuous silo overflow of approximately 1000 gpm is recom-

mended to maintain a stable and foam free liquid level at the fan pump suction.

· Assuming 100% of total gallonage through wet end downspouts, use 1.25 fps

(0.4 mps) to find liquid depth. (Maximum for good deaeration is depth 3 to 5

ft or 0.9 m to 1.8 m). (Nominal design depth should be 42 in.)

· All interior surfaces of the silo and wire pit to be tile-lined or equivalent.

· Use line size giving 6.0 to 8.0 fps between basis weight valve and fan pump

suction.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 28 Report1

· Alt. 1 of section ¥-Y or Alt. 2 of section Y-Y can be used for entrance into the

flow channel [14].

· Use Alt. 1 of section X-X for single dilution system.

· Use Alt. 2 of section X-X for double dilution system.

· Outlets may be positioned at any 90 ° angle to the silo.

· Use baffle with double dilution to prevent migration. Baffle height should be

approximately 3.0 ft above top edge of pump suction.

· The end of pipes going concentric to the pump suction should be cut at a 30 °

to 45 ° angle to prevent "burping."

A properly designed wire pit/silo arrangement will remove free air, provide a

constant fan pump head without vortexing, provide constant back pressure for the

basis weight valve, and provide a receptacle for introduction (and deaeration) of make-

up waters. A properly designed wire pit/silo arrangement should reduce air content
in stock to under 1.0%.

3.2.2 Voith's Recommendation

The level of the white water supply reservoir (silo, wire pit, or seal pit) must be kept

constant. Variations in this level will directly effect the quality of the final product

[36].

· This is best accomplished by means of an overflow. This allows a cushion for

sudden process changes, and long-term drainage variations.

· The maintenance of a constant level is essential. This level provides the inlet

pressure against the fan pump and the back pressure against the basis weight

control valve and various return flows.

.IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 29 Report1

The white water supply reservoir must provide adequate deaeration of the water.

· Retention time should be 0.5 to 1.0 min or greater.

· Downward velocities should not exceed 0.5 fps, to allow air a chance to rise to
the surface.

· Maximize surface area to provide maximum opportunity for the release of air.

· Discharges into the reservoir should avoid high velocities and "waterfalls" that

may entrain additional air.

Reservoir design must avoid stagnant areas (dead corners). Whenever possible, if

space is available, the use of a white water silo over a wire pit or tank is recommended.

· This minimizes the potential cascading of white water.

· Reduction in air entrainment reduces the need for and expense of foam retar-
dants.

· Inlet pressure to the fan pump is increased, reducing head and energy require-
ments.

· Operation of an auxiliary screen (vibrating) is enhanced as its discharge (if set

on the machine room floor elevation) can be directed into a flooded collection
tank.

· Start-up requires less "filling" of the system. Hence, stable operation is achieved

more readily.

3.2.3 Valmet's Recommendation

The silo is another area in which su_cient retention should be allowed for the liber-

ation of air. The downward velocity in the silo should be around 0. 5 ft/sec to allow

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 30 Report1

entrained air to escape and the retention time should be approximately 60-90 sec.

Often, the silo level is raised in order to increase the NPSHa for the fan pump or

mixing pump. This is detrimental to the process because the retention time is low-

ered and the air must travel a greater distance to be liberated, possibly worsening the
problem. The tray water must not be allowed to cascade into the silo for the same

reasons indicated in the overflow lines.

3.3 Fan Pump

In sizing a fan pump, care must be taken that the pump will operate at a maximum

impeller rpm over the entire range of flows and consistencies. There are two types of

fan pumps, one is the variable speed centrifugal and the other is the constant speed
one.

3.3.1 Voith's Recommendation

For the variable speed pumps, the following guidelines are suggested'

· Vane frequency at low rpm conditions should not be less than 60 Hz.

· Normally used when variation in flow downstream of the fan pump is not detri-

mental to system performance. Such as no flow sensitive cleaning equipment,

double dilution system, or a modified bypass system (if cleaners are not ex-

tremely sensitive to flow variations a small bypass can be utilized after the

cleaners to control a dP range across the cleaners, while offering variable speed

pump flow control to the headbox.)

· Energy efficient.

· Simplified, cleaner, and smaller approach flow system, through the elimination

of bypass flows and throttling control valves.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 31 Report1

· Reduce opportunity for pulsation generation.

· Reduce opportunity for consistency variations.

· Smooth headbox pressure control.

· During operation, the high points of the pump casing should be bled continu-

ously for air.

3.3.2 Beloit's Recommendation

· Pumps should be selected for a maximum impeller rpm to help raise the fre-
quency beyond the troublesome area.

· The impeller should have a maximum number of vanes.

· The impeller vanes should be split, staggered, and skewed if a double suction

pump is selected.

· The impeller must be properly balanced and obviously must not operate near

critical speeds. The impeller must be hydraulically as well as mechanically

balanced; that is, there should be provision for equal flow through each of the

impeller channels.

· Impeller deflection must be minimized.

· The suction head of the pump should well exceed maximum NPSH ("Net Pos-

itive Suction Head") requirements.

3.3.3 TIS' Recommendation

· Double suction type; split impeller with staggered vanes is recommended.

· Variable speed drive with good controls is recommended.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 32 Report1

Type I Type II

Figure 3-2' Mixing pipe arrangement.

· If oversized, avoid excessive throttling; reduce impeller diameter (check pump

manufacturer's recommendations) or reduce speed.

· The inlet should be large to maintain a low velocity.

· Locate inlet to obtain a symmetrical flow to the pump whenever possible.

3.4 Piping Arrangement

Proper flow velocities must be maintained in any piping system. As stock travels

through a system, it has a tendency to flocculate. If the proper velocity range is

maintained, the stock will be less likely to flocculate and remain well mixed without

creating undue turbulence.

Suppose we now have the uniform flows from thick stock, white water, and chemi-

cal additives, the mixing before the fan pump will determine the final stock approach-

ing the headbox. There are two ways of mixing the lines show n in Fig. 3-2, one is to

use concentric design and the other is to use an alternate top entry arrangement.

For the concentric design, the velocity of thick stock is higher than the white water

from the wire pit and silo. For the pipe intersection, less than 45 ° is recommended

to avoid burping.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 33 Report1

One of the most important design considerations for this system is how to provide

the best mixing of cleaned stock with white water in the fan pump. Cleaned stock

normally increases the velocity of the mixture as it flows toward the fan pump, the

final mixing takes place in the pump.

3.4.1 General Recommendations

· Make pipe lines as short as possible.

· Maintain flow velocity in the range of 5 to 15 ft/sec.

· Use noncorrosive pipe with smooth inside finish.

· Grind internal welds, remove weld spatter and polish with fine abrasive to assure

acceptable smoothness. Check for residual roughness using the cotton ball test,

particularly for pulps which easily form lumps and strings.

· Pipes should fit together smoothly. Fit gaskets carefully to avoid protrusion into

the flow. Particularly between screens and headbox, use ground metal-to-metal

flanges (Metlap, ¥anstone or equivalent).

· Avoid stagnant pockets of stock.

· Install drains at low points. Drain plugs should be flush with the inside wall.

3.4.2 TIS' Recommendation

· Piping to the headbox header should consist of a straight length which is two

or more pipe diameters long and is preceded by a long radius elbow, Venturi

elbow, or gradual transition.

· Slope pipes upwards (1 in or more per foot) from the fan pump to the header

to eliminate air pockets. This will also give better back drainage when the fan

pump is shut down.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 34 Report1

· When it is impossible to maintain a continuous upward slope to sweep out air

pockets, high points (such as the top of a pressure screen) should be vented

continuously. Successful venting, of course, requires that the vented area is

always above atmospheric pressure.

· High flow velocities (10 ft/sec and faster) will sweep air out of lines that do not

slope upward.

· Avoid air entrainment by cascades or vortices.

· Consider removal of troublesome air by a vacuum system.

· Adjustment of the recirculation flow from the end of a side entry tapered header

can be overly sensitive if this line and valve are over-sized. Each header is

designed for a specific recirculation, which is typically between 2 and 10% of

the headbox flow. The line and valve should be sized to pass approximately one

and a half times the design recirculation with the control valve fully open.

· A sight flo TM indicator on the header will assist in adjustment of the recirculation
flow.

· Walls should diverge from the mean flow direction at less than 5° to avoid

separation, and walls should converge at less than 45.o.

3.4.3 Voith's Recommendation

As a general rule, all horizontal piping should be up hill (in the direction of the flow).

· Thick stock (3- 4%) piping should be set at a minimum of 1.5 in rise per foot.

· Short runs of low consistency stock flow (0.25- 1%) may be set at 0.5 in rise

per foot.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 35 Report1

· Pipe velocities (other than controlled mixing points) should be no less than 6

fps to prevent settling of stock and slime generation.

· Provide for deaeration all high points in the piping lines. Deaeration lines should

be left slightly open at all times. Air vent lines should be separately piped to a
free surface discharge point.

3.5 Flow-Induced Vibration and Variations

The wire pit and screens, the screens and headbox, and parts of the headbox may

all form oscillating systems. In the case where the damping is not effective and the

resonance occurs, these oscillations will approach the headbox and eventually cause

basis weight variations by changing both head and consistency in the stuff box [36]
[37] [21] [14].

It is essential that with the piping system layout, all high-fiber content flows are

mixed in a uniform and smooth manner. The most efficient way to assure thorough
mixing is through the use of a collecting pipe. This method mixes individual return

flows and thick stock flow into a single pipe, prior to mixing with dilution water and

being introduced into the suction of the fan pump. Assuming mixing considerations

are properly followed, this method will eliminate the possibility of basis weight vari-

ations due to poor mixing, and it can also premix the return flows some distance

away from the fan pump. It is feasible to bring high fiber flows individually into the

suction line of the fan pump. This requires additional space to achieve proper mixing

conditions. Uncontrolled mixing by free discharge into the white water supply must
be avoided at all times.

Headbox recirculation should be returned to the fan pump suction through the

white water supply. This must be done in a manner that ensures uniform mixing,

while offering a separation between the recirculation piping and the remaining piping

in the approach flow system. Since pulsations introduced into the recirculation line

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 36 Report1

will have a direct effect upon the headbox, by disconnecting the recirculation from

the rest of the system, the influence of pulsations in the pump and the collecting pipe
upon the headbox can be greatly reduced.

To eliminate flow-induced vibration, the following general guidelines are commonly

used [4]-

· Simple hangers alone are not adequate for headers that are located under the
floor.

· Cross bracing must resist motion in the other two directions.

· Support or brace all piping, particularly at the headbox inlet.

· Support headers front and backl anchor to prevent CD vibration.

System sized for anticipated future higher volumes can be troublesome during

the start-up period. Temporary modifications to the fan pump system should be

considered, if a long period of sub-capacity operation is anticipated.

Immediately before the headbox (downstream of the screen) should be considered

a critical piping area and all design guidelines should be stringently followed.

For step diffuser headboxes on Fourdriniers which are equipped with a P.E. Tank

(Pulsation Elimination System), the tank serves as a hydraulic elbow, uniformly

distributing the flow (as delivered to the tank) across the full width of the inlet. This

provides a uniform distribution across the width of the machine. However, the tank

will not completely compensate for inefficiently distributed retention aid, etc., and

prior to the P.E. Tank, there should be a minimum of 3 pipe diameters of straight
run.

For pressure headboxes with highly filled sheets, the possibility of flow separation

(uneven filler distribution) in the elbows should be given special consideration. The

straight run of pipe should be 8 pipe diameters (minimum of 5). If the straight

run must be less than 6 diameters, then it is advisable to utilize a Venturi type

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 37 Report1

(aka acceleration) elbow. Further information is available on this design from Sulzer

Papertec Middletown Inc. (SPMI) as required [36].

Separate branch lines by 2 (branch) pipe diameters. Enter at 60 ° or, better, 45°.

Flow velocity should be low, less than 10 ft/sec. Match with main flow velocity.

The inlet closest to the pump should be the highest consistency. Avoid branch lines

entering diametrically opposite each other to prevent the flow of one affecting the

discharge of the second. Arrange branch inlets to avoid asymmetrical flows into the
fan pump suction.

3.6 Location of Valves

Valves control the pressure, rate of flow, and direction of fluids in accordance with

basic principles of flow [36] [20] [37] [21] [14].

No valve should be within eight pipe diameters of the header entrance. Variable

speed fan pump with good controls preferred. No in-line valve required. Coarse

control by a suitable control valve. Operate valves within optimum operating range

to avoid cavitation. Install valves before the screens, none between the screens and
headbox.

Use suitable control valve on the fan pump bypass, sized to give a maximum of

20% run-around. A second recirculation line with coarse control can be used to bring

the control valve into an optimum operating position.

Precede by 5 (preferably 8 or more) pipe diameters of straight pipe, and follow

by 5 (preferably 8) pipe diameters of straight pipe before tees or elbows. In locations

with space limitations, consider the use of vaned elbows.

Low elevations (submerged with respect to downstream pressure) for valves on

return lines are advisable to minimize the possibility of surging, resulting from the

trapped air after the valves. (This also minimizes negative pressures that can develop

from back flow at shutdown). All recirculation lines should go into pipeline leading

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 38 Report1

to the pump suction, not into the wire pit.

All control valves should be located to ensure there is a back pressure at all times.

This can be accomplished by locating all valves below the surface of the white water

supply. Valves may be located as to have downstream flow at a higher elevation than

the valve. When conditions dictate, a valve in a vertical flow downward line may be

placed above the surface of the white water supply, if there is a downstream pressure

drop greater than the physical elevation difference between the valve location and

the white water supply. The guiding factor is that the flow in the approach flow

system should be under positive pressure (except in specifically designed areas- i.e.,

deaeration tanks) at all times. Allowing flows in the approach flow system to generate

a natural vacuum will increase free air entrainment, foam generation, and pulsation

generation due to cavitation.

3.7 Elbow

3.7.1 Voith's Recommendation

Bends in flow should be made with long radius (Radius - 1.5 × Normal Diameter),
smooth flow elbows.

· Do not use mitred elbows due to the possibility of generating pulsations.

· Do not use vaned elbows due to the potential of fiber hang ups.

3.7.2 TIS' Recommendation

· Medium radius or vaned elbows are preferred. Medium radius elbows should be

followed by at least two pipe diameters of straight pipe.

· Two successive medium radius elbows should not lie in planes inclined at 90°

to each other.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 39 Report1

· Mitred vaned elbows are recommended in locations where uniform flow is desir-

able and a straight pipe run is impractical. In circular pipes, the vanes can be

welded into a band at the mitred section. The chord dimension can be selected

to provide a convenient number of vanes for any size of elbow, but generally

the vane spacing c/3 should lie between 2 in for small pipes and 5 in for large

pipes. The leading edges of the vanes should be wider than the longest fiber, or

3/8 in minimum, and the trailing edges rounded. End welds should be contin-

uous and polished. A modification of the above design includes transitions to a

rectangular mitred vaned section. This vaned section may be easier to fabricate

and, by including a curved wall _at the outside of the bend, gives better flow

control. The number of vanes of a vaned bend (no miter) design is based on the

feasibility of fabrication. All vanes must be carefully designed to avoid stapling

of long fibers. Interior welds must be ground smooth and pass the cotton ball
test.

· Discharge velocity from the ends of a tee should be greater than, or equal to,

the entrance velocity; the distance may be valved to restrict and stabilize the

discharge.

3.8 Discussions

Although many of the extracted guidelines from different companies and TIS are

similar, they do not lack confusion and contradictions. In conclusion, the additional

need in designing approach pipe systems is to have more precise and clearer guide-

lines, especially more key parameters based on deeper fundamental understanding.

It is found that most of the existing guidelines are based on the past engineering

experiences. Some of the design guidelines could be dated back for a few decades.

Very little information is available on the nonlinear transient behavior of the approach

ftow system. A proper way to achieve the optimized approach flow system design is

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
 Project F004 40 Report 1

really to have the computational tools to simulate individual components as well as

the whole system.
The common guidelines for the stuff box is to arrange the valve and stuff box over
the white water level. It is also conceivable that the stuff box can be treated as a
’ pulsation isolator between the refining system and the flow to the headbox. Although
certain hydraulic parameters are given in some guidelines, the detailed design is really
hard to master based on these simple rules extracted from the mass conservation law.
For example. how to optimize the vortex free stuff box free surface and the outlet is
never achieved.
As far as the wire pit and white white silo are concerned, the design is focusing on
the appropriate time for deaeration and stable water supply to the fan pump. How-
. .
I7 mixing . .
region 1before
F the* fan
f- pump IS
ever, the detail information Such as the really not
l 1 11

available. In fact, Valmet has a very different silo outlet geometry recommendation
from Beloit’s.
Commonly, variable centrifugal pumps are recommended by paper machine suppli-
ers, the needed controls of the pump are the stable inlet, impeller selections. However,
the information on how to avoid the deflection of the vanes, how the fan pump reacts
to certain variation during the papermaking process, and the detail design informa-
tion for low consistency stock is not available. In addition, for the most critical part
of pipe mixing, i.e. the mixing between the stock and white water before the fan
pump, there has been very little in-depth research done in the past.
Also, the guidelines for flow-induced vibration should really be considered much
more specifically than the current *ones.
In practice, due to the specific mill conditions and the change of certain com-
ponents within the approach flow systems, the existing guidelines are not always
followed. Therefore, it is very much needed to have a computer model to handle the
analysis of the whole approach flow system as well as the individual components.

XPST Confidential Information - Not for Public Disclosure

(For iPST Member Company’s Internal Use Only)
ProjectF004 Report1

Chapter 4

Research Approaches

The dynamics of the wet end includes pressure variation, flow, and consistency dy-

namics. The nonlinear dynamics computer modeling of the approach flow system

will help to analyze and to design the system with the consideration of perturbation

effects. The overall research approach would be two-fold:

· in-depth analysis of all components such as pressure attenuator, fan pump,

elbow, pipe mixing section, white water silo, stuff box, deaeration system, pres-

sure screen and cleaner within approach flow systems using finite element and

other computational procedures; and

· incorporate the computer simulation results into a global nonlinear dynamic

mathematical model, and obtain the solution with computational methods.

4.1 Engineering Approach

Starting from the Navier-Stokes equations, we can derive the following transient con-

trol equation between point 1 and 2

2 2

{_+V( 2 )+ +V(gz)}.dr- {v× (V×v)}-dr. (4.1)

1 P 1

IPST Confidential information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
°

Project
F004 42 Report
1

where v, p, p, dr, g and z stand for the flow velocity, pressure, fluid density, streamline
differential distance, gravity, and vertical Cartesian coordinate.

The r.h.s will be zero if either of the following holds:

· Positions 1 and 2 are on the same streamline; or

· V'×v-0.

Notice here, in applying the above relation to our stock, the density change around

the streamline might be considered. In the hydraulic control system, the pressure

transient behavior within the pipe can be estimated by the above equation. Of course,
the pipe flow direction is assumed to be the streamline direction in practice.

In engineering approach, pressure and flow dynamics have been modeled using
Bernoulli equations [39]

P_ -- P2 -_ A dt F Ap + s (4.2)

where Q, L, A, Ap and s denote the flow rate, path length, cross section area, pres-

sure source for the pipe containing the elevation difference between point 1 and 2,
and any heads provided by pumps. Since pumps receive energy from outside the

approach pipe system and are used to raise the energy level of the flow, the output of

the pump constitutes the total source of energy available within the hydraulic system.
Nevertheless, the total of the energy outputs from the system by components such

as the drive motor and hydraulic cylinders is not equal to the total energy available
from the pump because of friction and heat losses within the system. The proper
determination of the correct pump for a given application essentially involves match-

ing the pump to the required fluid power actuator giving consideration to pressure,

flow rates, available drive speeds and power, e_ciencies, cost, size, maintenance, and

the operational characteristics of the pump and system. These include pump noise,
vibration, natural frequency, and flow characteristics.

IPST ConfidentialInformation - Not for Public Disclosure

(ForIPST MemberCompany'sInternalUscOnly)
ProjectF004 43 Report
1

In addition to functioning in a stable, easily controlled manner, the fan pump

system should afford a minimum possibility of forming agglomeration, or bundles,

that cause sheet defects or breaks. Obviously, some detail analysis within the pump
is necessary in addition to the engineering approaches.

Eq. (4.2) was often found to be inadequate, especially in the modeling of pump
impeller vane deflection, screen vibration, and high frequency variations. Its modi-
fication is to include the continuous equation by introducing the fluid bulk modulus

For the consistency dynamics, the control equation can be often given as

dCo l_
dt - --(C_
T - Co) (4.3)

V
where C is the consistency, V is the tank volume, and _-- _ is time constant of the
tank.

4.2 Displacement/Pressure (u/p) Formulation

A number of finite element formulations have been proposed to model an acoustic

fluid for the analysis of fluid-structure interaction problems, namely, the displacement
formulation (se e Belytschko [8], Bathe and Hahn [6], Hamdi, et al [22], Belytschko and

Kennedy [9], Olson and Bathe [30]), the displacement potential and pressure formu-

lation (Morand and Ohayon [29]), and the velocity potential formulation (Everstine
[18], Olson and Bathe [31], Felippa and Ohayon [19]).

Primitive variable formulations have received considerable attention because they

do not require any special interface conditions or new solution strategies (for example,
in frequency calculations and response spectrum analysis). With the ever-increasing

availability of high-speed and large-capacity computers, this approach shows great

promise in general applications to the solution of a broad range of problems (specifi-

cally nonlinear problems) [7] [40] [41].

IPST ConfidentialInformation - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 44 Report1

In the mixed displacement/pressure formulation, we define a variational indicator

1-i -- f p2 -
{ _-_ u. fB __ )kp(_ p -}- V' u))dV _- f p SdS
u n
vf sf

where the variables are p, u, and the Lagrange multiplier Ap. We note that the first

two terms correspond to the usual strain energy (given in terms of the pressure)

and the potential of the externally applied body forces. The third term enforces the

constitutive relation. The last term is the potential due to any applied boundary

pressure on Sy.

Invoking the stationarity of II, we identify the Lagrange multiplier Ap to be the

pressure p and obtain the governing equations, with fs _ _pti,

Vp+pii - 0 (4.4)

V.u+ p = 0 (4.5)

with the boundary conditions

u. n - fin on $u (4.6)
p -- p on $f.

Applying the standard Galerkin discretization procedur e, we have for a typical

finite element

u - HU

p - HpP

V.u- (v-mu- SU

where H and Hp are the interpolation matrices, and U and P are the vectors of

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 45 Report1

solution variables [5].

The matrix equations of the u/p formulation are given as

i°o
01/
0 /i+ 0
where

M - fpH_HdV L = -fB_HpdV
vf vf

A -- _/1_ T
--THpHpdV R = -fH_S_pdS.
vs ss

In the u/p formulation we interpolate displacements and pressure as independent

variables, and we employ elements that satisfy the inf-sup condition. It is also im-

portant that the slip boundary conditions are introduced such that the requirements

of mass and momentum conservation around the fluid boundaries and fluid-structure

interfaces are satisfied.

Using the u/p formulation for the solution of frequencies, many (exact) zero fre-

quencies are calculated. The exact number of these zero frequencies can be predicted,

and in the solution the eigensolver simply shifts over them [5]. Also, the number of

zero frequencies can be reduced by use of the u-p-A formulation (see also reference

[7]).
This displacement/pressure finite element formulation for the analysis of acous-

tic/slosh fluid-structure interaction problems can be used effectively in this project.

4.3 Pressure Pulsation Attenuator

Extremely low frequency basis weight variations, from 0.001 Hz to 1 Hz, are often

related to thick stock flow. It is essential that the thick stock, at 3 to 4% consistency,

be fed at a uniform flow rate and stable consistency to the inlet of the fan pump.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 46 Report1

Acoustic fluid

Diaphragm

P+dP ._ P+dP'

Figure 4-1' Pressure pulsation attenuator mathematical model.

To finally achieve the uniform and consistent stock flow to the headbox, a properly

designed pressure pulsation attenuator has to be used to effectively eliminate both

the low- and high-range frequencies.

The above-discussed formulation can be directly applied to analyze the pressure

pulsation attenuator shown in Fig. 4-1. In fact, computation modeling of this sort

can also be applied to obtain optimized designs of different components.

4.4 Pipeline Simulation

Pipelines are nonlinear devices. The flow of fluids in pipes resembles, to a certain

degree, the flow of electric current in nonlinear resistances. Kirchoff's laws can be

applied directly (see Fig 4-2), however, the Ohm's law is replaced by the nonlinear

relation between flow rate Q and the friction loss AH

4 Q2
AH- S(Q) - £C2 (3.61og(C_Q/8dy)) 2 dS (4.8)

and

(4.9)
Q- S-_(AH)- _ AHds (-21og(2'825__ /XHd
C2L

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
Project
F004 47 Report
1

where C1 and C2 are the constants found from experimental results, and y, L and d
are the kinematic viscosity, pipe length, and pipe diameter.
The head loss AH can also be written as

e LU 2
AH- f (Re, d) d 2g (4.10)

where Re, e, and U are Reynolds number, surface roughness measure, and pipe flow
mean velocity, respectively.

From the Karman-Prandtl equation, for smooth pipes, we have

1
= 21Oglo(Re_ff) - 0.8. (4.11)

In actuality, the head loss will include the local head loss in turbulent flow, the

head loss due to sudden expansion, the loss in elbow (or pipe bends), the loss at

combining and dividing junctions. Of course, the concept of the equivalent pipe
length can also be extended.

In the study of transient flow, especially pressure surges in long pipelines, there

are close equivalents to the transmission line telegraph and telephone equations. In
the approach pipe flow system, fan pumps, pressure screens, cleaners, and elbows can
be simulated by different electrical devices.

Notice that to deal with the nonlinear relations in Eqs. (4.8) and (4.9), in practice
iterative methods have to be used.

For the complex pipeline systems, we apply the following

Pipelines in Series:

N N

AH_ - _ S_(Q) - AH (4.12)

i=1 i=1

Parallel Pipelines:

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 48 Report1

Qi Ii

Q2 _!_ Q4 I2 I4

Q3 13

Q1 + Q2 + Q3 + Q4 = 0 I1 + I2 + I3 + I4 - 0

1 2

DHz + DH5 + DH3 - 0 DV1 + DV5 + DV 3 - 0

DH2 + DH4- DH5 - 0 DV2 + DV4- DV 5 - 0

Figure 4-2' Analogy to electrical devices.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
Project
F004 49 Report
1

TypeI Type
II

\ x,

f x7 x,
\ _ V X
x,
\ x,
'x x,
\ x,
\ x,
\ x,
\ _/// X

,Nxx - ' xxxx

Figure 4-3' Silo design.

N N

Qi - _ SU_(AH) - Q (4.13)
i=1 i=1

and Kirchoff's laws to Partial Looped Pipelines, Branching Pipelines, and Pipeline

Networks. As discussed before, computational models of different approach flow sys-

tem components are going to be used to get the simplified equations such as Eqs. (4.8)
and (4.9). For example, in the analysis of the free vibration of fluid submerged im-

peller, the computational procedure with the u/p formulation as described above is
efficient and sufficiently accurate. A similar computational model can be applied

to the other components as well. Finally, we can incorporate the equations such

as Eqs. (4.8) and (4.9) into a global model of the approach flow system, and apply
the computational nonlinear dynamics procedure to analyze the effects of individual

components' variation on the system. A schematic hierarchy is shown in Fig. 4-4.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
Project
F004 50 Report
1

System Parameters Input

Main Menu

_3 _3 _3 _3

O O O O

0 0 0 0
0 0 0 0

System Design and Analysis

Modification

Figure 4-4' An overview of the computer simulation of the approach flow system
(eventually the whole wet end processes.)
IPST Confidential Information - Not for Public Disclosure
(For ]PST Member Company's Internal Use Only)
ProjectF004 51 Report1

_M:eI _:e yr
V

U U
Figure 4-5: Development of vortices.

4.5 Silo Design

Due to the different silo design guidelines [37] [14], in this research project, we will

develop a computer model to analyze some white water silo prototypes (shown in

Fig. 4-3) and discuss the advantages and disadvantages.

4.6 Vortex Forming

Air entrainment in the stock can result in pressure variations, machine direction basis

weight variations, slime and dirt spots, foaming, and pinholes. One of the interesting

problems is the vortices forming during the discharge outlets of chests, stuff boxes,

and wire pits (shown in Fig. 4-5). It is anticipated that proper CFD models can be

used to optimize the design process.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
Project
F004 52 Report
1

4.7 Application of Fluidics and Hydraulics

Fluidics is the technology of employing the general fluid phenomena of wall-attachment

and stream interaction in specially designed devices to perform the functions of sens-

ing, logic, and control [28] [16] [10] [27] [1] [13]. Some of the control procedures will
have the potential application to papermaking industries.

Major innovation in the utilization of fluids occurred in the 18th century as the
NeWcomen steam pump (1711) and Wart's steam engine (1769) were harnessed for the
rapidly increasing power requirements of the industrial revolution. In 1904 a German

aerodynamicist, Prandtl, proposed the control of flow separation in a wide-angled

diffuser illustrated in Fig. 4-6. Nicola Tesla filed a patent in 1916 for a "Valvular

conduit" which performed as a fluid diode, causing high resistance to fluid flow in

one direction and low resistance in the opposite direction shown in Fig. 4-7.

Figure 4-8 shows the effect upon the velocity profile caused by shifting the separa-
tion point. It can be seen that the velocity of the flow near the outer wall is increased
[33].
Figure 4-9 shows the turbulence amplifier. A laminar jet is produced from the

supply port. When the jet is disturbed and becomes turbulent before reaching the
outlet channel, the output signal falls from its previous value to a much smaller one.

Notice that the jet does not become laminar again immediately on removal of the

control signal, probably because of the feedback of turbulent flow from the region of
the outlet channel towards the control ports. The delay is found to be four times the

time for transport of a particle along the length of the jet.

4.8 Unsteady Flow in Piping System

Unsteady flow in piping systems is a common occurrence. Indeed, steady flow is so

rare that one might question the advisability of devoting so much time to a study of its

behavior. Although most of all hydraulic design is based on steady flow analysis and,

IPST ConfidentialInformation - Not for PublicDisclosure

(For IPST Member Company's Internal Use Only)
Project F004 53 Report 1

_:::::::::::::::::::
:i:!:!:i:i:!:!:!:i:!:!:i:!:!:i:i:i:i:i:!:!:!:!:!:!:!:!:!:!:!:!:i:i:i:i:!:!:!:!:!:i:i:i:!:!:!:!:i:!:!:!:!:i:!:!iii!iii!i!ii:ii
!!iiii!iiiii:i:i:i:i:!:!:i:!:i:i:i:i:i:i:i:i:i:i:!:!:i:i
!:!:!:!:!:!:!:!:!..._:!:i:i:!:i:!:!:!:i:i:i:i:i:i:i:!:!:!:i:i:::i::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::i:i:i:i:i:!:i:i:i:i:i:i:i:i:!:!:!:i:i:i:i:i:
.................-_......:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:::.::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

·. ======================

Suction _..........

_ !i_ _i_ _ _ _ ! _ i_._._ i_i_ i .!i_i_i !i!i i_i_ il !_i_i_i!_!:_!_i _i _i_i_i: i !i i i i _i_i_i_.i_i_ i_ _ _ _ _i_!i_i !_i_i_i_i_ _i_i! ! i_i! i _ i_i_:
Suction

!:!:!:i:!:i:!:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:i:iiiiiiiiiiiiii!i!iii!i!iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii!iiiiiiiii
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

Figure 4-6' Prandtl Diffuser.

Figure 4-7: Tesla's Diode.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
Project F004 54 Report 1

_//////////_
_ _ _

Figure 4-8' The velocity profile is changed by varying the control input.

II
., m

k. .r j

q[

C'
Jl j__ j J j_
P_ -_ -'b_ -b -'_
'_ Po

Po

Figure 4-9' Turbulence amplifier.

iPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
Project F004 55 Report 1

in many cases, the unsteadiness occurring in a pipeline system is of little consequence

because of its transient nature and small magnitude of change, in paper industry,

the unsteady flow within the approach flow piping is detrimental to the papermaking

process.

The control equation for the transient behavior within the pipe can be written as

[42]

cgp Oz 4,' dv
0-7+ + -7 - (4.14)
where d is the pipe diameter, r is the shear stress at the wall, and v is the average

velocity. Notice that none of the existing research in this area deals with nonlin-

ear dynamics involving consistency and pressure variations, although in practice the

perturbation due to various sources can be detrimental to the manufacturing pro-

cesses. It will be part of the approach flow system project to simulate the nonlinear
variations.

IPST Confidential Information- Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 Report1

Chapter 5

Research Planning

The following M.S. and Ph.D. projects could be generated from the research in this
area

(1) Computer Simulation of Approach Pipe System Dynamic Behaviors (M.S.)

(2) Mixing Behavior of White Water and Stocks Before the Fan Pump (M.S.)

(3) Mixing Behavior of Chemical Additives and Uniformity of Consistency (Ph.D.)

(4) Flow Separation and Cavitation (Ph.D.)

(5) Flow-induced Vibration of Approach Pipe Systems and Headboxes (Ph.D.)

(6) Nonlinear Dynamical Behavior of Approach Pipe Systems Subjected to Perturbation

(Ph.D.)

The fundamental research topics related to the above five projects are multidisci-

plinary. It includes the following hot subjects in computational mechanics areas'

(1) Turbulence Theory and Nonlinear Dynamics

(2) Fluid-Structure Interactions

(3) Both Spatial and Temporal Stability of Numerical Procedures

It is justified that a funded project should be established in the approach pipe

system area. This project will help to form an education and research center for study

in this area as well as the whole wet end papermaking processes.

Once the research is funded, we will start by analyzing the individual components

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 57 Report
1

1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6
Pulsation attenuator ......
Pipeflowmixture .....
Whitewatersilo _ _ _
Elbow ....
Writing
report _ _

Table 5-1' Tasks and time line.

within approach flow systems. The following problems will be focused on first'

· pressure pulsation attenuator;

· White water silo;

· Pipe mixture;

· Fan pump;

· Elbow; and

· Stuff box.

The second stage will be two-fold, firstly, we will continue to analyze specific

components within approach pipe flows; secondly, the whole system model will be

developed and analyzed.

Although at this stage, it would be hard to exactly follow the time lines, Table 5-1

is used to illustrate the tentative planning of this project for the first one and a half

years.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 Report1

Chapter 6

Conclusions

Regardless of dampening features of individual headbox designs, it is important to

minimize the system variations prior to delivery of stock to the headbox. This will

maximize the capability of the headbox to provide a uniform and consistent delivery

of fiber to the machine. The more stable the approach pipe system, the better the

end product. In fact, in modern paper machine designs, consistency variations are

of specific importance and it is nearly impossible to compensate for a swing in fiber

flow within the short dwell time of any headbox. This will only be accentuated as

the paper industry turns to more and more hydraulic headboxes and shorter forming
sections.

From the information provided by papermaking PAC members and other re-

sources, it seems that many possibilities exist for specific research focuses with the

proposed project. In fact, some of them can be ideal candidates for external projects

funded by individual companies or groups.

Based on the survey results and available guidelines from TAPPI TIS, Valmet,

Beloit, and Voith, it is concluded that more in-depth research is imperative in this

area. Since following the guideline does not mean trouble-free, and the details of the

approach flow system must be determined for individual cases, including the type

of plant, product, speed, and quality requirements for the specific application, the

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 59 Report1

development of computer simulators for the analysis as well as design purposes will

contribute tremendously to the paper industry.

In summary, the need for this project is strong and the support from the industry

is overwhelming. The main objective of this project is that through the development

of computer surrogates, we will finally be able to achieve, maintain, and master the

uniform rates and consistencies for all inflows, outflows, return flows, level controls,

in-line mixing, i.e., the complete process within the approach pipe flow system even if

the individual components and the whole system are under certain transient variation

or perturbation during the papermaking processes.

IPST Confidential Information - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
Project
F004 Report
1

References

[1] Pipework Design and Operation. IMechE Conference Publications, 1985.

[2] M. Barasch. Dynamic analysis of the approach flow system. CPPA 75th annual
meeting, technical section, pages A33-A37, 1989.

[3] M. Barasch. Dynamic analysis of the approach piping system. Tappi Engineering
Conference, pages 73-80, 1989.

[4] M. Barasch. Dynamic analysis of the approach flow system. Pulp _4 Paper
Canada, pages 49-52, 1991.

[5] K.J. Bathe. Finite Element Procedures. Prentice Hall, Englewood Cliffs, N.J.,
1996.

[6] K.J. Bathe and W.F. Hahn. On transient analysis of fluid-structure systems.
Computers _J Structures, 10:383-391, 1979.

[7] K.J. Bathe, C. Nitikitpaiboon, and X. Wang. A mixed displacement-based fi-

nite element formulation for acoustic fluid-structure interaction. Computers
Structures, 56(2/3)'225-237, 1995.

[8] T. Belytschko. Fluid-structure interaction. Computers _JStructures, 12'459-469,

1980.

[9] T.B. Belytschko and J.M. Kennedy. A fluid-structure finite element method
for the analysis of reactor safety problems. Nuclear Engineering and Design,
38:71-81, 1976.

[10] R.P. Benedict. Fundamentals of Pipe Flow. John Wiley & Sons, 1980.

[11] P. Boettcher. Headbox approach flow systems. Stock Preparation Short Course,
pages 41-45, 1984.

[12] D. Byers. Proper component fit is key design factor for headbox approach system.
Pulp _ Paper, pages 126-129, November 1991.

[13] A. Conway, editor. A Guide to Fluidics. Macdonald & Co. Ltd., 1971.

IPST ConfidentialInformation - Not for PublicDisclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 61 Report
1

[14] Beloit Corporation. Headbox piping recommendations: Design information.

November 1994.

[15] M.W. Day and J.C. Kerr. Stock approach technology' Systems design. Stock
Preparation Short Course, pages 91-95, 1988.

[16] P. Dransfield. Hydraulic Control Systems- Design and Analysis of Their Dy-
namics. Springer-Verlag, third edition, 1981. Lecture Notes in Control and
Information Sciences.

[17] J.C.W. Evans. Paper machines in the year 2000: Higher speeds, new press
designs. Trends and Developments in Papermaking, pages 50-51, 1985.

[18] G.C. Everstine. A symmetric potential formulation for fluid-structure interac-

tion. Journal of Sound and Vibration, 79(1)'157-160, 1981.

[19] C.A. Felippa and R. Ohayon. Mixed variational formulation of finite element
analysis of acoustoelastic/slosh fluid-structure interaction. Journal of Fluids and
Structures, 4:35-57, 1990.

[20] Fluid Mechanics Committee Tappi. Headbox piping system: General design
guide. Technical Information $heet(T.I.$.), 0410-06, 1995.

[21] Valmet-Tampella Ltd. Board Machinery Group. Special requirements for the
board machine short circulation. September 1985.

[22] M.A. Hamdi, Y. Ousset, and G. Verchery. A displacement method for the anal-
ysis of vibrations of coupled fluid-structure systems. International Journal for
Numerical Methods in Engineering, 13:139-150, 1978.

[23] O.L. Heissenberger, Jr. Design aspects for approach flow systems. Papermakers
Conference, pages 273-278, 1986.

[24] M. Johnson. The headbox approach system. Tappi Papermakers conference,

pages 515-522, 1991.

[25] J.C. Kerr. Stock prep approach systems. Stock Preparation Short Course, pages
93-98, 1990.

[26] K.D. Kurtz. Deaeration and the approach system. Stock Preparation Short
Course, pages 179-193, 1994.

[27] A. Lencastre. Handbook of Hydraulic Engineering. John Wiley & Sons, 1987.

[28] C.B. Lester. Hydraulics for Pipeliners, volume 1' Fundamentals. Gulf Publishing
Company, second edition, 1994.

IPST ConfidentialInformation - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)
ProjectF004 62 Report
1

[29] H. Morand and R. Ohayon. Substructure variational analysis of the vibrations

of coupled fluid-structure systems. Finite element results. International Journal
for Numerical Methods in Engineering, 14:741-755, 1979.

[30] L.G. Olson and K.J. Bathe. A study of displacement-based fluid finite elements
for calculating frequencies of fluid and fluid-structure systems. Nuclear Engi-
neering and Design, 76:137-151, 1983.

[31] L.G. Olson and K.J. Bathe. Analysis of fluid-structure interactions. A direct
symmetric coupled formulation based on the fluid velocity potential. Computers
_j Structures, 21(1/2)'21-32, 1985.

[32] D. Orloff. IPST Internal Communication, September 1995.

[33] R.S. Powell, M.M. Smith, and M.J.H. Sterling. Robust on-line data validation
for real-time monitoring of water distribution networks. In C. Maksimovid and
M. Radojkovi_, editors, Computational Modelling and Experimental Methods in
Hydraulics, pages 279-288. Elsevier Applied Science, 1989.

[34] M. Rajala. Viscosity perturbation damper connected with distribution header of

papermaking machine headbox. United States Patent, 3865687, February 1975.

[35] G.A. Smook. Handbook for pulp _ paper technologists. Angus Wilde Publications,
second edition, 1992.

[36] J. Stadelmann. Design of approach flow systems. Voith Inc.

[37] Valmet. Special requirements for the board machine short circulation. 1985.

[38] Vendert. Stock piping principles and their effect on paper machine performance.
1976.

[39] P. Virtanen, H. Koivo, and F. Cameron. Computer simulator for paper machine
wet end. Tampere University of Technology at Finland, 1987.

[40] X. Wang and K.J. Bathe. Displacement/pressure based finite element formula-
tions for acoustic fluid-structure interactions. International Journal for Numer-
ical Methods in Engineering, 40:2001-2017, 1997.

[41] X. Wang and K.J. Bathe. On mixed elements for acoustic fluid-structure in-
teractions. Mathematical Models 8j Methods in Applied Sciences, 7(3)'329-343,
1997.

[42] G.Z. Watters. Analysis and Control of Unsteady Flow in Pipelines. Butterworths,
1984.

IPST ConfidentialInformation - Not for Public Disclosure

(For IPST Member Company's Internal Use Only)