Sie sind auf Seite 1von 328

University of Wollongong Thesis Collections

University of Wollongong Thesis Collection


University of Wollongong

Year

Performance and capacity of centrifugal


gas cleaning devices
Mohamed S. Saad
University of Wollongong

Saad, Mohamed S, Performance and capacity of centrifugal gas cleaning devices, PhD thesis,
School of Mechanical, Materials and Mechatronic Engineering, University of Wollongong,
2006. http://ro.uow.edu.au/theses/661
This paper is posted at Research Online.
http://ro.uow.edu.au/theses/661

NOTE
This online version of the thesis may have different page formatting and pagination
from the paper copy held in the University of Wollongong Library.

UNIVERSITY OF WOLLONGONG
COPYRIGHT WARNING
You may print or download ONE copy of this document for the purpose of your own research or
study. The University does not authorise you to copy, communicate or otherwise make available
electronically to any other person any copyright material contained on this site. You are
reminded of the following:
Copyright owners are entitled to take legal action against persons who infringe their copyright. A
reproduction of material that is protected by copyright may be a copyright infringement. A court
may impose penalties and award damages in relation to offences and infringements relating to
copyright material. Higher penalties may apply, and higher damages may be awarded, for
offences and infringements involving the conversion of material into digital or electronic form.

Performance and Capacity of Centrifugal Gas


Cleaning Devices

A thesis submitted in fulfillment of the requirements


for the award of the degree of

Doctor of Philosophy
From

University of Wollongong
By

Mohamed S. SAAD
BSc, MEng.

School of Mechanical, Materials and Mechatronic Engineering.


Faculty of Engineering
2006

DECLARATION

I, Mohamed SAAD, declare that this thesis, submitted in fulfillment


of the requirements for the award of the degree of Doctor of Philosophy, in
the Faculty of Engineering at the University of Wollongong, is wholly my
own work unless otherwise referenced or acknowledged. The document has
not been submitted for qualifications at any other academic institution.

Wollongong, Australia

______________
Mohamed SAAD

ACKNOWLEDGMENT

I would like to thank my supervisor A/Prof. Peter WYPYCH for his


excellent supervision and invaluable guidance, support and encouragement
throughout the research project. His valuable suggestions, numerous
comments and criticisms during various stages of this research, including
the preparation of this work, are gratefully acknowledged.

I would also like to acknowledge with sincere appreciation the Ministry of


Higher Education of Libya and the University of Al-Tahadi for awarding
me a research scholarship through which the complete financial support for
this research was providing.

I would like to convey my appreciation to the technical staff in Bulk


Material Handling Laboratory for their friendship and unlimited
cooperation which made the research laboratory a pleasant place to work.
In particular, I would like to thank Mr. D. Cook, Mr. I. Frew, Mr. D. Hasti,
Mrs. W. Halford and Mr. I. McColm.

Finally and most importantly, this thesis is especially dedicated to my


family. To spirit of my parents, for their unfailing support and long
patience, I am extremely grateful. They always were the pillars upon which
I could lean to progress, God bless them. To my dearest wife, Fatima, for

ii

her support, understanding and sacrifice over these years and also to my
lovely sons, Suhaib and Safwan who were eagerly waiting for me every
night to come back home, although I couldnt spend as much time as I
wished with them, I am really grateful. I would also like to wholeheartedly
thank my best friend Ammar for his unlimited support and encouragement
over these years. Thank you all.

iii

ABSTRACT

The purpose of dust control systems is to capture, collect and dispose of


contaminant in an efficient manner. This research examines how to
improve the operational and collection efficiency of gas cleaning devices
via variations in geometry of different cyclone components. Unfortunately
many of the predictive models provide inaccurate and contradictory results.
Furthermore, many practical issues such as outlet and inlet configurations
have not been investigated properly or at all. This study investigates the
effect of cyclone outlet (vortex finder) diameter on cyclone pressure drop.
Two cyclone configurations were used: air discharging directly to
atmosphere; air discharging through a pipe connected to a filter. The
measured values of cyclone pressure drop were compared with pressure
drop predictions from various models (e.g. EEUA, 1997; Jacob et al., 1979;
Rhodes, 1998; Mason et al., 1983; and Zenz, 1999). This comparison
showed significant variations and differences compared with the
experimental results. The models of Jacob and Dhodapkar (1979) and
Mason et al. (1983) predicted similar values and were closest to the
experimental data. The research evaluated existing models and developed
new improved models for this purpose. A new theoretical model for
pressure drop prediction across the cyclone is presented based on the
consideration of the dissipative loss of flow in the cyclone system. Two

iv

different sizes of vortex finder (gas exit diameters) were used for this
modeling of pressure drop. The models of Stairmand (1949), Jacob
Dhodapkar (1979), Mason et al. (1983), Rhodes (1998), EEUA (1987) and
Zenz (1999) predicted significantly lower pressure drops than the
experimental values. The model of Barth (1956), with two values of k1 and
k2 for rounded and sharp edges, respectively, predicted significantly higher
values than the experimental data. Furthermore, the maximum solids flow
capacity of cyclone separators was investigated. Different bulk solids and
air flows were tested under different conditions: maximum solids flow rate
under pneumatic conveying conditions (before choking); choked gravity
flow from the test cyclone; and different gravity flow conditions from a
hopper. The results obtained in this study were compared with the
predictions of Beverloo et al. (1961), Brown (1961), Zenz (1962) and
Johanson (1965). Results show that the Johanson (1965) model provides
reasonable agreement with the experimental results.

TABLE OF CONTENTS

Acknowledgment ................................................................................ i
Abstract ............................................................................................. iii
Table of Contents ............................................................................... v
List of Figures .................................................................................. xv
List of Tables................................................................................. xxiii
List of Symbols ............................................................................ xxvii

Introduction.1
1.1 Background ........................................................................................ 1
1.2 The Purpose of Dust Control ............................................................. 2
1.3 Various Aspects of Dust Control ....................................................... 6
1.4 Gravitational Separation .................................................................... 9
1.4.1 Mechanical Description of Cyclone Separator...................... 10
1.4.2 Cyclone Applications............................................................. 15
1.4.2.1 Particulate collections ............................................... 15
1.4.2.2 Pre-cleaner ................................................................ 15
1.4.2.3 Fine particles............................................................. 15
1.4.2.4 Coarse particles......................................................... 16
1.5 Description of present research........................................................ 18
1.6 Objectives and thesis contents ......................................................... 19

vi

LITERATURE REVIEW........................................................................ 21
2.1 Introduction ...................................................................................... 21
2.2 Experimental Investigations............................................................. 22
2.2.1 Development of cyclone separator application in industry ... 23
2.2.2 Influence of geometrical configuration on cyclone
performance ........................................................................... 24
2.2.2.1 Cyclone geometry ..................................................... 24
2.2.2.2 Body diameter and dimensional ratio ....................... 27
2.2.2.3 Cone design and conical length ................................ 29
2.2.2.4 Dust outlet geometry................................................. 30
2.2.2.5 Gas outlet or vortex finder design ............................ 31
2.2.3 Operating variables on cyclone performance ........................ 33
2.2.3.1 Flow rate ................................................................... 33
2.2.3.2 Physical properties of the gas ................................... 34
2.2.3.3 Properties of the dust ................................................ 36
2.2.3.4 Concentration and dust loading ................................ 37
2.2.3.4.1 Pressure drop ............................................. 37
2.2.3.4.2 Collection efficiency ................................. 40
2.2.3.4.3 Dust rope phenomenon.............................. 41
2.3 Miscellaneous Cyclone Inlet Designed............................................ 42
2.3.1 Circular or pipe entry .......................................................... 43

vii

2.3.2 Tangential entry ..................................................................... 43


2.3.3 Swirl vane entry ..................................................................... 44
2.3.4 Wrap-round entry................................................................... 45
2.4 Classification of Cyclone Separators ............................................... 46
2.4.1 Conventional cyclones........................................................... 46
2.4.2 High efficiency cyclones ....................................................... 47
2.4.3 Wet or irrigated cyclone ........................................................ 47
2.4.4 Multi cyclone separators........................................................ 47

THE THEORY OF CYCLONE PRESSURE DROP AND


COLLECTION EFFICIENCY ............................................................... 52
3.1 Introduction ...................................................................................... 52
3.2 Performance Modeling..................................................................... 54
3.2.1 Flow pattern ........................................................................... 54
3.2.2 Cyclone velocities and pressure distribution ......................... 55
3.3 Pressure Drop ................................................................................... 59
3.3.1 Vortex finder sizes ................................................................. 62
3.3.1.1 Swirling flow in vortex finder .................................. 63
3.4 Models for Cyclone Pressure Drop.................................................. 65
3.4.1 Models based on estimating the dissipative loss ................... 65
3.4.1.1 Stairmand model ....................................................... 65

viii

3.4.1.2 Barth model............................................................... 66


3.4.1.3 Core model................................................................ 67
3.4.2 Purely empirical models ........................................................ 69
3.4.2.1 Shepherd and Lapple models.................................... 69
3.4.2.2 Casal and Martinez-Benet models ............................ 69
3.5 Other Selected Pressure Drop Models ............................................. 69
3.5.1 EEUA model.......................................................................... 69
3.5.2 Jacob and Dhodapkar model.................................................. 71
3.5.3 Rhodes model ........................................................................ 71
3.5.4 Mason et al. model................................................................. 72
3.5.5 Zenz model ............................................................................ 73
3.6 Solids Mass Flow Rate..................................................................... 74
3.6.1 Maximum solids flow capacity.............................................. 75

EXPERIMENTAL FACILITIES AND TECHNIQUES...................... 78


4.1 Description of Test Rig .................................................................... 78
4.1.1 Air supply and control ........................................................... 85
4.1.2 Feeding vessel and receiving bin........................................... 87
4.1.3 Rotary valve........................................................................... 88
4.1.4 Blow tank............................................................................... 89
4.1.5 Conveying lines ..................................................................... 91

ix

4.1.6 Cyclone separator .................................................................. 92


4.1.7 Material specification ............................................................ 93
4.2 Instrumentation and Data Acquisition ............................................. 93
4.2.1 Mass flow-rate of air.............................................................. 94
4.2.2 Mass flow-rate of solids......................................................... 94
4.2.3 Pressure drop.......................................................................... 94
4.2.4 Data acquisition system ......................................................... 95
4.2.5 Data processing...................................................................... 96
4.3 Calibration...................................................................................... 101
4.3.1 Load cells calibration........................................................... 101
4.4 Test Materials and Properties......................................................... 104
4.4.1 Test materials....................................................................... 104
4.4.2 Particle size and distribution................................................ 105
4.4.3 Loose poured bulk density................................................... 108
4.4.4 Particle density..................................................................... 110
4.4.5 Angle of repose.................................................................... 110
4.4.5.1 Poured angle of repose............................................ 111
4.4.5.2 Drained angle of repose .......................................... 113
4.5 Cyclone Experimental Procedure................................................... 116
4.5.1 Cyclone performance air only .......................................... 116
4.5.2 Cyclone performance with solids ..................................... 117

4.6 Phantom Vision Digital Analysis System...................................... 124


4.6.1 Measurement steps............................................................... 125

EXPERIMENTAL INVESTIGATIONS INTO CYCLONE


PRESSURE DROP AND SOLIDS FLOW CAPACITY.................... 129
5.1 Introduction .................................................................................... 129
5.1.1 Pressure drop........................................................................ 129
5.1.1.1 Flow straightener .................................................... 130
5.1.1.1.1 The test model ......................................... 130
5.1.1.1.2 Experimental results ................................ 131
5.1.1.1.2.1 Influence of flow
straightener ............................ 131
5.1.1.2 Vortex finder........................................................... 135
5.1.1.2.1 Experimental scheme .............................. 135
5.1.1.2.2 Air discharged directly to atmosphere .... 136
5.1.1.2.3 Test procedure ......................................... 138
5.1.1.2.4 Air discharged through a pipe to a filter . 138
5.1.1.2.5 Influence of the vortex finder.................. 139
5.2 Pressure Drop Modeling ................................................................ 144
5.2.1 Modeling with air discharging to atmosphere ..................... 144
5.2.2 Modeling with vortex finder connected to the filter............ 145
5.3 Development of New Pressure Drop Model.................................. 147

xi

5.4 Maximum Solids Flow Capacity.................................................... 158


5.4.1 Experimental Scheme .......................................................... 159
5.4.2 Capacity Limitation ............................................................. 159
5.5

Modeling of Gravity Flow Discharge ............................................ 162


5.5.1 Type of Flows and Velocity Variations............................... 168
5.5.2 Average Velocity of Bulk Material in a Mass-Flow
Hopper ................................................................................. 169
5.5.3 Experimental Scheme .......................................................... 175
5.5.4 Particle Velocity Analysis ................................................... 176
5.5.5 Test Results.......................................................................... 176

COMPARISON WITH THEORY OF EXPERIMENTAL DATA ON


PRESSURE DROP AND SOLIDS CAPACITY ................................. 178
6.1 Introduction .................................................................................... 178
6.1.1 Pressure Drop....................................................................... 178
6.1.1.1 Comparison of 105mm I.D. and 130mm I.D.
Vortex Finders ........................................................ 178
6.1.1.2 Comparison of Experimental Data and Existing
Models..................................................................... 180
6.1.1.3 Comparison of Experimental Data with Pressure
Drop Model Based on Dissipative Losses .............. 183

xii

6.2 Maximum Solids Capacity and Gravity Flow Discharge .............. 188

DISCUSSION.......................................................................................... 195
7.1 Introduction ................................................................................... 195
7.2 Cyclone Pressure Drop................................................................... 195
7.3 Pressure Drop Prediction ............................................................... 197
7.4 New Model Of Pressure Drop........................................................ 197
7.5 Maximum Solids Capacity and Gravity Flow Discharge .............. 198

CONCLUSION AND FUTURE WORK.............................................. 200


8.1 Conclusion...................................................................................... 200
8.1.1 Introduction.......................................................................... 200
8.1.2 Pressure Drop....................................................................... 201
8.1.3 Maximum Mass Capacity and Gravity Flow Discharge ..... 203
8.2 Future Work ................................................................................... 206

REFERENCES ....................................................................................... 208


APPENDIX A

................................................................................... 225

A.1 CYCLONE PRESSURE DROP EXPERIMENTAL WORK ....... 225


A.1.1 With Proper Inside Diameter of Vortex Finder 105mm I.D.
Flow Straightener Discharged Directly to Atmosphere

xiii

And the cyclone Separator Connected to Drum. ................. 225


A.1.2 With Proper Inside Diameter of Vortex Finder 130mm I.D.
Flow Straightener Discharged Directly to Atmosphere
And the cyclone Separator Connected to Drum. ................. 227
A.1.3 With Proper Inside Diameter of Vortex Finder 105mm I.D.
Flow Straightener Connected Via a Pipe to A Filter
And the cyclone Separator Connected to Drum. ................. 229
A.1.4 With Proper Inside Diameter of Vortex Finder 130mm I.D.
Flow Straightener Connected Via A Pipe to A Filter
And the cyclone Separator Connected to Drum. ................. 231
A.2 ANNUBAR CONVERSION GRAPH .......................................... 233
A.2.1 GNT-10 (1.5 inches 6.00 inch H2O)................................. 233
A.2.2 GNT-10 (1.5 inches 30.00 inch H2O)............................... 234
APPENDIX B

FRICTION LOSS CALCULATIONS

235

B.1 105mm I.D. vortex finder (Exit, Discharge to a filter from a 90


elbow, Round) ................................................................................ 235
B.2 130mm I.D. vortex finder and Transition. Round to Round (Exit,
Discharge to a filter from a 90 Elbow) ......................................... 235
B.3 Total pressure loss coefficient for transition: Transitions, Round to
Round ............................................................................................. 237

xiv

B.4 Local loss coefficient, Entries Duct mounted in wall (Hood, NonEnclosing, Flanged and Unflanged)............................................... 237
APPENDIX C

FLOW STRAIGHTENER PRESSURE DROP


EXPERIMENTAL WORK AND LOSS
COEFFICIENT ....................................................... 246

C.1 4 inch or (105 mm I.D. Flow Straightner) ..................................... 245


C.2 5 inch or (130 mm I.D. Flow Straightner) ..................................... 245
APPENDIX D

PRESSURE DROP PREDICTION


FOR FIVE MODELS ..............................................249

APPENDIX E

NEW PRESSURE DROP MODEL ....................... 255

APPENDIX F

PARTICLE VELOCITY EXPERIMENTAL


WORK ...................................................................... 263

F.1 Particle Velocity at the Cyclone Outlet Opening for Different


proportion of air flow to the blow tank feeder (before choking)... 263
F.2 Particle Velocity at the Cyclone Outlet Opening - Gravity Flow
Conditions (choked flow)............................................................ 263
APPENDIX G

................................................................................... 271

G.1 Mass flow rate prediction for Plastic pellets, Corn, and Rape seed
(canola) with different cone angles of cyclone separator , using four
theoretical models of Beverloo et al. (1961), Brown (1961), Zenz
(1962) and Johansone (1965)......................................................... 271
APPENDIX H

BULK MATERIAL PROPERTIES ...................... 274

H.1 Particle size Distribution................................................................ 274

xv

H.2 Instantaneous Yied Loci (IYL) and Wall Yiel Loci (WYL) Measured
for all Test Materials ...................................................................... 281
APPENDIX I

PUBLICATIONS DURING P.h.D


CANDIDATURE ..................................................... 291

xvi

LIST OF FIGURES

Figure
1.1

Caption

Page

Types of materials that can be separated


Hoffman et al. (2002).................................................................... 5

1.2

Particle sizes of some materials and suitable methods for


removing them from gas stream Hoffmann et al. (2002) ............. 6

1.3

Schematic diagram of a reverse flow cyclone separator ............ 11

1.4

General configuration of cyclone separation.............................. 12

1.5

Typical cyclone separator system ............................................... 13

1.6

Typical series arrangement for cyclone separators..................... 13

1.7

Plan of typical parallel arrangements of cyclone separator


Hoffmann et al. (2002)................................................................ 14

1.8

Elevation of typical parallel arrangements of cyclone separator


USACE (1988)............................................................................ 14

2.1

Typical conventional cyclone component and dimension


Labels .......................................................................................... 22

2.2

Cyclone entries: (a) circular or pipe inlet; (b) slot or tangential


inlet; (c) axial inlet with swirl vanes; (d) wrap-round inlet ..... 43

2.3

Types of cyclones in common use.............................................. 48

2.4

Typical arrangements for multi cyclone configurations............. 49

xvii

2.5

Relative effect of cyclone dimensions on efficiency.................. 50

3.1

Schematic diagram of a reverse flow cyclone separator ............ 53

3.2

Flow Pattern in the cyclone ter Linden (1949) .......................... 55

3.3

Velocity and pressure distribution in a cyclone


ter Linden (1949) ........................................................................ 58
(a) variation of tangential velocity vt , and radial velocity vr
(b) total and static pressure at different points in a cyclone
(c) variation of vertical velocity vh

3.4

Schematic of core model Hoffmann et al. (2002)....................... 68

3.5

Solids mass flow in relation to outlet sizing Zenz (1975) .......... 77

4.1

Layout of cyclone test rig air discharging directly to


Atmosphere ................................................................................. 80

4.2

Layout of cyclone test rig air discharging through pipe


connected to filter ....................................................................... 81

4.3

General layout of cyclone testing ............................................... 82

4.4

Annubar....................................................................................... 83

4.5

Cyclone DP Meters ..................................................................... 83

4.6

Cyclone water manometer .......................................................... 84

4.7

Flow straightener and vortex finder............................................ 84

4.8

The annubar station..................................................................... 85

xviii

4.9

General arrangement of compressed air supply.......................... 86

4.10

Rotary valve ................................................................................ 89

4.11

Blow tank Feeder ........................................................................ 90

4.12

General layout of blow tank feeding system .............................. 90

4.13

Samples of materials used and tested by the cyclone system in


this study ..................................................................................... 91

4.14

Air pressure tapping position...................................................... 95

4.15

Data acquisition system .............................................................. 96

4.16

Typical experimental plots created by Excel Spreadsheet


program ....................................................................................... 99

4.17

Calibration of feeding bin load cells......................................... 103

4.18

Calibration of receiving bin load cells...................................... 103

4.19

Regular and irregular shaped particles...................................... 105

4.20

Particle size distributions (Plastic Pellets)................................ 107

4.21

Particle size distributions (Corn) .............................................. 107

4.22

Particle size distributions (Canola) ........................................... 108

4.23

General configuration of poured angle of repose testing ......... 113

4.24

General configuration of drained angle of repose testing......... 115

4.25

Camera setup and recording window........................................ 124

4.26

Schematic diagram of cyclone apparatus rig with High Speed


Camera ...................................................................................... 128

xix

5.1

Schematic of flow straightener used to determine pressure


drop ........................................................................................... 132

5.2

Typical 105mm I.D. and 130mm I.D flow straightener ........... 132

5.3

Variation in pressure drop with F.S. and without F.S. ............. 134

5.4

Experimental and predicted pressure drop relations for different


flow straighteners...................................................................... 135

5.5

Conventional vortex finder without flow straightener.............. 136

5.6

Influence of 105mm I.D vortex finder on cyclone performance


(air discharging to atmosphere) ................................................ 140

5.7

Influence of 130mm I.D vortex finder on cyclone performance


(air discharging to atmosphere). ............................................... 141

5.8

Influence of 105mm I.D vortex finder on cyclone performance


(air discharging through a pipe connected to a filter)............... 142

5.9

Influence of 130mm I.D vortex finder on cyclone performance


(air discharging through a pipe connected to a filter)............... 143

5.10

Total pressure drop predictions by various models (105mm I.D


and 130mm I.D) with vortex finder discharging to
atmosphere. ............................................................................... 145

5.11

Total pressure drop prediction by various models (105mm I.D


and 130mm I.D) with vortex finder connected to a filter......... 146

xx

5.12

Pressure losses at different parts in a cyclone - directed to


atmosphere ................................................................................ 153

5.13

Pressure losses at different parts in a cyclone - connected to a


filter........................................................................................... 155

5.14

Total pressure drop due to dissipative losses in cyclone separator


(air discharged to atmosphere).................................................. 156

5.15

Total pressure drop due to dissipative losses in cyclone separator


(air discharged through a filter) ................................................ 157

5.16

General system of cyclone test rig............................................ 161

5.17

Cyclone separator with two observation windows ................... 161

5.18

Influence of cone angle on cyclone mass flow rate for plastic


pellets ........................................................................................ 165

5.19

Cyclone gravity flow discharge ................................................ 167

5.20

Different gravity flow conditions using a conical hopper


( =30) ..................................................................................... 167

5.21

Mass flow limits for Axi-Symmetric and Plane Flow Silo, Craig
(1996) and Wypych (2005)....................................................... 168

5.22

Velocity profile of bulk material in hopper, Craig (1996) ....... 174

5.23

Displacement of the material in hopper, Craig (1996) ............. 175

6.1

Comparison of the 105mm I.D. and 130mm I.D. vortex finders


(are discharged to atmosphere) ................................................. 179

xxi

6.2

Comparison of the 105mm I.D. and 130mm I.D. vortex finders


(connected to a filter) ................................................................ 179

6.3

Comparison of experimental data and existing models for a


vortex finder discharging to atmosphere .................................. 181

6.4

Comparison of experimental data and existing models of a vortex


finder connected to a filter ........................................................ 182

6.5

Comparison of experimental data and theoretical models of


vortex finder (discharging to atmosphere)................................ 186

6.6

Comparison of experimental data and theoretical models with


vortex finder connected to filter................................................ 187

A.1

Annubar conversion chart (6 inch H2O) ................................... 232

A.2

Annubar conversion chart (30 inch H2O) ................................. 233

B.1

Show the number of vortex spirals against the inlet


air velocity................................................................................. 238

B.2

Shows the function (K) verses area ratio (De/Dc) 2 .................. 238

B.3

Pressure loss of (EEUA model) via inlet velocity .................... 239

B.4

Pressure loss of (Jacob et al. model) via inlet velocity............. 240

B.5

Pressure loss of (Rhodes model) via inlet velocity................... 241

B.6

Pressure loss of (Mason et al. model) via inlet velocity........... 242

B.7

Pressure loss of (Zenz model) via inlet velocity....................... 243

B.8

Bend loss coefficient in 90 Round Elbows.............................. 244

xxii

C.1

Flow Straightener Loss Coefficient


(105mm I.D. and 130mm I.D.) ................................................. 248

H.1

Particle size distribution (old rape seed)................................... 276

H.2

Particle size distribution (old plastic pellets)............................ 276

H.3

Particle size distribution (old corn)........................................... 277

H.4

Particle size distribution (new plastic pellets) .......................... 279

H.5

Particle size distribution (new rape seed) ................................. 279

H.6

Particle size distribution (new corn) ......................................... 280

H.7

Effective Angle of Internal Friction (Plastic Pellets) ............... 281

H.8

Instantaneous Yield Loci (Plastic Pellets) ................................ 282

H.9

Wall Friction Angle (plastic pellets)......................................... 282

H.10

Wall Yield Loci (Plastic Pellets) .............................................. 283

H.11

Wall Friction Angle (plastic pellets)......................................... 283

H.12

Wall Yield Loci (Plastic Pellets) .............................................. 284

H.13

Effective Angle of Internal Friction (Corn).............................. 284

H.14

Instantaneous yield Loci (Corn) ............................................... 285

H.15

Wall Friction Angle (Corn)....................................................... 285

H.16

Wall Yield Loci (Corn)............................................................. 286

H.17

Wall Friction Angle (Corn)....................................................... 286

H.18

Wall Yield Loci (Corn)............................................................. 287

H.19

Effective Angle of Internal Friction (Rape Seed)..................... 287

xxiii

H.20

Instantaneous yield Loci (Rape Seed) ...................................... 288

H.21

Wall Friction Angle (Rape Seed).............................................. 288

H.22

Wall Yield Loci (Rape Seed).................................................... 289

H.23

Wall Friction Angle (Rape Seed).............................................. 289

H.24

Wall Yield Loci (Rape Seed).................................................... 290

xxiv

LIST OF TABLES

Table

Caption

Page

1.1

Typical size ranges of some common particle types....................... 4

2.1

Effect of cyclone parameter variation on collection


efficiency Davidson (2000) ........................................................... 26

2.2

Standard design for reverse-flow cyclones ................................... 27

2.3

Cyclone Classification................................................................... 49

3.1

Several equations for predicting pressure loss based on


number of inlet velocity heads Leith and Lee (1997) ................... 62

4.1

Cyclone dimensions and operating conditions.............................. 92

4.2

Mass of material and voltage measured by load cells................. 102

4.3

Physical properties of materials tested ........................................ 104

4.4

Maximum capacity of material to be separated........................... 118

4.5

Choking and escaping of the material from the cyclone............. 120

5.1

Maximum solids capacity of cyclone (just before choking) ....... 160

5.2

Bulk material properties .............................................................. 166

5.3

Maximum gravity discharge rate using cyclone (choked-flow) . 177

5.4

Maximum gravity discharge rate for different flow conditions


through hopper (Fig. 5.20)........................................................... 177

xxv

6.1

Maximum capacity of cyclone separator (just before choking).. 190

6.2

Maximum gravity discharge rate using cyclone (choked-flow) . 190

6.3

Experimental and predicted values of maximum gravity mass


flowrate of solids ......................................................................... 191

6.4

Physical properties of the old materials tested. ........................... 193

6.5

Physical properties of the new materials tested .......................... 193

6.6

Maximum solids capacity (just before choking) and maximum


gravity discharge rate (choked-flow) of cyclone ........................ 193

A.1.1 Data spreadsheet (105mm I.D.) to atmosphere ........................... 225


A.1.2 Data spreadsheet (130mm I.D.) to atmosphere ........................... 227
A.1.3 Data spreadsheet (105mm I.D.) connected to filter .................... 229
A.1.4 Data spreadsheet (130mm I.D.) connected to filter .................... 231
B.1

Data spreadsheet local loss coefficient..................................... 235

B.2

Data spreadsheet local loss coefficient..................................... 237

C.1

Data spreadsheet (105mm I.D.) F.S. pressure drop .................... 246

C.2

Data spreadsheet (130mm I.D.) F.S. pressure drop .................... 247

D.1

Pressure drop prediction of (EEUA Model)................................ 249

D.2

Pressure drop prediction of (Jacob et al. Model) ........................ 250

D.3

Pressure drop prediction of (Rhodes Model) .............................. 251

D.4

Pressure drop prediction of (Mason et al. Model)....................... 252

D.5

Pressure drop prediction of (Zenz Model) .................................. 253

xxvi

E.1

Data spreadsheet - new Pressure drop model of


(EEUA Model) ............................................................................ 255

E.2

Data spreadsheet - new Pressure drop model of


(Jacob et al. Model) .................................................................... 256

E.3

Data spreadsheet - new Pressure drop model of


(Rhodes Model) ........................................................................... 257

E.4

Data spreadsheet - new Pressure drop model of (Mason et al.


Model) ......................................................................................... 258

E.5

Data spreadsheet - new Pressure drop model of (Zenz Model) .. 259

E.6

Data spreadsheet - new Pressure drop model of


(Stairmand Model)....................................................................... 260

E.7

Data spreadsheet - new Pressure drop model of (Barth Model) . 261

F.1a

Data spreadsheet of Particles Velocity for different proportion of


air flow to the blow tank (0.25 to 0.04)....................................... 263

F.1b

Data spreadsheet of Particles Velocity for different proportion of


air flow to the blow tank (0.25 to 0.02)....................................... 264

F.1c

Data spreadsheet of Particles Velocity for different proportion of


air flow to the blow tank (0.2 to 0.02)......................................... 265

F.1d

Data spreadsheet of Particles Velocity for different proportion of


air flow to the blow tank (0.15 to 0.02)....................................... 266

xxvii

F.2a

Data spreadsheet of (Plastic pellets) velocity at the cyclone outlet


(choked flow)............................................................................ 267

F.2b

Data spreadsheet of (Corn) velocity at the cyclone outlet


(choked flow)............................................................................ 268

F.2c

Data spreadsheet of (Rape seed) velocity at the cyclone outlet


(choked flow)............................................................................ 269

G.1a

Data spreadsheet of maximum mass flow rate prediction (Plastic


pellets).......................................................................................... 270

G.1b

Data spreadsheet of maximum mass flow rate prediction


(Rape seed) .................................................................................. 271

G.1c

Data spreadsheet of maximum mass flow rate prediction


(Corn)........................................................................................... 272

H.1

Data spreadsheet of particle size distributions (old materials).... 274

H.2

Data spreadsheet of particle size distributions (new materials) .. 277

xxviii

LIST OF SYMBOLES

cross sectional area of the silo

[m2]

A0

outlet cross sectional area

[m2]

AR

total wall area of the cyclone body

[m2]

constant at section 2.2.3.4

[-]

condition at section 2.2.3

[-]

height of the cyclone inlet as defined in Fig. 2.1

average vertical acceleration of the material

ac

bulk material acceleration in hopper due to


convergence of the channel

av

[m]
[m/s2]

[m/s2]

bulk material acceleration in the hopper due to


increase in the velocity at the hopper outlet after
the discharge

[m/s2]

dust outlet diameter as defined in Fig. 2.1

[m]

barrel at section 3.5.5

width of the cyclone inlet as defined in Fig. 2.1

condition at section 2.2.3

[-]

constant at section 2.2.3.4

[-]

constant at section 3.5 as shown in Fig. 3.4

[-]

[-]
[m]

xxix

loss coefficient at section 3.5.4

[-]

constant at section 5.5

[-]

concentration of particles in inlet gas stream


at section 2.2.3

[g/m3]

vortex core at section 3.4

[-]

cyclone at section 3.5

[-]

CS

in the surface CS, as shown in Fig.3.4

[-]

diameter

the distance between the starting point to the ending

[m]

point for a single particle at section 4.6.1

[m]

Dd

diameter of cyclone dust outlet

[m]

De

diameter of gas exit (vortex finder)

[m]

DP

differential pressure drop between the entrance point


and the gas exit point

[Pa]
[m]

D0

hopper outlet diameter

dp

particle diameter

din

inlet diameter at section 3.5

dhi

inlet hydraulic diameter at section 3.5.5

d50

cut particle diameter (50% efficiency)

Eu

Euler number, P/(1/2 v2)

exit at section 3.5

[mm]
[m]
[inch]
[m]
[-]

xxx

ff

critical flow factor based on minimum opining dimension

[-]

ffa

actual flow factor based on actual opining dimension

[-]

wall friction factor

[-]

friction factor in Stairmand equation = f/2

[-]

gravity acceleration

total height of the cyclone as defined in Fig. 2.1

total height of the sample cone at section 4.4.5

[mm]

H1

height of the sample plate from the top of the stand

[mm]

H2

height from the tip of the sample cone to the top of


the stand

[m/s2]
[m]

[mm]

H() :

factor to take into account the variation in hopper type

[-]

height of the cyclone barrel

[m]

hc

height of the cyclone cone

[m]

hh

vertical height of hopper

[m]

constant in Barths pressure drop model at


section 3.4.1.2

[-]

flow straightener loss coefficient

[-]

proportionality constant at section 2.2.3

[-]

particle shape constant at section 5.5

[-]

length of the sample plate

[mm]

mass flow of solids

[kg/s]

xxxi

Mc

mass flow rate of solids collected

[kg/s]

Me

mass flow rate of solids entrainment

[kg/s]

Mf

mass flow rate of solids fed

[kg/s]

Mfa

mass flow rate of air

[kg/s]

Mi

mass flow rate of solids input

[kg/s]

M.S :

mild steel sheet

[-]

geometry parameter (m =1 for a conical hopper)

[-]

ms

mass of solids flow rate at section 3.6.1

NH

number of velocity heads

Ns

number of the spirals traverse by gas stream

[-]

vortex exponent, which equal 1 for an ideal gas

[-]

-1 for rotational as solid body

[-]

0.5 - 0.8 (in outer vortex)

[-]

Pi

inlet pressure

[atm]

volumetric flow rate

[m3/s]

term appearing in Stairmands pressure drop model

radius at section 3.4

[m]

Rc

vortex core radius

[m]

Rce

ratio of vortex core to vortex finder radii

Re

vortex finder radius

Re

Reynolds number, (vD)/

[lb/s]
[]

[-]

[-]
[m]
[-]

xxxii

rotational radius

[m]

reverse flow at section 3.5

height of cyclone vortex finder

[m]

SP

static pressure drop

[Pa]

S.S

stainless steel (304-2B)

gas temperature

Vav

average velocity of bulk material in hopper

[m/s]

Ve

gas velocity in vortex finder (exit duct)

[m/s]

Vm

average velocity of bulk material discharging from

[-]

[-]
[C]

the outlet

[m/s]

Vi

gas inlet velocity

[m/s]

Vt

tangential velocity

[m/s]

Vtcs

tangential velocity component in the surface CS

[m/s]

zh

depth below cylinder/hopper transition

hopper/cyclone cone angle, measured from the vertical

[]

solids drained angle of repose

[]

effective angle of internal friction

[]

total pressure drop

[Pa]

Pr

total pressure drop of gas flow reversal

[Pa]

Pe

total pressure drop of gas exit contraction

[Pa]

total interval time for the particle

[sec]

[m]

xxxiii

the difference distance between the starting point (x1, y1)


Coordinates and the finishing point (x2, y2) coordinate
for a single particle

[m]

loss factor, Table (3.1)

[-]

efficiency

[-]

oc

over collection efficiency

[-]

ratio of maximum tangential gas velocity, Table (3.1)

[-]

gas viscosity

[kg/ms]

gas density

[kg/m3]

solids bulk density

[kg/m3]

major consolidating stress

[Pa]

stress acting in equilibrium arch

[Pa]

ratio of maximum tangential gas velocity to velocity

within as entry, Table (3.1)

[-]

wall friction angle

[]

CHAPTER 1
INTRODUCTION

1.1

Background

It is well known that particulate pollution is a societal concern and has been
recognized as a problem for many decades. Airborne particles clearly cause
increase in incidence of disease of the air passages and constitute a real
environmental problem in some urban societies. Also, particles impact on
downstream equipment, such as blowers and turbines, causing erosion
damage. Therefore, many governments are imposing more stringent limits
on particulate emissions from the process industries. The challenge is to
meet these restrictions with robust and efficient technology at minimum
cost.

The necessity for increasing the protection of the environment and greater
raw material utilization increases the need for more highly efficient and
well-designed gas cleaning devices. It is well known that gas cleaning
device applications have been used widely in many industrial processes,
such as chemicals, metals, wood, paper, food and automotives. Gas

Chapter 1

Introduction

cleaning devices are used for a variety of process reasons. For example,
manufacturers may want to recover a valuable product from a gas stream.
Another common application is the separation of solids and gas at the end
of a pneumatic conveying line.

Gas cleaning devices are used either for the collection of dust or the
scrubbing of smokes or fumes. On the other hand, not only is protections of
the environment important, but also gas-cleaning devices are often installed
for industrial hygiene and safety reasons. Systems are designed to collect
contaminants such as welding fumes, grinding dusts, and dusts associated
with bag breaking and dumping operations Jacob and Dhodapkar (1979).
Also, they are used to collect fine particles and large particles, such as
metal shavings, sawdust, and even plastic pellets for the loading of
tankers/containers. Nevertheless, particles as fine as smoke and fume also
can be collected to some extent.

1.2

The Purpose of Dust Control

Harold (1973) reported that dust collection processes are concerned with
removal and disposal of suspended solids in gases for purpose of:
(i) Air pollution control as in fly ash removal from power plant
flue gases.

Chapter 1

(ii)

Introduction

Equipment maintenance reduction as in filtration of engineintake air.

(iii)

Safety or health-hazard elimination as in collection of


siliceous, metallic dust around grinding and drilling equipment
and in some metallurgical operations and flour dusts from
bagging operations.

(iv)

Product quality improvement as in air cleaning in the


production of pharmaceutical products and photographic film.

(v)

Recovery of a valuable product as in collection of dusts from


dryers and smelters.

(vi)

Powdered product collection as in pneumatic conveying; the


spray drying of milk, eggs and soap; and the manufacture of
high purity zinc oxide and carbon black.

Typical size ranges and types of some common particle types are given in
Table 1.1 based on the study of Jacob and Dhodapkar (1979). Particles in
gas stream vary in terms of size, density, shape, stickiness, friability,
erosive ness, surface, charge and other characteristics Hoffmann and Stein
(2002). Thus, separation equipment must be capable of processing a very
large variety of material from pellets to sub-micron powders, from hard
minerals, like garnet sand to soft food products like rolled oats. Fig. 1.1

Chapter 1

Introduction

shows a large number of substances that can and have been successfully
conveyed and subsequently separated in modern separation equipment,
including cyclones, bag filters and electro filters. Fig. 1.2 shows the
approximate size ranges of a number of particle types and the methods
suitable for removing them from gas stream.

Table 1.1 Typical size ranges of some common particle types

Typical Particles

Size range (m)

Carbon black

0.01 0.3

Tobacco smoke

0.01 1

Spray dried milk

0.1 10

Paint pigments

0.1 5

Ground talc

0.5 50

Fly ash

1 200

Pollens

10 100

Cement dust

3 100

Human hair

5 200

Hydraulic nozzle drops

50 5000

Beach sand

80 2000

Chapter 1

Introduction

Please see print copy for Figure 1.1

Figure 1.1 Types of materials that can be separated, Hoffman et al. (2002)

Chapter 1

Introduction

Please see print copy for Figure 1.2

Figure 1.2 Particle sizes of some materials and suitable methods for
removing them from gas stream, Hoffmann et al. (2002)

1.3

Various Aspects of Dust Control

The various aspects of dust control are:


(1)

Identification and definition of the problem, including the


characteristics of the dust, such as size, density, toxicity,
flammability, etc.

(2)

Containment and capture of the dust. Sometimes it is very difficult to


contain and/or collect the airborne particles, but there are three main
methods available: total enclosure, partial enclosure and hooding.

Chapter 1

(3)

Introduction

For the successful operation of any dust control plant, a proper


ducting system to transfer the dusty air gas needs to be designed.

(4)

Gas cooling or conditioning prior to dust collection: there are many


reasons dust-laden gases are often conditioned prior to collection.
For example, to recover heat and reduce the volumes of gas to be
handled or to protect filter fabrics or corrosion-protective linings
against high temperature. On the other hand, conditioning may assist
to separate the difficult particles, by increasing the humidity of a dry
gas stream.

(5)

Separation of the dust from the air or gas stream numerous


equipments is available for dust control, such as cyclones and inertial
separators, wet scrubbers, electrostatic precipitators and fabric filters.

The following physical factors can have a dominant effect on the design
and operation of gas cleaning systems Jacob and Dhodapkar (1979):

Size distribution of particles


Gas flow rate
Dust loading in the gas
Particle density
Particle shape and shape distribution
Mode of collection (wet or dry)

Chapter 1

Introduction

Powder flow properties


Attrition tendency
Erosion tendency
Electrical properties (Electrostatic precipitators)

There are many occasions during the processing and handling of particulate
solids, where particles must be removed from gas. It is well known that gas
may be used to transport some products such as powder within a process.
The efficient separation of the product from the gas at the end of the
transport line plays a very important part of this method of transportation.
In the combustion of solid fuels, fine particles of fuel ash become
suspended in the combustion gases and must be removed before the gases
discharge to the environment.

In any application, the size of the particles to be removed from the gas,
determine the separation method to be used. For example, particles larger
than 100 m can be separated easily by gravity settling, particles less than
10 m need more energy and other methods, such as filtration, wet
scrubbing and electrostatic precipitation Rhodes (1998).
Generally, there are four main methods for separating particles from gas
streams. The four methods are:

Chapter 1

Introduction

* Gravitational separation
* Inertial separation
* Filtration/Interception
* Electrostatic precipitation

1.4

Gravitational Separation

This is considered as the simplest and earliest type of dust collection


equipment, and consists of a chamber in which the gas velocity is reduced
to enable dust to settle out under the action of gravity.
Gravity separators are generally built in the form of long, empty,
horizontal, rectangular chambers with an inlet at one end and an outlet at
the side or top of the other end.

This study focuses on the device classified under gravitational separation,


known as cyclone separator. These devices are generally not suitable for
the separation of particles less than 10 m. They are best suited for
separating coarse particles larger than 20 m in diameter and as primary
separation devices with downstream electrostatic precipitator or fabric
filter.

Chapter 1

Introduction

10

1.4.1 Mechanical Description of Cyclone Separator


Cyclone separators are one of the most common types of industrial gas
cleaning device. They have the advantages of continuously separating
materials without stopping, and exhibiting relatively low pressure drops as
a function of flow rate. However, cyclone performance is still relatively
poor in terms of separating fine particles compared to filters. The most
common type of cyclones is known as the reverse flow type, as shown in
Fig. 1.3. The dust-laden gas enters tangentially into a cylindrical or conical
chamber and leaves through a central opening at the top part of the cyclone
body. Vortex motion or spiraling gas flow pattern creates a strong
centrifugal force field in which dust particles by virtue of their inertia,
separate from the carrier gas stream. They then migrate along the cyclone
walls by gas flow and gravity and fall into a storage receiver via the dust
outlet at the base of the conical lower part of the cyclone, see Fig. 1.3.
Centrifugal separating force or acceleration may range from five times
gravity in very large diameter, low resistance cyclones, to 2500 times
gravity in very small diameter, high resistance cyclones.

As shown in Fig.1.4, cyclone separators have no moving parts and consist


of: Gas entrance, Cylindrical body (barrel), Cone section causing the vortex
flow diameter to decrease until the gas reverses on itself, Gas outlet pipe or

Chapter 1

Introduction

11

vortex finder extending some distance into the cyclone body, and Dust
outlet. The immediate entrance to a cyclone is usually rectangular.

Please see print copy for Figure 1.3

Figure 1.3 Schematic diagram of a reverse flow cyclone separator

Chapter 1

Introduction

12

Please see print copy for Figure 1.4

Figure 1.4 General configuration of cyclone separation

In pneumatic conveying plant handling cyclone separators may also


combined with a fabric filter unit, if the bulk material is dusty see Fig. 1.5.

In some industries, where the solids concentration is high and the emissions
from just one separator stage would be too high, a second or even third
separator can be added in series or parallel with the first stage separator
collecting additional solids, Figs. 1.6, 1.7and 1.8.

Chapter 1

Introduction

Please see print copy for Figure 1.5

Figure 1.5 Typical cyclone separator system

Please see print copy for Figure 1.6

Figure 1.6 Typical series arrangement for cyclone separators

13

Chapter 1

Introduction

14

Please see print copy for Figure 1.7

Figure 1.7 Plan of typical parallel arrangements of cyclone separator,


Hoffmann et al. (2002)

Please see print copy for Figure 1.8

Figure 1.8 Elevation of typical parallel arrangements of cyclone separator,


USACE (1988)

Chapter 1

Introduction

15

1.4.2 Cyclone Applications

1.4.2.1

Particulate collections

Standard cyclones are used as particulate collection devices when the


particulate dust is coarse, and when collection efficiency is not a critical
requirement. However, their collection efficiencies are low compared to
other collection devices; cyclones are often used as pre-cleaners for other
equipment or as an additional cleaner to improve overall efficiency.

1.4.2.2

Pre-cleaner:

Cyclones are primarily used as pre-cleaners in solid fuel combustion


systems such as stoker fired coal burning boilers, where large coarse
particles may be generated. The most common application for such
cyclones is to be installed ahead of electrostatic precipitators. The reason is
that the cyclone exhibits increased collection efficiency during high gas
flow and dust loading conditions.

1.4.2.3

Fine particles

When fine sticky dust must be collected, cyclones more than 1.22 to 1.52 m
in diameter do not perform well. The use of small diameter multi cyclones

Chapter 1

Introduction

16

may provide better results, but may be subject to fouling. So, in this type,
usually two large diameter cyclones are employed in series.

1.4.2.4

Coarse particles

When using cyclones to separate coarse particles lager than 5 to 10 microns


in diameter with reasonable efficiency, inlet velocities must be takes into
account for cyclone design. This is to minimize erosion of cyclone walls
and to minimize degradation of coarser particles. Furthermore, cyclones are
also used for high-capacity solids flow applications: dense-phase
pneumatic conveying; pneumatic loading of tankers and shipping
containers.

Cyclone separation efficiency is dependent upon the mass of the particles.


The cyclone separator, which is classified under the inertial separation, has
long been established as a device for removing particles from gases.
Cyclone popularity is largely due to its simplicity of construction, lack of
moving parts, low cost and adaptability to a wide range of operating
conditions. Also the cyclone is compact in most applications, and the
collected product remains dry and normally, useful. The size of particles
that can be separated in a cyclone and the collection efficiency depend
essentially upon the difference in density of the solid particles and the

Chapter 1

Introduction

17

conveying gas, the solids concentration, the inlet gas velocity and the
dimensions of the cyclone body itself. Increasing the inlet velocity or
decreasing the cylinder diameter should normally result in an increase in
the collecting efficiency of finer particles. It should be noted that
decreasing the cylinder diameter will reduce the gas/solid throughput, in
addition to improving the collecting efficiency, and consequently more
cyclones will be needed. The dimensions of a cyclone designed for
optimum performance depend on its actual application (the nature of the
solid material to be separated and the separation efficiency required). For
high collecting efficiency the shape of the cyclone would be modified by
decreasing the cross-sectional area of the gas/solids inlet and the gas outlet
and reducing the depth to which the gas outlet duct extends into the
cyclone cylinder.

In recent years, more stringent environmental regulations require improved


dust collection efficiency in industrial installations creating a need for
further research aimed at improving the efficiency of industrial-scale
cyclones. Moreover, its desirable in some food processing industries, such
as the milk powder industry, to capture an extra fraction of the dust from
the effluent air stream in order to decrease the loss of valuable products in
secondary dust removal units such as filters Madhumita et al. (1998).

Chapter 1

Introduction

18

The most important parameters in cyclone operation are pressure drop and
collection efficiency. The overall collection efficiency is defined as the
ratio between the mass of solids collected by the cyclone in a time interval
and the mass flow rate of incoming solids. The pressure drop is given by
the difference between the static or total pressure at the cyclone entrance
and the gas exit. Both performance parameters are affected by the solids
loading. Although, there are many applications where solids loading are
high, cyclone performance is better known for loadings less than 1kg of
solid/kg of gas Fassani et al. (2000).

1.5

Description of Present Research

In order to obtain a reliable performance of a cyclone separator, there is a


need to understand the relationship between cyclone pressure drop and air
mass flow rate and the ability of the cyclone to separate solids from the gas
stream and the parameters that affect this relationship. The first study on
cyclone separators was recorded on 1886 and since that time, cyclone
separator development has continued. To date, due to its importance in the
industry, the cyclone separator has been investigated experimentally and
theoretically to improve their efficiency by introducing new design and
operation variables Stairmand (1952), Dirgo and Leith (1985), Kim and
Lee (1990), Hoffmann et al. (1992), Plomp et al. (1996), Buttner (1999),

Chapter 1

Introduction

19

Madhumita et al. (2000), Fassani et al. (2000), Obermair and Standinger


(2001), Avci and Karagoz (2003). As a result, a number of modifications to
the conventional cyclone design have been established and these have
improved collection efficiency Smith et al. (1982), DeOtte (1990), Liden
and Gudmundsson (1997), Youngmin et al. (2000) and Kim et al. (2001).
Despite efforts to develop theoretical expressions for the prediction of
pressure drop and collection efficiency based on the dimensions of the
cyclone and on the properties of gas and solid material to be separated,
none has really proved to be satisfactory and reliance must be placed on
experimental data for cyclone performance. Also, as far as the author
knows, no one has researched the solids flow capacity limitations of
cyclones. Many of the predictive models provide inaccurate and
contradictory results. Furthermore, outlet and inlet configurations have not
been investigated properly or at all.

Hence, it is imperative to investigate the factors affecting operational


performance and collection efficiency of the separation system as well as
maximum solids throughput.

1.6

Objectives and Thesis Contents

The main aim of this thesis is to investigate different cyclone geometries


and configuration for the purpose of improving operating performance and

Chapter 1

Introduction

20

model prediction. To achieve this overall aim, particular objectives of the


work are listed below:
Reviewing published literature to assess the current state of cyclone
separators.
Carry out extensive air-only experimental investigations into cyclone
pressure drop (using different vortex finder diameter).
Use the new database to explore the accuracy of cyclone pressure
drop models (especially in relation to the effect of the vortex finder).
Compare the experimental results with pressure drop prediction from
various models.
Investigate an over all pressure drop model based on several existing
cyclone theories and experimental findings.
Determine maximum solids flow capacity of cyclone (for various
industrial applications).
Compare the experimental findings of maximum capacity and
gravity discharge flow of cyclone with maximum solids predicted
from various models (Beverloo et al. (1961), Brown (1961), Zenz
(1962) and Johanson (1965).





 







    
      
                
       !     "   
 !  # " 
    "  $

 % "   
"    " &!'
!  "  "          !' !  
"     (   '    ! 
 "       
  !     )* *+  
    !      !       
     !    



 

   

 

      ! ,  -


./0/1  ./2/1, ./31,"./4/1 ./501
6./5718 -./731 # -./7/1( 
 9.//31 .::1 ! ;#.::01


Please see print copy for Figure 2.1



$ !          






9 '    ! !
   !      

 

   

0 

                  


!       


  !"#"

       !         !
     " !
  " " = "  
.::1 >!           ! !    
      !    !    ?   
      !      !    
!   #     " 
'   !  #
3:3:::: 0 @   
   ,  ./31  A ./5/1     
      "   -
 .//51 <    .//71 A    .:::1      .::1
8  !           .::1  
!    !    
-   .//51   B   .//71      
        
"  !      


 

   

2 

   C # ./531 B   .//71       .::1 $
' " !!  
          
       !    $   !  
        7::   !   
8 -./731     
 % ," !
-.//51  D
   "       *   
 !      %    

   $! !"% 

' !"%    !    
      ,    - ./0/ /2:1 !
   )+       .    "
   1 "
!    

  <

9   !             
                  



   

 

3 

     "


          "      
!    " !
  )    +!D" 

,              !'   E
    !               
  !'. 1  
  B .:::1&

E 

 

    
 


.1


"   
   

E&

    








 

   

4 

   !    


<.:::1



Please see print copy for Table 2.1


   

 

5 

 ,    !D" 











Please see print copy for Table 2.2



 ( 8 8  


           "
             !    
   !
"     "     
    %"   
       F       ! 
 ./551      
      
      

 

   

7 

./071     !          
           ! 
 B .:::1"     
   

"! (G  .///1   ! "      
          >      
       
  !'     "  
        !      ' 
./2/1   "     
        !  
   !'  !'
 ./551

"!'  <.//:1  
 .//31 "   "          !'
       ' "          "
C -.///1 !.::01  !   
         
    !.*,1
 !!      $

 

   

/ 

$ '            @8 H


43  43 "    !        "
       @8 H 33     .::1 >  
               
  @8H33   " 
   "        "  
"     "   
"    , .::1

0  8   - 

9  !    !     
        "! !     
!      , ./331
    ./551 "             
  !'! " " 
"     " "!  
     !     !    
      !
.::1    '   
         
     !'"


 

   

0: 

 I .::1   


 #       !        
   "  
 !   .::01            
        !      %   
       "!      !
         "     


B   ( ./701 
 '    "              "
           "! ,#   
 ./701       .::1      
         
"          "  
  

2 8

         !    
         
    ? "%   


 

   

0 

       !       
>    ,   .::1 "!      
            !  
      ! !
2:

8       
   D     
        $
      ./331   >    .::1  !   
           
>    ,   .::1   "    
     !            


3 <>J'$ 8 

J'   #              
              "  
      ,#     ./701      
              
"         


 

   

0 

!'   # #   - ./7/1    !'  
           "
    !'     "% "     ;
 -.//:1"  ' # 
   #      "
  2 "7772@ 
            " " 
  ( ./341 J   #     '    
     :2  :7 "       
    !      :3 9   9$ 
.//01      "    !'     
    !     !'    "   
           >     
C   - .///1        !'      
      !           
   "         !'    
       .::1
  ,  .::31  '      !'      
                 !
"%           !'   (   !'

 

   

00 

              
   
 "% "   "    '  -   .::0
::0 1"' !'   
 "   -.::0 1   
        
"   !      !   !'     
     " !'  
    %./7/1 
<     .//71          "
! "  ! !'  

E!    '  !

     

 !           $
'   -   .::01         !'
      "  
  "   

- . /!"% 

0 $"





   

 

02 

J    "    %        


  "  " !
        "      "  
     ./5519 '  
"   "              
      ,./71-
./731C -.///1 I .::1

C   - .///1  !   "      
       "         
    "  

  I    .::1     "    
    "     "      
  "    
 

0 <

       
" &
H%K  
 @

.1

"



   

 

* 

03 

.1

%*   


K*"

.0@1

* 

.1

D   

.%@01

* 

.L%1


               !
 !     "        
    "  "       
./551A  8 %./5/1 9.//31
  "   !  "
   &

:3

::    
   
::
 

"





H

.01

!    

'    !


       -
./731               
  



   

 

04 

  "      ./551 


  &

: 3

::  

 



 
::
 
 

.21


"

  H

   .%@01

  H

     .%@01


00 8

9  !    
               !
   !     #      $
' ,    ./331          
  
                "
 " !      ./551 8
.///1% B  .::1  
!            
        #    :   


 

   

05 

!"  9!   9 -


./731               
   

02 !

02 8

      " 
             
   "?   
 9    "  
    

 !  "  
"      
     '           % 
     '       !  
                   
 ! "       
    " 
         !       ,    -=

 

   

07 

./0/1 :0    !  


  5M 4M"    "! 07
N: 0 35N: 0 % @% 05N: 0 34N: 0 
% @% " !  6./571
"                 
   0:M! "'" 
  :40N: 0 % @%    
 2N: 0 2N: 0 % @%
"            
 "    "    
"   " !  %./71
  "     (%    9  ./751 "  ! 
                  
 "    " !       "   >  
 ; ./7/1 ".   
D1    !     
"  %        D J  ! 
"      "
    " "      "   !   
 ""      
!              "  


   

 

0/ 

          6   ./571       
   !    "          9  
9.//71"  "  " ! 
      ! "     "     
 $  <  .:::1  !
57 5@   :32  
  !     "   
  !      8
  .///1      !2   43 @  
    :4 2    %   @       
       
  "          "      
      ! ( ./241     ?  
                 
   &


  


   

.31


"                    
.@01O  "

 

   

2: 

.E@1 6   ./571   % ./71 !       
?  P!  !"  
.//1

02    

       
              
       #   
    -   ./2/1          
        B           
  "           
 "   ! 
!         !         
   "      - 
.//51  ,    ./331        
  "               A   
8 % ./5/1 "          
                  "
    9   - ./731 "      
!       "         
    " 


 

   

2 

  .//71            ! 3 @ 0    
           



 $C #.///1 !  


   !     

020 8
   

9 ! !   !   
          "        
        
    !       "   "   
 " !  !"  
 , ./41 > !      !     
   , ./41 F ./51 6   ./571  
 #
./731

   "!   F./51
"   " ?        
     '  "!
    " "       " #  
       

 

   

2 

6./571 !      "  


        .3:@01   
3:@0 "      ! 
" 

 #./731     "0::


8         D    !     
D        %         
"   3  "            "
 :" ""    
 "      

-

0!"


8          P       
    "              
                    
     " 
             " !   
!   "  $  !  
     !        #     "  
  




   

 

20 

0 Q=  



                   
  9  "              " 
 $

0    

9   $  
   $ "      
       !  '
    "          C # ./53
///1


Please see print copy for Figure 2.2




$  &.1Q= P. 1Q=


   P.1' ""! P. 1Q"D = 


 

   

22 

,  !               
           ./071       
               
9!   '      
   %     
                  
!    , .::1 !
    

00 ,"J  

    " Q"! ="  
$! "    
 '      " !'    
   "                 
       '       !   
               .//1 
"% %            !       
   "      ."   "   ! 1
 "    "
      !            
  B     !         " 



   

 

23 

     "   !  


        !   
./551

02 BD
  

"D Q!= "  $  
    "          !   "
  %              
      !        !'
        "  "  !'     
        ' "    "   !'
 %  "    D   
 "  , .::1

"! !"   
   
,  !

      !    



!            '     


    ,   !    !     
     8> .//:1 9   9$  .//41
<   ,  .//51   -   - .::01 8> .//:1

 

   

24 

!     ! "    !  


           !      
          "   <  
, .//51      D  
"  #       
!8D>!  F"  "
"  - -.::01"     
    "   
  8>.//:1 < , .//51

1

!  !"#


                  
        
   $0!  
       , ./331

2  !   

      "           
             
    $          



 

   

25 

2    



      
       !      !     
      !         
   "    
     "   $4
 5

20 B   

   "    !
    !  "          
               ? 
  "  D  

22 9 

B  !   !          
       
    
     F   "       
     $2


 

   

27 

    #   !       '


 ! 



Please see print copy for Figure 2.3




$0   

 

   

2/ 

Please see print copy for Figure 2.4


$2     


 0  


 !  
 
  
9D 
  
."1
  
 

<"   !  


.1
.@1
 
. 0 @1
.M1
:25D
3D2/7
4D02
3:D7:
/20
- :/ ::25D
2/7D2/3 32D02 7:D/3
:/20
:3D:0:3
23  5250D2/3 32D02 /:D/3

- :/ ::25D
2/7D2/3
32
/:D/3
:/20
( 

.1
D05






   

 

3: 

::

  M

/:
7:

,  

5:

9   

4:

-  

3:
2:
0:
:
:
:
:

:

:

0:

2:

3:

4:

5:

7:

/:

::

# 


$3
!      

9 !    !' 
                
       
    E       !      
    '    

!  !             
              
    R        !   ! 
        $   ' 
 !    "   
     !



   

 

3 

      


         "       
"    !    ! 

8             
! !!         
'.  1 
   !     !'      
! D   $      &

.1

!   !         !


'  

. 1

    ' ! D  '   ! 


      .    !'  
1

.1

  "    '     
  .  
!' 1

CHAPTER 3
THE THEORY OF CYCLONE
PRESSURE DROP AND COLLECTION EFFICIENCY

3.1

Introduction

Understanding the relationship between pressure drop, particle velocity,


volumetric concentration and solid flow rate in a cyclone separator is
important to design and operate a reliable cyclone system. For example, to
determine the pressure drop and mass flow rate discharge for a cyclone at a
certain gas velocity, or to evaluate the saturation capacity of cyclones
operating at a given air mass flow and delivery pressure, the relationship
between all these parameters is important. Therefore, this topic has been
investigated both experimentally and theoretically and have continued to
the present since the cyclone has been first patented in 1884s to develop a
reliable theoretical model to design a cyclone separator unit Dirgo and
Leith (1985). However, due to the complex three-dimensional fluid flow in
cyclones the exact mechanisms of removing particles are still not fully
understood.

52

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

53

To describe the flow in the reverse flow cyclone, see Fig.3.1, the dust laden
air stream enters tangentially at the top of the barrel section of the cyclone
and travels downward into the cone section of the cyclone forming an outer
vortex. The increasing inlet velocity causes an increase in centrifugal force
on the particles. This helps to separate the particles from the air stream.
When the air reaches the bottom of the cone, an inner vortex is created
flowing back up the center of the cone. The air leaves the cyclone through
the center cylinder at the top of the cyclone while the particles fall into the
dust collection bin connected to the bottom of the cyclone.

Please see print copy for Figure 3.1

Figure 3.1 Schematic diagram of a reverse flow cyclone separator

Chapter 3

3.2

The Theory of Cyclone Pressure Drop and Collection Efficiency

54

Performance Modeling

3.2.1 Flow Pattern

The pattern flow in a cyclone is illustrated in Fig 3.2, the air flow entering
the cyclone and tangentially pressed to the wall by acting of centrifugal
force and flow down as an outer spiral flow path close to the wall. From
this outer spiral, the spinning gas flows inwards towards the center and
moving upwards into the outlet duct. The zone diameter between the outer
and inner spiral is approximately equal to the outlet duct.

Cyclone flow pattern has been extensively investigated by Sheperd and


Lapple (1939), ter Linden (1949), Walton (1974), Kelsall (1952), Iozia
(1989), Griffiths et al. (1996), Leith and Lee (1997), and Gil et al. (2002)
and the overall trend of gas of motion has been generally confirmed.
However, there is no flow pattern model yet available, which will provide
accurate prediction of the velocity components at any point in a cyclone
separator. All of the experimental works have been concentrated on several
cyclones but with similar design. Although, no complete theory has yet
been given or is likely to be forthcoming.

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

55

Please see print copy for Figure 3.2

Figure 3.2 Flow Pattern in the cyclone, Ter Linden (1949)

3.2.2 Cyclone Velocities and Pressure Distribution

The total pressure at any point in a cyclone is the sum of the static pressure
and velocity pressure at that point. Total pressure decreases slowly from a
maximum value at the cyclone wall to a minimum value near the cyclone
axis. The high velocity produced in the cyclone body causes a high velocity
pressure and the static pressure becomes negative Leith and Lee (1984).

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

56

The velocity pressure in the cyclone can be so high with the high tangential
velocity presented, so the static pressure becomes negative relative to the
atmosphere. The static pressure within the central core can be negative and
the zone of negative static pressure may extend from core through the dust
outlet into the dust collection bin. Therefore, cyclone dust hopper should
always be airtight to prevent dust air drawn up from the bin into central
core and then out from the cyclone. It is believed that the pressure drop in a
cyclone depends on the cyclone design and on its operation parameters
such as inlet velocity.

The air flow in the cyclone rotational field is rather complex and can
complicate the understanding and modeling of cyclone pressure drop. It
can be described by three velocity components such as tangential velocity,
axial velocity and radial velocity. Ter Linden (1949) is considered as the
first paper who measured the flow field in a cyclone and described the
tangential velocity distribution and the pressure drop distribution, as shown
in Figs. 3.3a, b and c. After the air stream enters into cyclone, it will be
spiral downward because of the effect of centrifugal, gravitational and
friction forces. In the outer vortex, the total air flow consists of a tangential
velocity, radial velocity and axial velocity. The tangential velocity, vt, is

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

57

the dominant velocity and related to the distance, r, from the cyclone axis
by Eq. (1):

v t r n = Cons.
where

(1)

vt

tangential velocity

(m/s)

rotational radius

(m)

vortex exponent, which equals 1 for an ideal gas.

-1 for rotational as solid body

0.5 0.8 (in outer vortex)

Actual values of n are between 0.5 and 1, depending upon the radius of the
cyclone body and gas temperature Hoffman and Stein (2002).The
tangential velocity in the conical part of the cyclone increases with the
decreasing of the rotational radius in the outer vortex, it increases to the
maximum at the boundary of the outer and inner vortex. Tangential
velocities may be lower than the gas inlet velocity at the cyclone wall, but
can increase rapidly at some distance from the wall. Tangential velocity
considers a greater in the conical part than in the cylindrical part.

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

58

The tangential gas velocity within the core decreases with decreasing
rotational radius, and drops to a value near zero at the cyclone axis. The
pressure distribution varies from cyclone design to another. The rotation of
the air flow in a cyclone is considered as the main reason to cause pressure
drop.

Please see print copy for Figure 3.3

Figure 3.3 Velocity and pressure distribution in a cyclone ter Linden (1949)
(a) variation of tangential velocity vt , and radial velocity vr
(b) total and static pressure at different points in a cyclone
(c) variation of vertical velocity vh

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

59

Total pressure drop decreases from a maximum value at the cyclone wall to
a minimum value near the cyclone axis. With the higher tangential velocity
present can lead to higher pressure gradient.

The vertical and radial cyclone velocities are observed to be less than the
tangential velocity. The gas flow motion in these directions is considered as
a downward motion at the cyclone wall and upward motion toward the
cyclone axis. The radial velocity measurements are reported to be
approximately equal along the entire height of the cyclone. Although the
measurements are subject to inaccuracy due to the turbulence, the
investigation of radial velocity is nonexistent, compared with tangential
and vertical velocities ter Linden (1949) and Leith and Lee (1984).

3.3

Pressure Drop

Cyclone pressure drop is considered as one of the most important


parameters from the point of view to evaluate cyclone performance. Also,
cyclone pressure drop together with the operation of the cyclone is a major
factor to be considered in the design of any cyclone separation system.
Many models have been developed to estimate pressure drop, among them
Shepherd and Lapple (1939), Alexander (1949), Stairmand (1949, 1952),
Barth (1956), Casal and Martines (1983), and Avci and Karakoz (2001).

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

60

However, these models are either empirical or theoretical; none of them


predicts the pressure drop accurately for a wide range of cyclone designs.
It is known that the cyclone pressure drop is a function of the cyclone
dimensions, particle properties and its operating conditions such as inlet
velocity. Shepherd and Lapple (1939, 1940) and Hoffmann and stein
(2002) determined the optimum dimensions of cyclones based on body
diameter.

The pressure drop across a cyclone is caused by the geometry


configuration, the wall friction, change in flow direction, and the
dissipation loss in the vortex finder (outlet tube). Because of these various
factors, the cyclone pressure drop is mainly composed of the following:
The entry pressure loss in the tangential inlet duct.
The frictional loss along the spiral travel distance in the outer vortex
(caused by friction wall).
The pressure loss caused the change in air flow direction from the
outer to the inner vortex flow.
The pressure loss due to gas flow through the vortex finder.

The losses in the cyclone entrance are often negligible, and the losses in the
cyclone body are significant, while the losses in the vortex finder are the

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

61

largest Avci and Karagoz (2001) and Hoffman and Stein (2002). Cyclone
pressure loss can be expressed as a number of inlet velocity heads, NH.
Velocity heads can be converted to pressure loss units:

= N H 0.5 g Vi2

(2)

where the number of inlet velocity heads, NH, will be constant for any
cyclone design although the pressure loss, P, varies with different
operating conditions. Thus, pressure drop for a cyclone can be established
by determining NH experimentally for a particular cyclone design Leith and
Lee (1997). Values of NH are listed in Table 2.2 for a several sets of
cyclone dimension ratios. Moreover, many analytical expressions for
determining NH from cyclone geometry have been investigated, and several
are listed in Table 3.1

On the other hand, Shepherd and Lapple (1940) gave a simpler expression
to use, and while it does not include all cyclone dimensions, appears to it
give good results Casal and Martinez (1983).

N H = 16

ab
D e2

(3)

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

62

Table 3.1 Several equations for predicting pressure loss based on


number of inlet velocity heads Leith and Lee (1997)

Please see print copy for Table 3.1

3.3.1 Vortex Finder Sizes

The reverse flow cyclone with tangential inlet velocity consists of eight
main parts, as shown in Fig.2.1: gas inlet height (a), gas inlet width (b),
cyclone body diameter (D), cyclone cylinder height (h), cyclone body
height (H), cyclone solids outlet diameter (B), gas exit (vortex finder)
diameter (De) and vortex finder height (S). It is well known that these eight
dimensions describe cyclone pressure drop Saltzman and Hochstrasser
(1983), Iozia and Leith (1989), Kim and Lee (1990), Moore and Mcfarland

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

63

(1993), Zhu and Lee (1999), and Xiang and Lee (2001). These researchers
have found that the vortex finder size is very important dimension that
significantly affects the cyclone pressure drop and plays a critical role in
defining the flow pattern inside the cyclone chamber. Many studies have
been done on cyclone design and these have mentioned an effect of vortex
finder size on pressure drop. For example, Saltzman and Hochstrasser
(1983) studied the design and performance of cyclones with a different
combination of three cyclone cone lengths and three vortex finder
diameters. Kim and Lee (1990) illustrated how the ratio of the diameters of
cyclones body (D) and the vortex finder (De) affected pressure drop.
Moore and Mcfarland (1993) also tested three different cyclone body
diameters with three different outlet tube sizes. Different combinations
were used to vary the ratio of the outlet tube diameter to cyclone body
diameter. A functional relationship has been provided to facilitate design
through permitting calculation of the cyclone size for a given flow rate,
cutpoint diameter, and ratio of body diameter to outlet tube diameter.

3.3.1.1

Swirling Flow in Vortex Finder

It is generally understand that the vortex finder plays a significant role in


defining the flow field inside the cyclone, including the pattern of the outer
and inner spiral flows. The normal procedure to measure cyclone pressure

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

64

drop in industry is to measure the static pressure using a standard pressure


tapping at the wall on the upstream and downstream pipes or ducts. In a
cyclone separator this method of measuring pressure can not work
properly, because of the swirl motion in the exiting gas. The complexity of
this motion inside the cyclone separator has caused difficulties in
understanding well the details of the vortex flow mechanism Walton
(1974). The action of the vortex flow in the cyclone separator is
complicated and affected by different instability phenomena such as
recirculation zone phenomena, vortex breakdown phenomena and
processing vortex core phenomena Yazdabadi et al. (1994) and Griffiths et
al. (1996). This was described earlier by Shepherd and Lapple (1939) and
also recently confirmed by Hoffmann et al. (2001) and Hoffmann and stein
(2002). A significant dynamic pressure is stored in the swirl flow, and the
static pressure is not regular in the swirling motion.

To increase the understanding in the above areas, experimental work has


been undertaken in Bulk Material Handling Laboratory at University of
Wollongong to explore the effects of vortex finder size and different flow
rates on cyclone performance. Also, additional tests have been carried in
this work to increase understanding of the swirl flow and it is effects on

Chapter 3

65

The Theory of Cyclone Pressure Drop and Collection Efficiency

pressure drop by establishing more stable vortex structures of the cyclone


vortex flow.

Models for Cyclone Pressure Drop

3.4

3.4.1 Models Based on Estimating the Dissipative Loss

3.4.1.1

Stairmand Model

Stairmand (1949) determined the velocity distribution in the cyclone from


a moment-of-momentum balance, and then determined the pressure drop in
terms of entrance and exit losses combined with the loss of static pressure
in the swirl. Stairmand gave the following formulae to predict the pressure
drop:

4 ab

2 2(D b)
=
=
+

1
Eu
1
2
q
in

+ 2 2
D
1/ 2Vin2
e

De

(4)

with
De

2(D b)

q=

0.5

De
4A R G
+
+
ab
2(D b)
2A R G
ab

0.5

2
D 2 D e2
(D + D d ) 2 D D d
+ D (H H c ) + D e S +
AR =
Hc +

4
2
2

(5)

0.5

(6)

Chapter 3

where

The Theory of Cyclone Pressure Drop and Collection Efficiency

AR

66

the total wall area of the cyclone body

(the inner walls of the lid, the cylindrical and conical sections and the outer
wall of the vortex finder), and G is wall friction factor = (f/2), which
Stairmand set to equal 0.005 for condition which normally apply to flow
in cyclones Hoffman and Stein (2002).

3.4.1.2

Barth Model

In (1956) Barth reported that the dissipative loss as separate contributions


is from: the cyclone inlet, the cyclone body and the vortex finder. Also, he
stated that in cyclone design the inlet losses can be avoided, and also that
the pressure loss in the body is due to the decrease in dynamic pressure at
the imaginary friction surface, which extends down from the vortex finder.

body
De
1
=
2
2

1/ 2 vin D v

(
H
S
)
e
f
v tCS 0.5De

2
v tCS


v
e

(7)

Furthermore, estimated that the loss in the vortex finder by semi-empirical


approach

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

v tCS
v tCS
e

=
+
K
1 / 2 ve2 ve
v
e

67

4/ 3

(8)

Where, K is found empirically to take on the values 3.41 and 4.4 for vortex
finders with rounded and sharp edges, respectively.

3.4.1.3

Core Model

This model has been derived by Hoffmann et al. (2002) to determine the
pressure drop across a cyclones vortex finder only. It calculates the loss in
static pressure in the vortex finder pipe. The flow in the vortex finder is
assumed to compose of an outer annulus and a potential vortex flow with a
uniform axial velocity surrounding a core of solid-body rotation. Within a
core of radius, rc, the vortex undergoes solid-body rotation with negligible
axial flow, see Fig. 3.4.

e =

g
2 1 2
1
ve + 2 vtCS

1
+

2 2
2 (1 RCe
)
RCe

(9)

This equation shows that both axial and tangential velocity component
contribute to the static pressure drop across the vortex finder.

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

Note: R Ce =

68

RC
Re

ratio of vortex core to vortex finder radii.

vtCS

tangential velocity component in the surface CS.

ve

gas velocity in vortex finder (external)

for free vortex region the tangential velocity at radius Re, can be expressed
as:

v tCS =

Cons tan t
Re

Please see print copy for Figure 3.4

Figure 3.4 Schematic of core model Hoffmann et al. (2002)

(10)

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

69

3.4.2 Purely Empirical Models

There are two empirical models that have been used widely:
3.4.2.1

Shepherd and Lapple (1940) suggested:

NH =
3.4.2.2

16 a b
D e2

(11)

Casal and Martinez-Benet (1983) gave:

ab
N H = 3.33 + 11.3 2
De

3.5

(12)

Other Selected Pressure Drop Models

3.5.1 EEUA Model

EEUA (1997) focuses on three flow losses associated with the cyclone to
determine pressure drop.
Barrel friction (Pa):

B =

2 f g v in2 D c N S
d in

(13)

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

70

where the Reynolds number for determining the friction factor f is based
on the inlet area.

Re =

d in g v in2
(14)

12 g

and the inlet diameter d in =

4(inlet area )
inlet perimeter

Gas flow reversal (Pa):

r =

g vin2
(15)

Exit contraction (Pa):

exit = 0.5 g ( e2 - c2 + k e2 )

where

(16)

gas exit velocity

(feet/sec)

cyclone gas velocity

(feet/sec)

and k is based on the ratio of barrel to exit tube area (De/Dc). The cyclone
pressure drop P is the sum of the three individual pressure drops.

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

71

3.5.2 Jacob and Dhodapkar Model

This model expresses pressure drop in terms of the number of velocity


heads NH and suggested that experimental data be used to confirm this
relationship.

Without

such

data,

Jacob

and

Dhodapkar

(1979)

recommended that the number of velocity heads be calculated from:

H =

16 a b
D e2

= 0.003 g v in2 N H

(17)

(H2O gauge)

(18)

3.5.3 Rhodes Model

This model is based on a resistance coefficient and the Euler number Eu,
which relates the cyclone pressure drop P (Pa) to a characteristic velocity

v, Rhodes (1998).

g 2

= E u
2

(19)

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

72

The characteristic velocity v (m/s) can be defined for gas cyclones in


various ways but the simplest and most appropriate definition is based on
the cross-section of the cylindrical body of the cyclone:

where

4q
D c2

gas flow rate

Dc

cyclone inside diameter (m)

(20)

(m3/s)

The Euler number represents the ratio of pressure forces to the internal
forces acting on a fluid element. Its value is practically constant for a given
cyclone geometry, independent of the cyclone body diameter.

3.5.4 Mason et al., Model

This model provides a useful approximate prediction of pressure drop


across a cyclone (Pa) Mason et al. (1983):

= C

area of gas inlet


inlet velocityhead
area of gas outlet

(21)

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

73

where the loss coefficient, C can be estimated from the Nomograph of


Mason et al. (1983)

3.5.5 Zenz Model

As with the EEUA Model, this model illustrates three flow losses
associated with cyclone pressure drop in (H2O gauge) Zenz (1999):
Barrel loss (H2O gauge):

B =

f D c g v in2 S
26.2 d hi

(22)

Where: f is the friction factor obtained from a function of Re.

Reversal loss (H2O gauge):

r =

g v in2
335

(23)

Exit contraction (H2O gauge):

e = 0.00298 g (e2 - 2B + ke2 )

(24)

Chapter 3

74

The Theory of Cyclone Pressure Drop and Collection Efficiency

3.6 Solids Mass Flow Rate


The performance of any solid separation is estimated by either its overall
collection efficiency (oc) or fractional collection efficiency (fc). The
overall efficiency is defined as:

OC= (Mc / Mi ) = [(Mi Me ) Mi ]

(25)

where, Mc is the mass flow rate of solids collected by the cyclone for an
input of solids flow, Mi. The difference between these two flows is the loss
of solids due to entrainment, Me.

For the estimation of the overall collection efficiency, the solids include all
the particles irrespective of their grain sizes. However, the collection
efficiency of any solids separator depends on the grain sizes of the
particles.

Hoffmann et al. (2002) have explained the three particle fractions in


cyclone separators and describe their mass flow rates as Mf (mass fed), Mc (mass
collected), and

Me (mass emitted). Based on a mass balance:

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

Mf = Mc + Me

75

(26)

and the overall efficiency is defined as the mass fraction of input solids
captured by the cyclone system:

Mc
M
Mc
=1 e =
Mf
Mf Mc + Me

(27)

On the other hand, the estimation of the efficiency can be observed by


collecting samples and weighing any two of the fractions. It is clear that the
size and the density of the solids play a substantial role in measuring
cyclone efficiency.

3.6.1 Maximum Solids Flow Capacity

Despite the intense research efforts have been done to study gas cleaning
devices particularly cyclone separators and the factors that are affected
their operations. It appears that there has been a complete lack of thought
of well investigated or confirmed models based on maximum solids flow
capacity of cyclones. Zenz (1975) explained that the solids flow capacity of
cyclones can be determined from the relationship for solids gravity flow

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

76

through an orifice of cyclone dust outlet as given by the following


equation:

1
b B
tan d

ms =

(28)

Zenz reported that solids are poured into the top of a pipe fixed beneath
the cyclone ate different rates, depending up on the cyclones feed rates.
On the other side, Zenz rearrangement of Eq.(28) solved for outlet pipe
diameter as:

B 2.5 = 153

where:

ms
b

(29)

outlet diameter

(inches)

ms

solids mass collection

(pounds/sec)

drained angle of repose

solids bulk density in the pipe (pounds/ft3)

Now, it has been arqued in terms of Eq.(28) and Eq.(29) that Zenz have
been developed; Eq. (29) considers too conservative as long as solids flow
do not chocked. If solids are dumped instantaneously into a cone see Fig.

Chapter 3

The Theory of Cyclone Pressure Drop and Collection Efficiency

77

3.5a (the cone representing the cyclone cone), the solids will drain out of it
is bottom nozzle (which is representing the cyclone outlet pipe) at rate that
can be calculated from Eq. (28) or Eq. (29). However, the solids are poured
into the cone at a rate such that the bottom opening is not chocked with
solids see Fig. 3.5b (the same mechanism in the actual cyclone cone), the
materials can pass through the cone at a rate several times as great as that
can be calculated from either equation (28) or (29).

Please see print copy for Figure 3.5

Figure 3.5 Solids mass flow in relation to outlet sizing Zenz (1975)

CHAPTER 4
EXPERIMENTAL FACILITIES AND TECHNIQUES

Experimental studies have been undertaken to understand the effect of


geometrical parameters and operating conditions on gas solid separation
equipment (viz. the cyclone), and also evaluate the theoretical analysis.

4.1

Description of Test Rig

The cyclone separator test rig, as depicted in Figs 4.1, 4.2 and 4.3 was
designed and fabricated to investigate the factors that affect cyclone
separation performance. The following apparatus was used with air only
for the first stage of experiments and then particles were employed for the
second stage.

Generally, the test rig consists of the following components:

air supply

annubar to measure the air mass flow rate throughout the system Fig.
4.4.
pressure regulator to control the inlet pressure to the unit
flow control valve to constrict and control the passage of flow

78

Chapter 4

Experimental Facilities and Techniques

79

temperature gauge to measure the air temperature entering the


cyclone
differential pressure meter (DP meter) in order to measure the
pressure drop across the cyclone (30 H2O full-scale meter was used
to obtain accurate readings), see Fig. 4.5.
two water manometers were used to measure the pressure across the
cyclone as a reference source Fig. 4.6.
two vortex finder diameters (105 and 130mm I.D.) Fig. 4.7.
flow straighteners (105mm I.D. & 130mm I.D.) used at the exit of
the cyclone to straighten the flow before discharge, see Fig. 4.7.
annubar station- composed of supply pressure gauge and annubar
DP meters (30 inch H2O and 6 inch H2O full-scale) Fig. 4.8.
cyclone to be tested
the feeding hopper and the receiving bin
blow tank feeder

Chapter 4

80

Figure 4.1 Layout of cyclone test rig air discharging directly to atmosphere.

Experimental Facilities and Techniques

Please see print copy for Figure 4.1

Chapter 4

81

Figure 4.2 Layout of cyclone test rig air discharging through pipe connected to filter

Experimental Facilities and Techniques

Please see print copy for Figure 4.2

Chapter 4

82

Figure 4.3 General layout of cyclone test rig.

Experimental Facilities and Techniques

Please see print copy for Figure 4.3

Chapter 4

Experimental Facilities and Techniques

Please see print copy for Figure 4.4

Figure 4.4 Annubar

Please see print copy for Figure 4.5

Figure 4.5 Cyclone DP Meters

83

Chapter 4

Experimental Facilities and Techniques

Please see print copy for Figure 4.6

Figure 4.6 Cyclone water manometer

Please see print copy for Figure 4.7

Figure 4.7 Flow straightener and vortex finder

84

Chapter 4

Experimental Facilities and Techniques

85

Please see print copy for Figure 4.8

Figure 4.8 The annubar station

4.1.1 Air Supply and Control

The air supply unit consists of a compressor, air dryer to pressurize and dry
the air. It is clear that the major aim in the selection of an air supply system
for the experimental investigations is to provide sufficient compressed air
at ambient temperature for all experiments stages. In the Bulk Solids
Handling Laboratory, air at a maximum pressure head of 800 kPa g is
available, supplied from the following rotary screw compressors:
Atlas Copco electric powered Model GA-308, 3.1 m 3 min 1
free air delivery.
Ingersoll Rand diesel powered Model P375-WP, 10.6 m 3
min 1 free air delivery.

Chapter 4

86

Experimental Facilities and Techniques

Ingersoll Rand diesel powered Model P850-WGM, 24.1 m 3


min 1 free air delivery.

To monitor the performance of the cyclone separator with different


configurations and operating conditions, any combination of the previous
compressors can be employed for the test rig. The compressors are
connected to an aftercooler, two refrigerated air dryers, air receiver (1.75
m 3 capacity) and pressure regulator followed by a flow control valve that
maintains a fairly constant pressure at the annubar. Fig 4.9 shows a general
arrangement of the air supply system.
P 8 5 0 -W G M
2 4 .1 m 3 /m in

D ie s e l P o w e re d R o ta ry S c re w
C o m p re s s o r o r In g e rs o ll R a n d
C o m p re s s o r

P 3 7 5 -W D
1 0 .6 m 3 /m in
A n n u b ar

Air Receiver

P re s s u re
re g u la to r

Ref. Dryer

1.75 m3

T o T est
R ig

F ilte r

Dryer

A fte r c o o le r

A tla s C o p c o
C o m p re s s o r
3 .1 m 3 /m in

Figure 4.9 General arrangement of compressed air supply

Chapter 4

Experimental Facilities and Techniques

87

However, for this thesis project, the output of only one compressor (P850WGM, 24.1 m 3 /min) was required.

4.1.2 Feeding Vessel and Receiving Bin

A feeding bin with a discharge blow tank mounted underneath supplies the
material to be conveyed and collected. The blow tank is supported by load
cells, see Fig 4.3. The mass of material discharged from the blow tank can
be measured by the load cells. The blow tank is connected to the cyclone
by a 105mm I.D steel pipe line.

The particulate material stored in the feeding vessel was conveyed through
the main line at various rates by varying the air flow distribution to the
blow tank. The gas-solid mixture discharging from the feeding blow tank
then enters the cyclone where the solids are separated from the carrier gas
stream by the virtue of centrifugal force and fall down into the receiving
bin and the gas exits via a central opening at the top.

The mass of the received material can be measured by the load cells which
support the receiving bin. The cyclone pressure tappings to measure the
pressure drop are located at the cyclone entrance section and at the end of

Chapter 4

Experimental Facilities and Techniques

88

cyclone outlet pipe. The pressure drop can be varied by selecting two
different diameters of gas outlet pipe (vortex finder) Figs 4.1, 4.2 and 4.3.

4.1.3 Rotary Valve

Rotary valves are devices that are most widely used in all pneumatic
conveying feeding applications. They are designed to collect and transfer
dry granular solids and powders into the conveying gas system. There are
two main types of rotary valve: Drop-through and Blow-through, Marcus et
al. (1990).

During the initial stages of the experimental work in this study, a dropthrough rotary valve was used as the product feeder, see Fig. 4.10, and was
mounted underneath the feeding bin. The materials were dropped into a
feeding shoe which linked the valve to the pipeline. Unfortunately, the
rotary valve could not achieve the feed rates needed to achieve and also
explore the maximum solids capacity of cyclone separators. Hence, it was
found necessary to utilize a blow tank feeder for the investigations into the
maximum solids capacity of cyclone separator, see Figs. 4.11 and 4.12.

Chapter 4

Experimental Facilities and Techniques

89

Please see print copy for Figure 4.10

Figure 4.10 Rotary valve

4.1.4 Blow Tank

Blow tanks are a common high-pressure feeding device. There are two
main configurations of blow tank: top discharge and bottom discharge.
Blow tank operation usually requires high-pressure compressed air. The
blow tank used in this study has a capacity of about 0.5 m3, is mounted on
shear-beam load cells and comprises a fluidizing-discharge cone and outlet
valve see Figs. 4.11 and 4.12.

Chapter 4

Experimental Facilities and Techniques

Please see print copy for Figure 4.11

Figure 4.11 Blow tank Feeder

Please see print copy for Figure 4.12

Figure 4.12 General layout of blow tank feeding system

90

Chapter 4

Experimental Facilities and Techniques

91

Plastic pellets, corn, and rape seed (mean particle size: 4.3 mm and particle
density: 920.5 kg/m3; 7.4 mm and 1380 kg/m3; 1.6 mm and 1139.6 kg/m3,
respectively) were employed as the test materials, see Fig.4.13a, b and c
respectively.

Please see print copy for Figure 4.13

Figure 4.13 Samples of materials used and tested by the cyclone system in
this study

4.1.5 Conveying Lines

The lines used in the experimental work to convey the various materials
consisted of 105 mm I.D mild steel pipe. Pressure tappings were located
along the mild steel pipe.

Chapter 4

Experimental Facilities and Techniques

92

4.1.6 Cyclone Separator

Figs. 4.1, 4.2 and 4.3 show the general configuration of the test cyclone.
This unit and other associated connections were constructed to provide
easy changing of other geometrical parts. Table 4.1 provides the details and
diameters of the cyclone that was tested and summary of relevant operating
conditions.

Table 4.1 Cyclone dimensions and operating conditions

Cyclone Dimensions
(see Fig. 2.1)

D = 400 mm
a = 155 mm

Operating Conditions

g = 1.2 kg/m 3
g = 0.000018 Pa s
Tg = 20 C (approx.)

b = 52 mm
De = 105 mm, and
De = 130 mm
S = 200 mm

Max.v in = 25 m/s
Air supply pressure = 600 kPa g
Air supply annubar = GNT-1.5 inch
Ambient air temperature = 20 C
(approx.)

h = 500 mm

Ambient air pressure = 100 kPa abs

H = 1300 mm

(approx.)

B = 82 mm

Ambient air humidity = 60 to 80 %


(approx.)

= 11

Chapter 4

Experimental Facilities and Techniques

93

4.1.7 Material Specification

Cyclones separators can be made from a variety of materials. The final


selection of wall material depends on the erosion characteristics of the dust,
the corrosion characteristics of the gas, and the operating pressure and
temperature of the cyclone itself. Generally, the test cyclone was
constructed of mild steel. Also, two glass plate windows on the top plate
(roof) and outlet dust pipe were installed to observe flow patterns and
measure particle velocity, respectively.

4.2 Instrumentation and Data Acquisition

All the instrumentation was selected to measure directly the following


parameters during the experimental work on the cyclone separator:
Mass flow rate of air (Mfa)
Mass flow rate of solids (Ms)
Static air pressure at the entrance of cyclone (SP)
The differential pressure between the entrance point and the gas exit
point at the top of cyclone (DP)

Chapter 4

Experimental Facilities and Techniques

94

4.2.1 Mass Flow-Rate of Air

Valves of annubar differential pressure and static pressure were recorded


and converted to mass flow rate by using a computer program. Conversion
charts also were available for set-up and checking purposes.

4.2.2 Mass Flow-Rate of Solids

Shear-beam-type load cells were used to support the feeding bin and the
receiving bin, see Fig. 4.3. The mass of material discharging from or
loading into these containers over a period of time can be measured by
these load cells.

4.2.3 Pressure Drop

The static pressure and differential pressure (DP) of the cyclone separator
were measured by using pressure tapping points, see Fig. 4.14. The first
pressure tap was located at the inlet section before the cyclone entrance to
determine the static pressure and the second pressure tap was located at the
exit section after the flow straightener on the top of the cyclone (to ensure a
non-swirling flow and avoid any dynamic effects). The differential pressure
between the entrance and exit were monitored by using a DP meter and
water manometer (in parallel) to ensure good accuracy.

Chapter 4

Experimental Facilities and Techniques

95

Please see print copy for Figure 4.14

Figure 4.14 Air pressure tapping position

All the DP meters used in the experimental work were checked with a
water manometer Fig. 4.6.

4.2.4 Data Acquisition System

A computer - based data acquisition system (DAS), see Fig. 4.15, was used
to collect all the experimental data.
This system consisted of electronic devices (Load cells, DP pressure and
Annubar) that were plugged into a zero box. The signals then were sent to a
DataTaker 500 using an RS232 com cable that was connected to a
desktop computer. Computer software named Delogger read the data
from the DataTaker 500 and then recorded and saved the data. The raw
data was processed and analyzed by a computer program called Excel
spreadsheet.

Chapter 4

96

Experimental Facilities and Techniques

Annubar

COMPUTER
(data are stored directly onto
computer hard disk or
floppy disk)

Differential Pressure Across Cyclone

DATATAKER

Upstream or Static Pressure

Annubar
Differential
Pressure

Total Pipeline Pressure

DP

ZERO BOX

Feeding Bin Load Cells

Receiving Bin Load Cells

Annubar
Supply
Pressure

Figure 4.15 Data acquisition system

4.2.5 Data Processing

The computer program that was used for data processing was called Excel
spreadsheet and was developed by the researchers of the Bulk Solids
Handling research group. The program saves the raw data in Delogger
and then inputs it into the Excel spreadsheet for processing. This data

Chapter 4

97

Experimental Facilities and Techniques

processing program can read the data files directly and display the signals
as required. The major functions of Excel spreadsheet are:
1. plot and calculate mass flow-rate with respect to cycle time.
2. calculate flow characteristics (e.g. average pressure and solid mass
flow rate).
Typical experimental plots are provided in Fig. 4.16.
Furthermore, it is used for statistical calculation of some major parameters,
including the average value of the mass flow-rate of air, average static and
differential pressure and solids mass flow rate discharging from the feeding
blow tank and receiving bin.
EXPERIMENT NO. 11

TEST DATE: 29/10/2004

150

Mass of Solids (kg)

125
Blow Tank
Load Cells

100

Receiving Bin
Load Cells

75
50
25
0
0

50

100

150

Cycle Time (sec)

(a) Load cell chart (mass of solids)

200

Chapter 4

98

Experimental Facilities and Techniques

0.32

Air Mass Flowrate (kg/s)

0.28
0.24
0.20
0.16
0.12
0.08
0.04
0.00
0

15

30

45

60

75

90

105

120

135

150

120

135

150

Cycle Time (sec)

(b) Air mass flowrate chart


900

Annubar DP (mm H2O)

800
700
600
500
400
300
200
100
0
0

15

30

45

60

75

90

Cycle Time (sec)

(c) Annubar DP chart

105

Chapter 4

99

Experimental Facilities and Techniques

Differential Pressure (kPa g)

10.5
9.0

SP----inlet static pressure


SPD----cyclone static pressure

SP

7.5
SPD

6.0
4.5
3.0
1.5
0.0
0

15

30

45

60
75
90
Cycle Time (sec)

105

120

135

150

(d) Differential pressure chart

Total Pipeline Pressure (kPa g)

30
Blowtank Pressure

Inetial
Pressurisation of
blow tank

25

Total Pressure

20
15
10
5
0
0

15

30

45

60

75

90

105

120

135

150

Cycle Time (sec)

(e) Pipeline pressure chart

Figure 4.16 Typical experimental plots created by Excel Spreadsheet


program

Chapter 4

Experimental Facilities and Techniques

100

The static differential pressure chart, Fig. 4.16d, shows that the cyclone
inlet static pressure and cyclone static pressure drop decrease dramatically
at around 60 seconds then material is fed into the pipeline and cyclone. The
static pressure and static pressure drop remain steady during the time of
conveying and then increase slightly at 112 seconds when the material feed
is stopped or finishes. It is interesting to note that the final air-only static
pressure and static pressure drop do not return to their initial values (vis.
8.5 kPa g and 6.5 kPa g, respectively). The above findings confirm the idea
that the pressure drop across a cyclone separator decreases when solid
particles are introduced into the flow, and remains low when the solid
particles are conveyed completely from the cyclone, Yuu ey al. (1978) and
Beekmans and Morin (1987).

Figure 4.16e shows the initial pressurization of the blow tank (before
opening its outlet or discharge valve) and the conveying pressure. At the
end of each test and after the material was conveyed entirely from the
cyclone test separator, it was observed that the blow tank pressure
remained high. The reason for this is after conveying of the material is
complete the blow tank discharge valve remains open and the blow tank
pressure transducer records an air only back pressure.

Chapter 4

4.3

Experimental Facilities and Techniques

101

Calibration

4.3.1 Load Cells Calibration

Load cells, which are used to support the feeding bin and receiving bin,
determine the mass of material loading into or discharging from the
containers. The calibration of the load cells is carried out by filling the feed
hopper with product as following:
1. Remove any previously used material from the rig and clean out the
rig with a high flow-rate of air. Record the voltage output of all load
cells;
2. Record the voltage output of the load cells that support the blow tank
while it is empty (0 kg);
3. Load a given mass of a product (say 120 kg) into the blow tank.
Then record the voltage output of the load cells;
4. Record the voltage output of the load cells that support the receiving
bin while it is empty (0 kg);
5. Convey all the material to the receiving bin and record the voltage
output from the load cells of the receiving bin.

Table 4.2 shows the recorded results of mass and voltage. The test material
used in this case was rape seed.

Chapter 4

102

Experimental Facilities and Techniques

The calibration results of the load cells are illustrated in Figs. 4.17 and 4.18.
The linearity is quite good for each set of load cells. The calibration factors
for the load cells of the blow tank and receiving bin were 187.06 kg/mv
and 182.48 kg/mv, respectively.

Table 4.2 Mass of material and voltage measured by load cells

Blow Tank
Mass of Product

Receiving Bin
mv

(kg)

Mass of Product

mv

(kg)

2.9801

1.9881

40

3.2542

40

2.2623

80

3.3124

80

2.4743

120

3.6216

120

2.6457

Chapter 4

103

Experimental Facilities and Techniques

140
y = 187.06x - 557.46

Mass of Product (kg)

120
100
80
60
40
20
0
0

0.5

1.5

2.5

3.5

Voltage Response of Load Cells (mv)

Figure 4.17 Calibration of feeding bin load cells

140

Mass of Product (kg)

120

y = 182.48x - 362.78

100
80
60
40
20
0
0

0.5

1.5

2.5

Voltage Response of Load Cells (mv)

Figure 4.18 Calibration of receiving bin load cells

Chapter 4

4.4

104

Experimental Facilities and Techniques

Test Materials and Properties

4.4.1 Test Materials

It is clear that cyclone performance and other kinds of gas cleaning devices
are affected by the properties of the materials to be collected. Hence for
improving the performance of these devices in particular gas - solid
separator equipment, the influence of the properties of the materials must
be considered and recorded properly. The following sections introduce
these properties and their measurement methods. The materials that were
chosen in this study were Plastic pellets, rape seeds and corn Table 4.3.

Table 4.3 Physical properties of materials tested

Material

Solid Density

Bulk Density

Median Particle

tested

(kg/m3)

(kg/m3)

Diameter (mm)

Plastic Pellets

920

531

4.3

1140

621

1.6

1380

670

7.4

Rape Seed
(Canola)
Corn

Chapter 4

105

Experimental Facilities and Techniques

4.4.2 Particle Size and Distribution

Particle size and distribution characteristics are most often used in gas-solid
separation. Also, particle size analysis of product is needed to determine
conveying velocities of cyclone separator and other types of feeding
devices. However, there are many definitions of particle size, but none has
been adopted as a comprehensive standard because of the variation of
particle geometrical shapes. The term particle size means the average
dimension of the particle. For a spherical particle shape, see Fig. 4.19a.
The size for this particle can be defined easily as its actual diameter.
However, for non spherical or irregular particles shaped, see Fig. 4.19b, c,
the definition of particle size depends on the method of measurement. In an
attempt to deal with the size of irregular particles, the term most often used
is equivalent diameter. This refers to the diameter of a sphere which
exhibits the same physical behavior as the particle when subjected to the
same sizing method, for example, a sphere particle having the same
projected area, volume or mass or just passing through a sieve aperture.

(a)

(b)

(c)

Figure 4.19 Regular and irregular shaped particles

Chapter 4

Experimental Facilities and Techniques

106

Thus the measurement of the size (equivalent diameter) of particles is


depends on the method used to determine that parameter.
There are many methods that can be used for determining the size
distribution of particulate materials, such as woven wire sieves, optical
microscopy, sedimentation, x-ray absorption and centrifugal setting,
cascade impactor, aerodynamic sizing and modern optical methods (e.g.
laser diffraction) Cheremisinoff et al. (1986) and Wang (1995). Among all
of these methods, woven wire sieves or mechanical sieving is the most
widely used method for determining the size distribution and it is well
known by engineers and researchers, as it covers a wide range of particle
size in industry. Such equipment is either manually or mechanically shaken
for a specific period of time, resulting in a proportion of particles retained
on each screen. Thus the particle size and distribution measured by sieving
can be defined by quoting the apertures of the screens (e.g. mass of
particles passing through or being retained on the apertures).
The most useful method to determine the particle size and distribution data
is to plot the particle size or sieve size against the mass percentage of the
sample in a spreadsheet, Figs.4.20, 4.21 and 4.22.
A common method used for a assigning a characteristic figure to this
information is by determining the median size. This is defined as the
particle size which represents 50 % of the sample by mass.

Chapter 4

107

Experimental Facilities and Techniques

mass>d (%)

Cumulative Distribution

100

50

0
1

d (mm)
d50= (4.4mm)

10

Figure 4.20 Particle size distribution (Plastic Pellets)

mass>d (%)

Cumulative Distribution

100

50

0
1

10
d(mm)
d50=7.5mm

Figure 4.21 Particle size distribution (Corn)

100

Chapter 4

108

Experimental Facilities and Techniques

mass>d (%)

Cumulative Distribution

100

50

0
1

10

d(mm)
d50=1.7mm

Figure 4.22 Particle size distribution (Canola)

Note: Due to the limitation sieve sizes available for the product of rape
seed (canola), the curve is not appears as smooth as the other products
(plastic pellets and corn).

4.4.3 Loose Poured Bulk Density

Loose poured density (sometimes called the poured bulk density) is defined
as the mass of bulk solids divided by the total volume of the particles
(including voids). For a particular solid, the bulk density does not have a
unique value and it varies with the condition of the bulk solid (e.g.

Chapter 4

Experimental Facilities and Techniques

109

consolidation pressure). It is dependent on the particle density, particle


shape and the way that particles are packed or positioned in terms of one
another. For example, bulk solids that are conveyed pneumatically may be
aerated and have a lower bulk density. So, it is not always easy to
determine the bulk density of a product under changing consolidation
conditions. In cyclone separator systems, where the bulk solids flow and
accumulate by virtue of centrifugal force and gravity, it can be
considered that the bulk density of solids within a cyclone approaches a
loose poured condition.

The following method was used to determine loose-poured bulk density:


i Pour a certain amount of bulk solid carefully and gently into a
measuring cylinder. The measuring cylinder must be held at an
angle of 45 degrees to the horizontal (avoid compaction when
pouring the solids).
ii Note the volume occupied by the bulk solid ( when the cylinder is in
the vertical position ).
iii Weigh the cylinder with bulk solid and calculate the mass of the
bulk solid.

iv Determine the loose poured bulk density by dividing the mass of


the bulk solids by the poured volume of the bulk solid.

Chapter 4

Experimental Facilities and Techniques

110

4.4.4 Particle Density

In addition to particle size, the particle density plays an important role in


determining particle motion in fluids, and hence cyclones. Particle density
is defined as the average mass of a single particle divided by its actual
volume. The larger the dust particle the more easily it is separated in a
cyclone. This is due the fact that with increasing particle size the
centrifugal force will increase too, which will help to force the particles to
the cyclone wall. The difference in particle density to the gas stream
density also plays a very important role. If this difference in density
increases, then, the efficiency will increase somewhat. Particle density is
often be measured by using a pycnometer or water displacement method.

4.4.5 Angle of Repose

The poured angle of repose is defined as the angle of the free surface of a
pile of product with the horizontal plane. This angle is helpful in
determining the design of hoppers and bins as well as spouting to move this
product by gravity. An angle of repose less than 30 indicates that the
product is very free flowing. Angles of repose greater than 30 and less
than 45 is free flowing and the angle above 45 is sluggish. There are two

Chapter 4

Experimental Facilities and Techniques

111

different angles of repose: poured angle of repose and drained angle of


repose Gerchow (1986).

4.4.5.1

Poured Angle of Repose

Angle of repose can be obtained by the following steps:


(i)

Collect the poured angle of repose apparatus.

(ii)

Set the funnel plate height to an approximate height that you


think the product may pile up to. Use the pegs to hold the plate
into certain position.

(iii)

Fill the funnel with sample whilst holding the plate under the
opening so that no sample falls out. When the funnel is full with
sufficient sample move the plate to allow the sample to flow out
onto the tray, (if the sample spread too far on the plate the
funnel height should be lowered).

(iv)

Agitate the sample in the funnel to assist flow (if needed) and
obtain steady flow of material.

(v)

Continue to pour sample until the tip of the product cone seals
the funnel stem. Hence, no more material can flow from the
funnel.

(vi)

Lift off the funnel to allow the heap to come to a point.

Chapter 4

Experimental Facilities and Techniques

112

(vii) Measure the length of the base of the cone (at four different
points) in millimeters.
(viii) Measure the height of the bottom of the funnel stem. i.e. the top
of the heap.
(ix)

Measure the slope angle of the cone using the sprit level
compass. Do this in four different areas at least. Note these
values.

The angle also can be measured by using the digital spirit level:
- Turn on spirit level.
- Check zero.
- Carefully place the level at the base of the cone and lay down
on cone.
- Read angle.
- Write down results.
- Repeat this on four sides of the cone. Average the results
(x)

Remove and clean up sample.

(xi)

Redo (ii) to (viii) three times, so that four tests in total are
completed.

Chapter 4

Experimental Facilities and Techniques

113

Please see print copy for Figure 4.23

Figure 4.23 General configuration of poured angle of repose testing

4.4.5.2

Drained Angle of Repose

Drained Angle of Repose can be obtained by the following steps Fig. 4.24:

(i)

Collect Drained Angle of Repose apparatus.

(ii)

Check that the discharge valve is closed.

Chapter 4

Experimental Facilities and Techniques

114

(iii)

Measure the full length of the sample plate (L).

(iv)

Measure the height of the sample plate from the top of the stand.
Do this by placing a ruler across the top of the stand and another
ruler (with the markings starting from the edge) gently touching
the tip of the cone and reading the height where it lines up with
the first ruler.(H1) Make sure that the ruler used to measure
height starts at zero at the edge of the ruler.

(v)

Fill the container with sample (try to direct the sample onto the
sample plate) high enough to cover sample plate, and to the top
of the funnel. When it looks like the sample cone and plate will
no longer take any more material, stop the filling.

(vi)

When the sample is stable, open the discharge valve and allow
the sample to flow out.

(vii) Do not tap the stand to assist with the discharge.


(viii) When all the sample has discharged from the funnel carefully
measure the height of the tip of the cone to the top of the stand.
Do this by placing a ruler across the top of the stand and another
ruler (with the markings starting from the edge) gently touching
the tip of the cone and reading the height where it lines up with
the first ruler. (H2) Make sure that the ruler used to measure
height starts at zero at the edge of the ruler not in from the edge.

Chapter 4

Experimental Facilities and Techniques

115

(ix)

Clean away sample.

(x)

Redo (v) to (x) three times so a total of four tests have been
completed.

The results can be calculating theoretically for both poured and drained
angle of repose by:
H = H1 - H2
tan = H/ (L/2)

Please see print copy for Figure 4.24

Figure 4.24 General configuration of drained angle of repose testing

Chapter 4

Experimental Facilities and Techniques

116

4.5 Cyclone Experimental Procedure

After all the instruments were installed and calibrated as required, the test
rig was checked for proper operation. After finishing the necessary checks,
experiments were carried out using the procedures described in the
following section.
The principal aspects measured in the experiments were effect of outlet
piping and high solids concentration for different bulk materials on cyclone
performance and the saturation capacity of a cyclone separator system.

4.5.1 Cyclone Performance - Air Only

These tests were carried out by running air only through the system at a
given air flow rate. Different inlet velocities were employed, and pressure
drop difference across the cyclone was measured. The static pressure at the
cyclone entrance also was measured. In this stage of experiment, two
cyclone configurations were used: air discharging directly to atmosphere;
air discharging through a pipe connected to a filter. The effect of vortex
finder on pressure drop was investigated by selecting two different
diameters of vortex finder mounted at the top of the cyclone. This allowed
the determination of the relationship between the air flow rate and the static

Chapter 4

Experimental Facilities and Techniques

117

or total pressure at cyclone inlet and also between inlet velocities and total
pressure drop.

4.5.2 Cyclone Performance - with Solids

The procedures in this test were similar to the air-only performance test
except that products were introduced to the system. In addition to repeating
the measurements that were taken in the air-only performance test, the
mass variations of the feeding and receiving bins were monitored using
load cells to allow the determination of conveying rate.
Using the blow tank feeder, the materials were conveyed from the feeding
bin to the cyclone separator unit, where the mechanism of particle
separation took place inside the cyclone under ambient conditions or under
certain cyclone operating conditions. The maximum capacity of the
materials being collected by the cyclone system was measured for each
material using the Data Logger Program. This was achieved by
increasing the proportion of air flow to the blow tank feeder. The results
are summarized in Table 4.4.

Furthermore, Table 4.5 shows data for tests where particles did come out
from the cyclone using the same operating conditions. During the time of
testing and by the assistance of glass windows on the roof of the cyclone,

Chapter 4

118

Experimental Facilities and Techniques

building up the solids flow patterns inside the cyclone and choking flow
phenomena in the cyclone dust outlet were visually observed. The
materials were conveyed from the feeding bin to the cyclone using the
blow tank feeder and the cycle time of particles to be separated from the
cyclone was taken since the product entered the cyclone body, using a
stopwatch. It was observed that the conveying air flow rate selection played
the main rule in terms of built up the materials inside the cyclone body, (e.g.
reducing air flow rate with keeping blow tank air flow constant), see Table
4.5.

Table 4.4 Maximum capacity of material to be separated

Plastic Pellets
Test

Blow tank Total air mass flowrate

no.

air flow

Capacity

Visual

(kg/s at 1.2 kg/m3)

(kg/s)

observation

0.25

2.95

No build up of

(max.)

material inside

(kg/s)
1

0.04

cyclone, and no
particles being
lost from cyclone.

0.02

0.2

1.5

0.02

0.15

1.58

Chapter 4

119

Experimental Facilities and Techniques

Rape Seed
Test

Blow tank Total air mass flowrate

no.

air flow

Capacity

Visual

(kg/s at 1.2 kg/m3)

(kg/s)

observation

0.25

3.1

No build up of

(kg/s)
1

0.01

(max.)

material

inside

cyclone, and no
particles

being

lost from cyclone.

0.01

0.2

1.96

0.01

0.15

2.02

Dust outlet
blocked in first 3
second for about
1 second

Corn
Test

Blow tank Total air mass flowrate

no.

air flow

Capacity

Visual

(kg/s at 1.2 kg/m3)

(kg/s)

observation

0.25

2.82

No build up of

(kg/s)
1

0.04

(max.)

material

inside

cyclone, and no
particles

being

lost from cyclone.

0.03

0.2

2.05

0.02

0.15

1.33

Chapter 4

120

Experimental Facilities and Techniques

It was observing also that at the end of conveying of the materials, the
escape of pressurized in the blow tank sometimes causes a momentary
increase in mass solids (ms) resulting a choking (just for a few seconds) in
the cyclone dust outlet, as shown in Table 4.5.

Table 4.5 Choking and escaping of the material from the cyclone

Plastic Pellets
Test

Blow tank Total air mass flowrate

no.

air flow

Capacity

Visual

(kg/s at 1.2 kg/m3)

(kg/s)

observation

0.25

2.88

(kg/s)
1

0.05

Build up of
material inside
cyclone after 4
seconds (3/4
cone) and
147.27g particles
out.

0.04

0.2

2.7

Build up of
material after 15
second (3/4 cone)
and a few
particles out.

Chapter 4

121

Experimental Facilities and Techniques

Rape Seed
Test

Blow tank Total air mass flowrate

Capacity

Visual

no.

air flow

(kg/s at 1.2 kg/m3)

(kg/s)

observation

0.25

3.3

(kg/s)
1

0.02

Build up of
material inside
cyclone after 150
second (3/4 cone)
and 233g particles
out.

0.02

0.2

3.17

Build up of
material inside
cyclone after 3
seconds (1/2
cone) and 227g
particles out.

0.02

0.15

3.17

Build up of
material inside
cyclone after 150
second (3/4 cone)
and 229g particles
out.

Corn
Capacity

Visual

(kg/s at 1.2 kg/m3)

(kg/s)

observation

0.25

3.8

Test

Blow tank Total air mass flowrate

no.

air flow
(kg/s)

0.05

Build up of
material inside
cyclone after 40

Chapter 4

122

Experimental Facilities and Techniques

seconds for
about 2
seconds.

0.05

0.2

3.39

Build up of
material inside
cyclone after 24
seconds (whole
cone).

0.05

0.15

3.7

Build up of
material inside
cyclone after 11
seconds (whole
cyclone).

0.04

0.15

3.07

Build up of
material inside
cyclone after 60
second (whole
cone).

0.03

0.15

2.25

Choking flow
at the end of
conveying for 3
seconds

The operating procedures of the separation performance tests were as


follows:
i.

Set the blow tank air and the total conveying air required for the
experiment by adjusting the airflow control valves.

Chapter 4

ii.

Experimental Facilities and Techniques

123

Open the valve of the feeding bin and load the material into the blow
tank, Fig 4.11 and 4.12, then close the valve after loading.

iii.

Set up the data acquisition system.

iv.

Start the data acquisition system.

v.

Switch on the supply air slowly up to the maximum set point.

vi.

After about 20 seconds open the outlet valve of blow tank to convey
the material into the cyclone system.

vii.

After all the material has been conveyed into the cyclone system,
close the discharge valve of the blow tank.

viii.

At the end of separation, keep the air blowing for a while and then
close the air supply valve.

ix.

Repeat the previous steps from (i) to (viii) after changing the blow
tank and total conveying air.

In order to ensure good accuracy of results and to know exactly the


separation mechanism and the saturation point of the cyclone, two
observation windows were installed at the top of cyclone. All the above
tests were performed under visual observation and under ambient and
steady-state conditions.

Chapter 4

Experimental Facilities and Techniques

124

4.6 Phantom Vision Digital Analysis System

The High Speed video image capture and Phantom Vision software were
used for particle flow analysis. The Phantom High speed Motion
Acquisition and Analysis software had been designed to be used in the
windows operating environment. This software allows the users to enter,
adjust and view all images and data acquisition settings in one convenient
window. A variety of aperture lenses can be used for focusing by the High
Speed Digital Camera, see Fig. 4.25.

Please see print copy for Figure 4.25

Figure 4.25 Camera setup and recording window

Chapter 4

Experimental Facilities and Techniques

125

Phantom offers a variety of processing functions. These functions can be


used on their own or in combination to adjust and clarify the image or to
discover other features and details in terms of processing images. For
example, the tune function can be used to adjust the brightness, contrast or
gamma of the image. The smooth function is an automatic way of
providing a more smooth or refined image. By selecting the sharpness
function, it will automatically improve a low contrast image. Colour
scheme function is also available to make the image clearer and the
particles more visible.

4.6.1 Measurement Steps

The measurement functions for speed and acceleration can only be applied
to cine files (single images extracted from cine files do not contain the
timing information to calculate the velocity).
Once all the information required are collected from the images, they were
imported into Microsoft Excel program.

The Measurement Procedure is described below:


1- Select the measurement units, choose either metric or any other
standard units;

Chapter 4

126

Experimental Facilities and Techniques

2- Scale the image by entering a known dimension of any object (using


a ruler) or linear dimension/scale in the image. Phantom
automatically applies this scale to the entire cine;
3- Set an origin to measure position, distance, displacement, speed or
acceleration and a reference point must be established. Once selected,
(x, y) coordinates (0, 0) will automatically be assigned to that point
for each image in the cine;
4- Select five frames for instance, for a certain single particle (six
images will be displayed in the cine):

(time interval per frame, t) = (number of images 1)


= 6 1 = 5 sec.
5- (x pixels) is the difference distance between the starting point (x1,
y1) coordinates and the ending point (x2, y2) coordinates for a single
particle;
6- The constant of (0.0011 m) from the image: Depends on the distance
between the camera and the object that needs to be measured;
If the distance between two known positions on the ruler (x1, y1) and
(x2, y2) was 0.15 m:
x = (x1 x2) pixels = 0.15 m
1 pixel =

0.15
= 0.0011 m
x

(constant)

Chapter 4

Experimental Facilities and Techniques

127

7- The particle distance D, from the starting point to the ending point
can be represented by:

D = x (0.0011)

(m)

8- The total interval time for the particle, t, can be determined as


follows:

t =

Number of Frames
Frams per Second

In this section two High Speed Video Camera setups were used to calculate
particle velocity: High Speed Camera at the top of cyclone entrance and
High Speed Camera at the bottom of cyclone (particle) outlet, Fig. 4.26.
The experimental apparatus consists of a steel pipeline of 105mm I.D.,
which is connected with a glass pipe with the same diameter at the cyclone
entrance section, and a glass tube of 82 mm I.D beneath the cyclone unit
(outlet opening).

Chapter 4

Experimental Facilities and Techniques

128

Please see print copy for Figure 4.26

Figure 4.26 Schematic diagram of cyclone apparatus rig with High Speed
Camera

By using the High Speed Camera (HSC) the particle flows were captured
from the horizontal and vertical glass pipes (HSC.1 and HSC.2), see
Fig.4.26 at a speed ranging between 1000 and 1635 frames per second.
Different air mass flow rates were used and the mass flow rate of the
materials was measured directly by using the Data Taker program.

CHAPTER 5
EXPERIMENTAL INVESTIGATIONS INTO CYCLONE
PRESSURE DROP AND SOLIDS FLOW CAPACITY

5.1

Introduction

5.1.1 Pressure drop

It is well known that one of the most important operating parameters in


cyclone systems is pressure drop. Small variations in cyclone geometry and
other parameters may cause significant changes in cyclone pressure drop.
In engineering practice, the determination of the ability of any part of
cyclone body to induce a separation of particles from the gas stream and
the parameters that affect the collection efficiency and pressure should be
the first step in designing a cyclone separator.

Consequently a number of studies have been reported to improve cyclone


performance and to reduce pressure drop. For example, modifications to
cyclone geometry have been tested, such as inlet and outlet gas geometry,
Hofmann et al. (1992). More recently, Lim et al. (2003b) proposed a
double cyclone and compared its efficiency parameters to that of a
conventional cyclone, by varying the vortex finder diameter and length.

129

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

130

Hence, it appears that the vortex finder has a significant effect on cyclone
performance. Furthermore, the flow pattern inside the vortex finder was
found to be unstable and too complex to estimate, Lim et al. (2003b).
For this reason a systematic experimental program has been designed and
undertaken in the Bulk Solids Handling Laboratory at Wollongong
University to investigate and improve cyclone performance. The
investigations were aimed at studying and controlling the instability of
vortex flow, as well as examining the effect of outlet piping (a vortex
finder) on cyclone performance. The experimental results obtained from
this work also were compared with pressure drop predictions from various
theoretical models. The experiments were conducted by inserting varying
sizes of flow straightener and vortex finder. Air only experimental
investigations into cyclone pressure drop also have been carried out.

5.1.1.1

Flow Straightener (F.S.)

5.1.1.1.1 The Test Model

The flow straightener test models and the rig were designed in the
laboratory. The details of the test rig are shown in Fig. 5.1. It was designed
to work with two different diameter flow straighteners (105mm I.D and
130mm I.D X 1D height each), as shown in Fig. 5.2, which were equipped

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

131

with two pieces of cylindrical pipe (105mm I.D and 130mm I.D). At the
inlet duct a pressure tapping was used to determine the difference between
static and atmospheric pressure in order to estimate the static pressure drop
of the flow straightener. The static pressure drop measurements across the
flow straightener were made using an inclined manometer for different
inlet air flows.

In order to achieve a uniform direct air flow from the vortex finder (to
allow accurate measurement of flow), experimental work has been
conducted with a flow straightener. Two different sizes of flow straightener
have been used (105mm I.D. and 130mm I.D.) and the length was (1D) for
both to provide enough space for flow guidance, see Fig. 5.2.

5.1.1.1.2

Experimental Results

5.1.1.1.2.1 Influence of Flow Straightener

In order to provide a straight flow inside the vortex finder, the flow
straightener was inserted in the test rig model as shown in Fig. 5.1.The
measurements were made just outside at the outlet (about 500mm away
from flow straightener).The influence of the flow straightener on the flow
field and vortex finder pressure drop was observed. Two test model

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

132

configurations were used: measuring pressure difference both with and


without the flow straightener.

Please see print copy for Figure 5.1

Figure 5.1 Schematic of flow straightener used to determine pressure drop

Please see print copy for Figure 5.2

Figure 5.2 Typical 105mm I.D. and 130mm I.D flow straightener

Chapter 5

133

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

Two different flow straighteners were used to measure the pressure drop by
employing different velocities. Many experimental tests have been made,
and consistent results have been obtained, see Fig. 5.3.

The pressure difference as a function of the velocity can be seen clearly


from Fig. 5.3. The data from these experiments were then used to
determine corresponding loss coefficients. The detailed calculations are
presented in Appendix C. A loss coefficient of K=0.182 was found to
represent well both flow straighteners. A comparison between measured
values of pressure drop due to the flow straightener and predicted values
using K=0.182 is shown in Fig.5.4.

Pressure Difference, SPD (Pa)

140
with flow straightener

120

without flow straightener

100
80
60
40
20
0
0

10

15

20

25

Inlet Velocity, Vi (m/s)


(a) 105mm I.D. flow straightener pressure drop

30

134

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

Pressure Difference, SPD (Pa)

Chapter 5

70
with flow straightener
60

without flow straightener

50
40
30
20
10
0
0

10

15

20

Inlet Velocity, Vi (m/s)


(b) 130mm I.D. flow straightener pressure drop

Predicted Pressure Drop (Pa)

Figure 5.3 Variation in pressure drop with F.S. and without F.S.

80
70
60
50
40
30
20
10
0
0

20

40

60

Exp. Pressure Drop (Pa)

(a) 105 mm I.D. flow straightener

80

135

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

Predicted Pressure Drop (Pa)

Chapter 5

30
25
20
15
10
5
0
0

10

15

20

25

30

Exp.Pressure Drop (Pa)


(b) 130mm I.D. flow straightener

Figure 5.4 Experimental and predicted pressure drop relations for different
flow straighteners

5.1.1.2

Vortex Finder

5.1.1.2.1

Experimental Scheme

The typical configuration of the cyclone vortex finder test rig- (air only)
has been shown in Figs. 4.1 and 4.2. Compressed air is used under ambient
conditions, with the air leaving the cyclone either directly to atmosphere or
via a pipe connected to a filter, as shown in Fig. 4.2. The main dimensions
and operating conditions of the cyclone are presented in Fig. 2.1and Table

Chapter 5

136

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

4.1. The cyclone unit that was the objective of this study had a reversed
flow and tangential rectangular inlet, and was equipped with two different
vortex finder diameters (D =105mm I.D and 130mm I.D), see Fig. 5.5.
Two different cyclone configurations were investigated. The cyclone was
designed and fabricated so that the vortex finder could be separated from

200mm

100mm

the cyclone and easy interchanged

Figure 5.5 Conventional vortex


finder without flow
straightener.

5.1.1.2.2

Air Discharged Directly to Atmosphere

During this stage of the experiment only air was running through the
system at a given flow rate. The vortex finder and the flow straightener
diameters that were tested were 105mm I.D. and 130mm I.D., and these
were open to atmosphere. The flow straighteners were inserted to provide
uniform and a straight flow. The air flow rate was adjusted by a flow

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

137

control valve and pressure regulator located upstream. A pressure tapping


at the cyclone inlet was also used to measure the static pressure at this
stage. Two DP meters and two water manometers were installed on the
cyclone test rig in parallel in order to ensure correct flow measurements for
air pressure and air density changes. Inlet velocity measurements were
taken, up to a maximum of 25 m/s. Differential pressures across the
cyclone were measured by means of two DP meters and two water
manometers. Data readings were taken 4 times each at different flow rates.
No unusual incidents were reported during the experiments.

With the aim of determining the influence of the vortex finder diameter on
cyclone performance, several experiments were carried out, using only air
and vortex finder with two different diameters. Different cyclone inlet
velocities and air-flow rates were tested, in order to elucidate the effect of
these parameters on the vortex finder flow pattern. The pressure within the
cyclone separator was varied by selecting two different diameters of vortex
finder mounted at the top of the cyclone collector. The cyclone pressure
difference then was varied to observe the influence of the vortex finder
diameter on cyclone performance (i.e. pressure drop).

Chapter 5

5.1.1.2.3

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

138

Test Procedure

To start testing the first stage of the cyclone configurations, the air flow
was set from the air compressor and dryer. When the pressure level reached
a sufficient reading, the testing took place as follows:

air was passed through the rig using the flow control valve,
starting with a low flow rate;

readings were recorded after the system reached a satisfactory


and balanced level according to the various water manometers
and DP meters;

then the reading of the annubar was taken via the DP meter;
air flow rates were recorded in kg/sec from the calibration charts;
the previous steps were repeated after increasing the air flow.

5.1.1.2.4

Air Discharged Through a Pipe Connected to a Filter

The experimental procedures in this stage, as shown in Fig. 4.10, were


similar to testing with the previous configuration (air discharged to
atmosphere), except that the vortex finder was connected to the filter unit
and an additional manometer was used. Since this test rig was used for air
only, there was no need to collect any product. However, the bin may cause
the backpressure that would occur during the actual working of the cyclone

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

139

system. All measurements made in the test process when the air was
discharged directly to atmosphere were repeated. The total difference in
pressure as a function of cyclone inlet velocity was varied by using two
different sizes of vortex finder.

5.1.1.2.5

Influence of the Vortex Finder

The influence of changing the gas outlet diameter (vortex finder) on


cyclone performance is represented in Figs. 5.6, 5.7, 5.8, and 5.9. It can be
seen clearly that for a given vortex finder diameter, the cyclone pressure
drop increases with increasing gas inlet velocity, and the pressure drop
across the cyclone is increased by decreasing the gas outlet diameter
(vortex finder). Thus, the vortex finder diameter does have significant
effect on the cyclone pressure drop for a given inlet velocity.

Chapter 5

140

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

Total Pressure, TPi (Pa)

10000
9000
8000

105mm I.D.

7000
6000
5000
4000
3000
2000
1000
0
0

10

15

20

25

30

0.25

0.30

Inlet Velocity,Vi (m/s)

10000

Total Pressure TPi (Pa)

9000
8000

105mm I.D.

7000
6000
5000
4000
3000
2000
1000
0
0.00

0.05

0.10

0.15

0.20

Flow Rate(mf), (kg/s)

Figure 5.6 Influence of 105mm I.D vortex finder on cyclone performance


(air discharging to atmosphere)

Chapter 5

141

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

Total Pressure, TPi (Pa)

7000
6000

130mm I.D.

5000
4000
3000
2000
1000
0
0

10

15

20

25

30

0.25

0.3

Inlet Velocity,Vi (m/s)

7000

Total Pressure, TPi (Pa)

6000

130mm I.D.
5000
4000
3000
2000
1000
0
0

0.05

0.1

0.15

0.2

Flow Rate, mf (kg/s)

Figure 5.7 Influence of 130mm I.D vortex finder on cyclone performance


(air discharging to atmosphere)

Chapter 5

142

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

Total Pressure Difference, TPD (Pa)

7000
6000

105mm I.D.

5000
4000
3000
2000
1000
0
0

10

15

20

25

0.2

0.25

Inlet Velocity,Vi (m/s)

Total Pressure Difference, TPD(Pa)

7000
6000

105mm I.D.

5000
4000
3000
2000
1000
0
0

0.05

0.1

0.15

Flow Rate, mf (kg/s)

Figure 5.8 Influence of 105mm I.D vortex finder on cyclone performance


(air discharging through a pipe connected to a filter)

Total Pressure Difference, TPD (Pa)

Chapter 5

143

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

6000
5000

130mm I.D.

4000
3000
2000
1000
0
0

10

15

20

25

30

0.25

0.3

Total Pressure Difference, TPD (Pa)

Inlet Velocity,Vi (m/s)

6000
5000

130mm I.D.

4000
3000
2000
1000
0
0

0.05

0.1

0.15

0.2

Flow Rate, mf (kg/s)

Fig. 5.9 Influence of 130mm I.D vortex finder on cyclone performance (air
discharging through a pipe connected to a filter)

Chapter 5

5.2

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

144

Pressure Drop Modeling

Tests were carried out with varying outlet geometry and by discharging to
atmosphere directly or via a filter. Predictions of pressure drop from
various models were also calculated for these different conditions.

Note: the flow straightener was used in all experiments.

5.2.1 Modeling for Air Discharging to Atmosphere

Different vortex finder diameters with flow straightener were used (105mm
I.D and 130mm I.D) for modeling the pressure drop with air discharging
directly to atmosphere, as shown in Fig. 5.10.

Please see print copy for Figure 5.10

(a) 105mm I.D. vortex finder

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

145

Figure 5.10 continued

(b) 130mm I.D vortex finder

Figure 5.10 Total pressure drop predictions by various models (105mm I.D
and 130mm I.D) with vortex finder discharging to atmosphere

5.2.2 Modeling for Vortex Finder Connected to the Filter

Different vortex finder diameters with flow straightener were used (105mm
I.D and 130mm I.D) for modeling the pressure drop with the vortex finder
connected to a filter via 105mm I.D. piping, see Fig. 5.11.

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

146

Please see print copy for Figure 5.11

Figure 5.11 Total pressure drop prediction by various models (105mm I.D
and 130mm I.D) with vortex finder connected to a filter

Chapter 5

5.3

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

147

Development of New Pressure Drop Model

Various factors, as mentioned previously, contribute to the prediction of the


cyclone total pressure drop:

Pressure drop due to gas expansion loss at the cyclone entrance;


Pressure drop due to cyclone body (frictional) loss within the
cyclone;

Pressure drop due to gas flow reversal;


Pressure drop due to gas flow loss through the cyclone outlet
(including flow straightener)

Two cyclone configurations were used to develop a new model of cyclone


pressure drop: air discharging directly to atmosphere and air discharging
through a pipe connected to a filter. Two different sizes of vortex finder
were used, (105mm I.D. and 130mm I.D.) for a wide range of entrance
velocities. The predicted values of total pressure drop due to gas expansion
were taken into account at the cyclone inlet and outlet (air discharging
directly to atmosphere).

In the designing of cyclone separators it is important to know the pressure


loss which will be experienced under various conditions of working. There

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

148

are models that estimate the total pressure drop as entrance and exit losses
combined with the loss of static pressure in the swirl. Stairmand (1949)
determined the total pressure drop after he calculated the velocity
distribution in the cyclone from angular momentum and reported that in
practice little of the decrease in static pressure from the outer to the inner
part of the vortex can be recovered in the vortex finder, so that this can be
counted as lost.

Barth (1956) estimated the dissipative loss (i.e. the loss in the sum of the
static and dynamic pressure): P + v2 as separate contributions from the
inlet, the cyclone body and vortex finder losses. By using Eq. (8) in chapter
3 Barth estimated the loss in the vortex finder and used the two values of
constant K that represents the vortex finders with round and sharp edges (k1
and k2), respectively.

In addition the models of Jacob and Dhodapker (1979), Mason et al. (1983)
and Rhodes (1998) are used to predict the total pressure drop as entrance
and exit losses and are combined with the authors total pressure drop
model for the flow straightener.

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

149

The models of EEUA (1997) and Zenz (1999) illustrate three flow losses
associated with the cyclone to estimate the pressure drop. Thus, the total
pressure drop predicted using these models is based on summing the exit
expansion loss, the pressure drop due to flow straightener, the pressure
drop due to vortex finder (outlet duct), the pressure drop due to contraction
loss, the expansion loss at the gas inlet, and the entry friction loss.

The total pressure drop is made up of the separate losses in the inlet pipe, in
the cyclone body, and in the cyclone exit duct (vortex finder). In predicting
the total pressure drop in the cyclone each of these losses is considered
separately and combined with the other losses. For example, predicting the
total pressure drop through the cyclone using the models of Jacob et al.
(1979), Mason et al. (1983) and Rhodes (1998) is achieved by summing the
total pressure drop due to exit expansion loss, the flow straightener, the
cyclone body, the inlet expansion loss, and the entry friction loss.

The pressure drop across the cyclone is caused by the area changes, the
wall friction, change of the flow direction and the dissipation in the vortex
finder (outlet duct). Because of these effects, the cyclone pressure drop is
mainly composed of the following parts as shown in Figs. 5.12 and 5.13:

Chapter 5

150

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

the entry friction pressure loss (P1-2)


the inlet expansion loss

(P2-3)

the cyclone body loss

(P3-4)

the contraction loss

(P4-5)

the exit duct friction loss

(P5-6)

the flow straightener loss

(P6-7)

the exit expansion loss

(P7atm.)

The entry friction pressure loss (inlet duct)

(P1-2)

In this part the pressure loss is caused by friction gas/surface wall and the
total pressure drop can be determined by the following expression:

g v 2 L
1 = f
2 D

Where

(1)

P1

total pressure drop

(Pa)

Darcy friction factor

total length of cyclone inlet duct

hydraulic diameter of cyclone inlet duct (mm)

velocity of air entering cyclone

(mm)

(m/s)

Chapter 5

151

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

The inlet expansion loss (P2-3)


In this part the pressure loss is caused by the inlet duct area changes. It can
be calculated by:

= C 0 g V 2 2

Where

(2)

dynamic pressure loss in the inlet duct

C0

local loss coefficient, entries (Appendix B)

velocity of air entering cyclone

The cyclone body loss

(Pa)

(m/s)

(P3-4)

The total pressure in this part of the cyclone can be predicted by using the
Shepherd and Lapple (1939) empirical equation:

4 = 1 2 g V 2 N H

Where

N H = 16

(3)

ab
D e2

Note: the models of Jacob and Dhodapker (1979), Mason et al. (1983) and
Rhodes (1998) have used Eq.(3) for predicting total pressure drop. The

Chapter 5

152

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

models of EEUA (1997) and Zenz (1999) have used Eq. (4) to predict the
total pressure drop in the cyclone body:

Where

f D c g N s V 2

(4)

26.2 d hi

Dc

cyclone body diameter

Ns

number of the vortex spirals

dhi

hydraulic diameter

The contraction loss

(inch)

(inch)

(P4-5)

In this part the pressure loss is caused by the reverse flow of the gas at the
point that air leaves the cyclone body through the exit duct. The following
expression has been also used to determine this part of pressure loss:

= C 0 g V 2 2

Where

C0

(5)

local loss coefficient, transition (Appendix B)

The exit duct friction loss (vortex finder)

(P5-6)

In this part the pressure loss is caused by friction gas/surface wall and the
total pressure drop can be determined by using Eq.(1).

Chapter 5

153

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

The flow straightener loss

(P6-7)

The pressure loss in this part is caused when the spinning air stream passes
throughout the straightener. It can be determined as follows:

6 = K g V 2 2
Where

(6)

flow straightener loss coefficient (determined by


experiment).

The exit expansion loss

(P7atm)

In this part the pressure loss is caused by the expansion of air leaving the
vortex finder. It can be calculated by using Eq. (2):

Atm.
7
6
1

2 3
5
4

Figure 5.12 Pressure losses at different


parts in a cyclone (directed to atmosphere)

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

154

The total pressure (TP) is determined by summing the seven individual


losses associated with the cyclone as following:

TP7 = 0

(7)

TP6 = TP7 + P7

(8)

TP5 = TP6 + P5 + P6

(9)

TP4 = TP5 + P3

(10)

TP3 = TP4 + P4

(11)

TP2 = TP3 + P2

(12)

TP1 = TP2 + P1

(13)

Expanding the above method to other cyclone configurations, Fig. 5.13, the
following general equations can be derived for cyclone total pressure drop
(TPD):

TPD = TP1-TP7

(direct to atmosphere)

(14)

(105mm I.D. connect to filter)

(15)

(130mm I.D. connect to filter)

(16)

or
TPD = TP1-TP9
or
TPD = TP1-TP10

Chapter 5

155

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

1
8

6
1

3
1
5
4

(a) 130mm I.D. vortex finder

2 3

5
4

(b) 105mm I.D. vortex finder

Figure 5.13 Pressure losses at different parts in a cyclone - connected to a


filter

The total pressure drop predictions of Jacob and Dhodapker (1979),


Rhodes (1998), Mason et al. (1983) and Stairmand (1949) are plotted
against air velocity and agree well with each other, as shown in Figs. 5.14
and 5.15.

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

Please see print copy for Figure 5.14

Figure 5.14 Total pressure drop due to dissipative losses in cyclone


separator (air discharged to atmosphere)

156

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

Please see print copy for Figure 5.15

Figure 5.15 Total pressure drop due to dissipative losses in cyclone


separator (air discharged through a filter)

157

Chapter 5

5.4

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

158

Maximum Solids Flow Capacity

Cyclone separators have been available for pneumatic conveying for many
years. The determination of the ability of this kind of mechanical
equipment to separate solids from a gas stream and the parameters that
affect the operating system and pressure drop should take priority in
designing or selecting any type of gas cleaning devices. Cyclone collection
efficiency can be quite high for coarse particles (e.g. larger than 1000m),
and the cyclone can operate at very high dust loadings, but to date no
research or design information exists on cyclone solids capacity limitations.
Industrial applications that require such information include high-capacity
pneumatic conveying (e.g. dense-phase) and the pneumatic loading and
unloading of road tankers and shipping containers. For this reason an
extensive test program has been undertaken to investigate and evaluate the
maximum mass flow rate of solids for a cyclone collection system.
Different bulk solids and air flows have been tested, and different
cyclone/hopper arrangements have been used for this purpose. The
experimental results obtained have also been compared with four
theoretical models of Beverloo et al. (1961), Brown (1961), Zenz (1962)
and Johanson (1965), using a Microsoft Excel Spreadsheet. This has
resulted in a recommended model and strategy for the design of such high
capacity cyclone separator system.

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

159

5.4.1 Experimental Scheme

The general layout of the cyclone test rig used in this study consists of a
compressor and dryer, as well as pressure, temperature and flow measuring
instruments, and a data acquisition system, see Fig 4.11. A blow tank
mounted underneath a feed bin supplies the material to be conveyed and
collected. For each test, the data acquisition system monitored the solids
mass flow rate via the load cells. The experimental plots show a steady and
linear increase in the mass of solids entering the receiving hopper, and a
steady and linear decrease for the blow tank see Fig.4.16a. Also, the test
cyclone in this work was provided with two observation windows, which
were installed on the top of the cyclone (i.e. in the roof).

5.4.2 Capacity Limitation

The materials were conveyed from the feeding hopper and blow tank to the
cyclone separator, where particle separation took place under ambient
conditions and for a maximum total conveying air flow rate of 0.25 kg/s.
The maximum flow capacity of the cyclone (before choking) was measured
for each product (by slowly increasing feed rate for a given air flow).
Table 5.1 shows the results measured for the maximum mass capacity of
the cyclone for each product.

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

160

Table 5.1 Maximum solids capacity of cyclone (just before choking)

Average maximum.
Material
capacity (kg/s)
Corn

3.8

Rape seed

3.1

Plastic pellets

2.9

Fig. 5.16 is a photograph of the cyclone test rig. The test rig consists of a
cyclone with tangential inlet duct that was used for the testing. 105 mm
I.D. inlet and outlet pipes were used, to gain understanding of the
separation mechanism for the product inside the cyclone, and also to
confirm that the particles are entering the cyclone body in a uniform stream
without bouncing. Furthermore, to observe the saturation point that the
product reached inside the cyclone body separator (at the product outlet),
two observation windows were designed and constructed on the top of the
test cyclone to allow visualization of the flow patterns, as shown in Fig.
5.17. Also, a computer system was used to collect and analyse the
experimental data, see Fig. 4.15.

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

Please see print copy for Figure 5.16

Figure 5.16 General system of cyclone test rig

Please see print copy for Figure 5.17

Figure 5.17 Cyclone separator with two observation windows

161

Chapter 5

5.5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

162

Modeling of Gravity Flow Discharge

Many investigations have been carried out with the aim of predicting the
gravity flow rate of solids discharging from hoppers and bins. The mass
flow rate of solids discharging from a hopper depends on the geometrical
parameters of the vessel, such as the hopper angle and outlet size, the
characteristics of the particles, such as particle size, particle size
distribution, moisture content and particle shape, as well as the wall
friction, Stepanoff (1969) and Gu (1991).

Tests were carried out on the laboratory to estimate the maximum gravity
flow discharge rate of particulate solids and the actual particle velocity
from the test cyclone for three different products (plastic pellets, corn and
rape seed). There have been many previous studies to predict and develop
the general flow rates experimentally and theoretically for granular solids
using different models, such as Beverloo et al. (1961), Brown (1961), Zenz
(1962), Johanson (1965), Brown et al. (1970), and Gu (1991).
Beverloo et al. (1961) presented an equation to estimate the flow rate:

mS = C d g0.5 (D0 - kdp )2.5

(17)

Chapter 5

163

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

where D0 is the hopper outlet diameter, C and k are constants, with C


between 0.575 and 0.595, and K is about 1.4.
Brown (1961) also presented the following maximum bulk solids flow rate
equation for conical hoppers:

b(B kdp )2.5 g 1 cos 0.5


ms =

3
4
2 sin

(18)

Johanson (1965) developed a model for predicting the maximum discharge


rate for free-flowing granular materials from hoppers and bins:

m S = b A 0 Vm

gB
Vm =
4 tan

where

(19)

0.5

(20)

A0

outlet cross sectional area

(m2)

Vm

average velocity of the particles

(m/s)

half angle of the cyclone cone

()

outlet dimension

(m)

acceleration due to gravity

(m/s2)

Chapter 5

164

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

ms

mass flow rate of solids

(kg/s)

bulk density of solids

(kg/m3)

From Eqs. (19) and (20), it can be seen that as the angle increases, the
mass flow rate decreases, see Fig 5.18.
The experimental results of Johanson (1965) also support this trend for a
range of bulk materials. Zenz (1962) also presented an equation for the
solids flow rates from bins, but with no consideration of the hopper cone
angle:

1
mS =
tan d
where

0.5
b B

d = drained angle of repose.

(21)

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

165

Please see print copy for Figure 5.18

Figure 5.18 Influence of cone angle on cyclone mass flow rate for plastic
pellets

In this investigation two types of gravity mass flow rate configuration were
considered: choked flow using the test cyclone with = 11 and an outlet
diameter of 82 mm (see Fig.5.19); and three different flow conditions using
a cone (see Fig. 5.20). To investigate the suggestions/concepts of Zenz
(1975), this cone ( =30, outlet diameter = 73mm) was used to achieve
three conditions: part flow (before choking), full-bore flow (almost
choking) and choking flow.

Chapter 5

166

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

The relevant bulk material properties utilized in this section of the thesis
are presented in Table 5.2. It is to be noted that all test results are for
different particle sizes (1.6mm, 4.3mm and 7.4mm). For more details of the
test procedures and results see Appendix H.

Note: the values of w are applicable to normal stresses greater than 5 kPa.

Table 5.2 Bulk Material Properties

Material

w []

[]

Particle size
d(mm)

M.S.

S.S

Corn

40

8.5

10.5

7.4

Rape Seed

33

1.6

Plastic Pellets

41

10

11

4.3

Where

effective angle of internal friction

wall friction angle

M.S. =

mild steel sheet

S.S

stainless steel sheet (304-2B)

Chapter 5

167

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

Cyclone
Plastic
Pellets

Video
Camera
Container

Load Cell

Figure 5.19 Cyclone gravity flow discharge

Cyclon

Plastic
Pellets

Valve
Control
Con
Video
Camera

Load
cell
(a)

(b)

(c)

Figure 5.20 Different gravity flow conditions using a conical hopper


( =30)

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

168

Also, it should be noted that the values of and w are functions of the
major consolidation stress. The values presented in Table 5.2, are based on
values of major consolidation stress that illustrate the flow stress conditions
in the cyclone model.

5.5.1 Type of Flows and Velocity Variations

There are two basic types of flow of bulk materials in silos: mass flow and
funnel flow. These flows are affected by the properties of the bulk material
and the geometry of the silos, see Fig. 5.21. Furthermore, the flow rate for
discharge of the bulk material in these two types is different.

Please see print copy for Figure 5.21

Figure 5.21 Mass flow limits for Axi-Symmetric and Plane Flow Silo,
Craig (1996) and Wypych (2005)

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

169

Based on the information, it appears that the limits for mass flow depend
on the half angle , the wall friction angle w , and the effective angle of
internal friction . On the other hand, funnel flow bins are characterised
by their squat hopper proportions or by their flat bottoms, see Fig. 5.21.
The flow type obtained during thee experimental works presented in this
section and in the results shown in Table 5.2 was mass-flow.

5.5.2

Average Velocity of Bulk Material in A Mass-Flow Hopper

It should be noted that to compute the mass - flow rate of solids from a
hopper, the type of the flow and the average velocity of the material
discharging from that hopper should be taken into account. The type of
flow that has been considered in this study is axially symmetric flow in a
conical hopper. Figs. 5.22 and 5.23 show the velocity profile of a bulk
material flowing through a hopper and displacement of the material against
time.

In an earlier study, Bosely et al. (1969) investigated the effect of hopper


geometry, particle size, and particle density on velocity profiles in hopper
discharge. They tried to avoid the effect of air pressure gradients by using
coarse particles (1 to 2.5 mm). The shape of the hopper was found to be the

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

170

dominant influence on the velocity profiles. They also observed the effect
of wall friction on the maximum velocity.
Johanson et al. (1962) made a detailed study of the stress and the velocity
for coarse bulk materials in converging channels. Craig (1996) considered
the accelerated flow of the bulk material in the outlet opening of the hopper
as shown in Fig. 5.22. Craig (1996) also asserted, following the work of
Johanson (1965), that the average vertical acceleration, a, can be expressed
as:

ff

a = g 1
ff
a

where

(22)

ff = critical flow factor based on minimum opening dimension


ffa = actual flow factor based on actual opening dimension

the average vertical acceleration of the material, a, at the discharge


opening, can be considered as two parts:

a = ac + av

(23)

Chapter 5

where

ac = bulk material acceleration in hopper due to convergence of


the channel

av

171

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

(m/s2)

= bulk material acceleration in the hopper due to increase in the


velocity at the hopper outlet after the discharge (m/s2)

The value of ac can be obtained as a function of the material velocity at the


opening, by applying the condition of continuity of the flowing material
without significant density change. In other words, the average vertical
velocity vav (zh), is inversely proportional to the cross sectional area A of
the channel so that:

v av ( z h ) =

Cons tan t
A

(24)

where: A is a function of (zh)


Then ac can be written in terms of velocity as:

dv (z ) dA(z h )
dv (z )
=
a c = Vav (z h ) av h = Vav (z h ) av h
dA(z h ) dz h
dz h

Vav (z h ) 2 dA(z h )
(25)
A(z h )
dz h

From a geometrical consideration of axi-symmetric and plane flow


hoppers, the relationship in general is:

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

where :

1 dA(z h ) (1 + m) tan
=
A(z h ) dz h
y

y=

172

(26)

half width of plane flow hopper, or, radius of


the conical hopper at outlet = B/2

Combining equation (24) and (25):

2 vav (zh )2 tan (1+ m)


ac =
B

(27)

Solving for av from Eqs. (22) and (27) results in:

ff 2vav(zh )2 tan (1+ m)


av = g 1
B
ffa

(28)

As the exit velocity increases, av decreases until the average steady state
velocity v av ss (z h = h h ) is reached and av becomes zero (av = 0). This
average exit velocity is given by:

Vm = Vav ss (z h = h h ) = (1

gB
ff
)
ffa 2 tan (1 + m)

(29)

Chapter 5

173

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

Eq. (29) can be used to predict the average exit velocity of cohesive bulk
materials at hopper outlets (via the inclusion of the flow factor ff).
Another method to predict the average steady-state velocity of a bulk
material at the hopper outlet is to use the correlation suggested by Beverloo
et al. (1961). Beverloo stated that the average flow rate, Q, of a free
flowing seed material from a conical mass-flow hopper was

Q = C b
where

g (B - kd p ) 2.5

(30)

C = 0.575 0.595
k = particle shape constant with an average value of 1.4

By applying the principle of continuity to Eq. (30), leads to the average


vertical velocity at the hopper outlet for a conical hopper

Vm = Vav ss (z h = h h )

4C g (B kd p ) 2.5
B2

(31)

Now, it should be noted that in this work it was found that the cyclone
mass flow rates achieved were steady and constant, and the height of the
material in the cyclone cone did not affect the flow rate.

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

174

Eq. (29) can be rewritten for the discharge of free-flowing material,


(

ff
= 0 ) from a conical hopper outlet as:
ff a

gB
vm =
4 tan

0 .5

(at hopper outlet)

(32)

and the mass of solids flow rate, ms, can be calculated from the following
general expression by using the steady-state velocity vm:

ms = b A 0 Vm

(33)

Please see print copy for Figure 5.22

Figure 5.22 Velocity profile of bulk material in hopper, Craig (1996)

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

175

Please see print copy for Figure 5.23

Figure 5.23 Displacement of the material in hopper, Craig (1996)

5.5.3 Experimental Scheme

Figs. 5.19 and 5.20 show the choked gravity flow discharge set-up from the
cyclone itself and different gravity flow conditions from a separate hopper.
The latter was used to explore and test some of the flow rate concepts
suggested by Zenz (1975). A Phantom High Speed Digital Video Camera
and Phantom Vision Software were used for particle analysis.

Chapter 5

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

176

5.5.4 Particle Velocity Analysis

The slow motion feature of the High Speed Video Camera was used to
analyze and measure the particle velocity. The time interval per frame for
each video was 1/1000 sec; the ruler was fixed on the glass pipes to be
photographed at the same time. Thus, the velocity of the particles could be
calculated from measuring the longitudinal length of the displacement on
the video per time, while the mean and the standard deviation (STDEV) for
each particle was also calculated, Utts (2004). It should be noted that 2
tests and 15 frames were recorded for each product (although sometimes 3
tests were recorded if needed, due to scatter). The measurement steps have
been discussed in detail in section 4.6.1, Chapter 4 and Appendix F.

5.5.5 Test Results

The results obtained from the cyclone choked flow and three different flow
condition tests are shown in Tables 5.3 and 5.4. These results were used to
compare/validate the previously considered theoretical models. The
maximum gravity flow discharge was compared with four theoretical
models of Beverloo et al. (1961), Brown (1961), Zenz (1962) and Johanson
(1965). Chapter 6 indicates the comparison in detail.

Chapter 5

177

Experimental Investigations into Cyclone


Pressure Drop and Solids Flow Capacity

Table 5.3 Maximum gravity discharge rate using cyclone (choked-flow)

Plastic Pellets

Max. gravity
discharge rate
(kg/s)
2.8

Av. Particle
Velocity
(m/s)
1.42

Corn

4.4

1.43

Rape Seed

4.4

1.54

11

Material

Av.STDEV.
%
1

Table 5.4 Maximum gravity discharge rate for different flow conditions
through hopper (Fig. 5.20)

Material

Type of
Flow

Part flow
Plastic
Pellets

Full-bore
flow
choked flow
Part flow

Corn

Full-bore
flow
Choked
flow
Part flow

Rape
seed

Full-bore
flow
Choked
flow

Av.
Gravity
Particle
Discharge Rate
Velocity
(kg/s)
(m/s)
0.42

2.04

2.27

1.95

1.5

0.66

0.93

2.16

3.47

2.11

2.12

0.68

0.78

2.18

3.11

2.15

2.55

0.82

Av.
STDEV
%
4
1
3
1
1
7
5
3
2

Comments

Before
choking
Just before
choking
Choked
Before
choking
Just before
choking
Choked
Before
choking
Just before
choking
Choked

CHAPTER 6
COMPARISON WITH THEORY OF EXPERIMENTAL
DATA ON PRESSURE DROP AND SOLIDS CAPACITY

6.1

Introduction

6.1.1 Pressure Drop

The experimental results obtained in this study were compared with


improved existing theories in order to put the data to better use. The
measured values of the cyclone pressure drop in the present study were
compared with pressure drop predictions from a number of different
models. This effort has resulted in a new procedure to predict cyclone
pressure drop within certain limits.

6.1.1.1

Comparison of 105mm I.D and 130mm I.D. vortex finders

An overview of the experimental results on the cyclone pressure drop for


two vortex finder configurations is shown in Figs. 6.1 and 6.2. As
expected, the figures show that the pressure drop becomes larger as the
inlet velocity is increased. In this study the De/D ratios (of the diameter of
the vortex finder to the diameter of the cyclone body) were 0.2625 and
0.325 for the 105mm and 130mm vortex finder diameters, respectively.
178

Chapter 6

179

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

The pressure drop when using 0.325 was significantly lower than for

Total Pressure Difference, TPD (Pa)

0.2625.
10000
9000
8000

105mm I.D

7000

130mm I.D

6000
5000
4000
3000
2000
1000
0

10

15

20

25

30

Intlet Velocity, Vi (m/s)

Figure 6.1 Comparison of the 105mm I.D and 130mm I.D. vortex finders
(air discharged to atmosphere)

Total Pressure Difference,TPD (Pa)

7000
6000

105mm I.D.

5000

130mm I.D.

4000
3000
2000
1000
0
0

10

15

20

25

30

Inlet Velocity,Vi (m/s)

Figure 6.2 Comparison of the 105mm I.D and 130mm I.D. vortex finder
(connected to a filter)

Chapter 6

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

180

The reason is that a decreased vortex-finder diameter contributes to a


larger pressure drop for the same inlet velocity. The 105mm I.D. and
130mm I.D vortex finders have quite similar pressure drops at low
velocities, but the pressure drop using the 105mm I.D device is
significantly higher than that of the 130mm I.D. at high velocities, see Figs.
6.1 and 6.2. Consequently, the vortex finder diameter does significantly
affect the overall cyclone pressure drop for a given inlet velocity.

6.1.1.2

Comparison of Experimental Data and Existing Models

The pressure drop values of five models were compared with the
experimental results obtained in this study for different inlet velocities and
two different sizes of vortex finder. The pressure drop has been plotted
against air velocity for different sizes of vortex finder, as shown in Figs.
6.3 and 6.4. From the results provided in Figs. 6.1, 6.2, 6.3, and 6.4 it can
be seen that:
All model predictions are significantly lower than the experimental
values;
The Jacob et al. (1979) and Mason et al. (1983) methods agree well
with each other and are the closest to the experimentally determined
curves;
Some models provide surprisingly low values of pressure drop.

Chapter 6

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

181

Please see print copy for Figure 6.3

Figure 6.3 Comparison of experimental data and existing models for a


vortex finder discharging to atmosphere

Chapter 6

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

182

Please see print copy for Figure 6.4

Figure 6.4 Comparison of experimental data and existing models of a


vortex finder connected to a filter

Chapter 6

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

183

The effect of the cyclone vortex finder on the cyclone pressure drop was
investigated. Two different diameters of vortex finder were used for a wide
range of entrance velocity. Furthermore, two cyclone configurations were
used: air discharging directly to atmosphere; and air discharging through a
pipe connected to a filter. The pressure drop for a 0.325 ratio of the
diameter of the vortex finder to the cyclone body diameter was lower than
for De/D = 0.2625. The predictions of the models showed significant
variations and differences compared with the experimental results. The
Jacob et al. (1979) and Mason et al. (1983) models, as mentioned
previously, predicted similar values and were closest to the experimental
data.
Possible reasons behind the variations between the experimentally
measured values of the pressure drop and the values predicted from the
models include the assumptions that have been used by the models and
perhaps some error in the experimental operations.

6.1.1.3

Comparison of Experimental Data with Pressure Drop Model


Based On Dissipative Losses

Theoretically, there are many factors that combine to predict the total
pressure drop across the cyclone separator, such as, local losses at the inlet,

Chapter 6

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

184

losses at the outlet, and losses within the cyclone body. Therefore, the total
pressure loss (P) is the sum of these losses.
A new theoretical model for pressure drop prediction across the cyclone
was presented based on the consideration of the dissipative loss of flow in
the cyclone system. The values of pressure drop predicted by this model
are compared with experimental data and other various models: Stairmand
(1949), Barth (1956), Jacob and Dhodapkar (1979), Mason et al. (1983),
EEUA (1997), Rhodes (1998), and Zenz (1999). The models of Barth and
Stairmand, which are based on considerations of physical phenomena,
predict an increase in the pressure drop with an increase in the inlet
velocity for both sets of experimental configurations, see Figs. 6.5 and 6.6.

Now, it should be noted that two different sizes of vortex finder (gas exit
diameters) were used for this modeling of pressure drop. In this section the
experimental results obtained were compared with existing model
predictions for the two types of cyclone configuration. From the results
provided in Figs. 6.5 and 6.6, it can be seen that:

(a)

some of the models such as EEUA (1997), and Zenz (1999) provide
predictions significantly lower than the experimental values;

Chapter 6

(b)

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

185

some of the model such as Jacob and Dhodapkar (1979) and


Stairmand (1949) provide prediction slightly lower than the
experimental values;

(c)

The model of Barth (k1, k2) predicts slightly higher values than the
experimental results.

It has been observed also that the predicted values of the total pressure drop
decrease when the conversion from the 105mm I.D. to the 130mm I.D.
vortex finder takes place.

Now to elucidate the excessive of Barths pressure drop predictions


compared with the experimental values. Barth estimated the pressure drop
in the vortex finder using Eq. (8), Chapter 3, and predict very high pressure
drop, as shown in Figs. 6.5 and 6.6. More recently, Hoffmann et al. (2002)
reported that the model of Barth predicts high pressure drop in the vortex
finder.

From the above, it appears that the Barth (k1) model and the models of
Jacob and Dhodapkar (1979) and Stairmand (1949) can be used
collectively to predict a possible envelope of cyclone pressure drop (i.e.
possible upper and lower limits). This is the new procedure recommended
for cyclone design.

Chapter 6

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

186

Please see print copy for Figure 6.5

Figure 6.5 Comparison of experimental data and theoretical models of


vortex finder (discharging to atmosphere)

Chapter 6

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

187

Please see print copy for Figure 6.6

Figure 6.6 Comparison of experimental data and theoretical models with


vortex finder connected to filter.

Chapter 6

6.2

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

188

Maximum Solids Capacity and Gravity Flow Discharge

The results obtained from the cyclone chocked flow and the tests for three
different flow conditions have been summarized in Chapter 5, see Tables
5.3 and 5.4. These results are used to compare/validate the previously
considered theoretical models. The maximum capacity cyclone tests
(before choking) and maximum gravity flow discharge were compared
with four theoretical models: Beverloo et al. (1961); Brown (1961); Zenz
(1962); and Johanson (1965).
By comparing the measured and predicted values in Tables 6.1 and 6.3, it
can be seen that:
the models of Beverloo, Brown, and Zenz predict flow rates much
lower than the gravity experimental values (choking and before
choking-flow).
the model of Zenz predicts flow rates much lower than the
experimental values of the cyclone separator (just before choking).
the predictions of Johanson are generally in much better agreement
with the experimental results.

For the gravity flow rate discharge work (cyclone choked-flow), the
differences between the mass flow predictions of Beverloo et al. (1961);

Chapter 6

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

189

Brown (1961); Johanson (1965); and Zenz (1962) and the experimental
results are considered in more detail, as follows:
The differences between the mass flow predictions of Beverloo et al.
(1961) and the experimental results were -45% to -59%.
The differences between the mass flow predictions of Brown (1961)
model and the experimental results were -15% to -35%.

For the Zenz (1962) model, the differences between the mass flow
predictions and the experimental results were -32% to -43%.

For the Johanson (1965) model, the differences are much lower: for
Plastic pellets product about +2 %, for the Corn product about -10
%, and for the Rape seed about -18 %.

For the maximum capacity cyclone tests (just before choking), the
variations between the values predicted by Zenz and the experimental
results for the three different products were significant (viz. -11% to -35%).
However, the Johanson (1965) model again provided reasonable results,
with variation of about -3%, +4% and +17% for plastic pellets, corn and
rape seed, respectively. Therefore, it was found from the work and results
presented that the flowrates computed by the Johanson model provided a
good agreement with the experimental results for the three different sized
materials.

Chapter 6

190

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

Table 6.1 Maximum capacity of cyclone separator (just before choking)

Material

Max. capacity
(kg/s)

Zenz

diff.
(%)

Plastic pellets

2.9

1.91

-35

Corn

3.8

2.53

-33

Rape seed

3.1

2.76

-11

Table 6.2 Maximum gravity discharge rate using cyclone (choked-flow)

Please see print copy for Table 6.2

Chapter 6

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

191

Table 6.3 Experimental and predicted values of maximum gravity mass


flowrate of solids

Please see print copy for Table 6.3

Chapter 6

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

192

It should be noted that the initial experimental work (e.g. Table 5.1) on
cyclone maximum capacity (just before choking) was carried out on three
different products (plastic pellets, corn, and rape seed) with certain physical
properties. Different results were obtained for each product using different
air flows.

Due to the deterioration of the organic products (corn and rape seed) and
the necessity of further experimentation on the choked gravity flow
discharge from the cyclone separator and the other gravity flow rate
conditions (e.g. measurement of particle velocity), new organic materials
had to be used. The three different materials used in this work were
updated and their physical properties were found to be quite different.
Tables 6.4 and 6.5 illustrate the physical properties of the old and new
materials tested.

Significant changes in results were obtained on the maximum capacity of


the cyclone separator (just before choking) using the old materials, and the
maximum gravity discharge rates using the cyclone (choked-flow), see
Table 6.6. For example, for a given material, the maximum gravity
discharge rate (choked-flow) of the cyclone should be less than the
maximum solids capacity (just before choking)

Chapter 6

193

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

Table 6.4 Physical properties of the old materials tested.


Material

Solid Density

Bulk Density

Median Particle

(kg/m3)

(kg/m3)

Diameter (mm)

Plastic Pellets

921

531

4.3

Rape Seed

1140

621

1.6

Corn

1380

670

7.4

Table 6.5 Physical properties of the new materials tested


Material

Solid Density

Bulk Density

Median Particle

(kg/m3)

(kg/m3)

Diameter (mm)

Plastic Pellets

921

531

4.3

Rape Seed

1152

674

1.9

Corn

1337

734

Table 6.6 Maximum solids capacity (just before choking) and maximum
gravity discharge rate (choked-flow) of cyclone.

Material

Max. capacity
(kg/s)
(Old material)

Max. gravity
(kg/s)
(New material)

Difference
%

Plastic Pellets

2.95

2.76

Rape Seed

3.8

4.4

16

Corn

3.1

4.4

42

Chapter 6

Comparison with Theory of Experimental Data


On Pressure Drop and Solids Capacity

194

For corn and rape seed (canola), Table 6.6 shows that the maximum gravity
flow (choked) discharge values are higher that the maximum values of
cyclone solids capacity. The main reason for the discrepancy in trend was
the unexpected variation in material properties (i.e. from the old to the new
material), as shown in Tables 6.4 and 6.5.

CHAPTER 7
DISCUSSION

7.1

Introduction

In the present study, experimental and prediction results are presented for
pressure drop (air discharged directly to atmosphere; air discharged
through a pipe connected to a filter) for different inlet velocities and two
different sizes of vortex finder. Also it summarises investigations into
maximum solids capacity and gravity flow discharge (maximum solids
flow without choking; choked gravity flow discharge from cyclone; and
three different gravity flow conditions from a hopper) of cyclone separators
for three different products (plastic pellets, corn and rape seed) and for a
maximum total conveying air flow rate of 0.25 kg/s.

7.2

Cyclone Pressure Drop

Figs. 5.6, 5.7, 5.8, and 5.9 show the experimental results on the cyclone
pressure drop plotted against inlet air velocity for two different vortex
finder configurations. According to the trends of the experimental data, the
figures show (as expected) that the pressure drop becomes larger as the
inlet velocity is increased. The ratios of the diameter of the vortex finder to
195

Chapter 7

Discussion

196

the diameter of the cyclone body (De/D) were 0.2625 and 0.325 for the
105mm and 130mm vortex finder diameters, respectively. The pressure
drop when using 0.325 was significantly lower than for 0.2625. The reason
for this is a decreased vortex-finder diameter contributes to a larger
pressure drop for the same inlet velocity. The 105mm I.D. and 130mm I.D
vortex finders have quite similar pressure drops at low velocities, but the
pressure drop using the 105mm I.D device is significantly higher than that
of the 130mm I.D. at high velocities, see Figs. 6.1 and 6.2. Consequently,
the vortex finder diameter does significantly affect the overall cyclone
pressure drop for a given inlet velocity (even though some pressure drop
models are not affected by vortex finder diameter changes).
The pressure drop over the cyclone separator was measured with a flow
straightener inserted inside the vortex finder (straight flow). The influence
of the flow straightener on the flow field and vortex finder pressure drop
was observed. Two test model configurations were used: measuring
pressure difference both with and without the flow straightener. Fig. 5.3
shows the pressure difference as a function of the velocity. As expected,
the curves for the two different sizes of flow straightener were quite
different. Furthermore, the pressure drop variation curve for the 130mm
I.D. flow straightener is lower than for the 105mm I.D. straightener, which
can be seen also from Fig. 5.3

Chapter 7

7.3

Discussion

197

Pressure Drop Prediction

Figs. 5.10 and 5.11 show the corresponding variation of the pressure drop
prediction of five models. The predicted pressure drop is found to increase
with increase air velocity (as expected). The predictions of the models for
two different sizes of vortex finder and two cyclone configurations (air
discharging directly to atmosphere; and air discharging through a pipe
connected to a filter) showed significant variations and differences
compared with the experimental results, see Figs. 6.3 and 6.4. The Jacob et
al. (1979) and Mason et al. (1983) models predicted similar values and
were closest to the experimental data, whereas some other models provided
surprisingly low values of pressure drop. Possible reasons behind the
variations between the experimentally measured values of the pressure
drop and the values predicted from the models include the assumptions that
have been used by the models and perhaps some error in the experimental
operations. However, by repeating tests and checking calibrations
continuously, the latter effect is expected to be minor.

7.4

New Model of Pressure Drop

New theoretical model for pressure drop prediction across the cyclone is
presented based on the consideration of the dissipative loss of flow in the

Chapter 7

Discussion

198

cyclone system. Two different sizes of vortex finder (gas exit diameters)
were used for this modeling of pressure drop. The models of Stairmand
(1949), Jacob et al. (1979), Mason et al. (1983), Rhodes (1998), EEUA
(1987) and Zenz (1999) predicted a significant lower pressure drop than the
experimental values. The model of Barth, with two values of k1 and k2 for
rounded and sharp edges, respectively, predicted significantly higher values
than the experimental data, see Figs. 6.5 and 6.6.

As explained in Chapter 6, the estimation of the pressure drop in the vortex


finder using Barth (1956) theory was the reason behind the excessive
pressure drop predictions compared with the experimental values. In other
word, the Barth (1956) model for vortex finder loss in quite conservative
and inaccurate (for the cyclone conditions considered in this study).

7.5

Maximum Solids Capacity and Gravity Flow Discharge

The initial experimental work on cyclone maximum capacity (just before


choking) was carried out on three products (plastic pellets, corn, and rape
seed) with certain physical properties. Different results were obtained for
each product using different air flows.

Due to the deterioration of the organic products (corn and rape seed) and
the necessity of further experimentation on the choked gravity flow

Chapter 7

Discussion

199

discharge from the cyclone separator and the other gravity flow rate
conditions (e.g. measurement of particle velocity), new organic materials
had to be used. The three different materials used in this work were
updated and their physical properties were found to be quite different.
Tables 6.4 and 6.5 illustrate the physical properties of the old and new
material tested.

A significant change in results was obtained for the maximum capacity of


the cyclone separator (just before choking) using the old materials, and the
maximum gravity discharge rates using the cyclone (choked-flow), see
Table 6.6. For example, for a given material, the maximum gravity
discharge rate (choked-flow) of the cyclone should be less than the
maximum solids capacity (just before choking).

Table 6.6 shows that the maximum gravity flow discharge values are
higher than the maximum values of cyclone solids capacity for corn and
rape seed materials. The main reason for the discrepancy in trend was the
unexpected variation in material properties (i.e. from the old to the new
material).

CHAPTER 8
CONCLUSION AND FUTURE WORK

8.1

Conclusion

8.1.1 Introduction

It was evident in the study carried out in this thesis that cyclone separators
are used extensively in process industries particularly in the design of
pneumatic conveying systems as gas cleaning devices or as gas-solids
separators. Due to their versatility such as low capital and maintenance
costs and simplicity of design and absence of moving parts, cyclones are
ideal for use as precleaners for more expensive final control devices such
as baghouses or electrostatic precipitators.

The performance of the cyclone separator is determined by two important


parameters, the collection efficiency and the pressure drop. Many
individual characteristics affect cyclone performance, such as the
influence of different operating conditions and the variation in the
geometry of different cyclone parts as well as the materials to be separated.
Experimental testing and practical knowledge are still very useful tools in

200

Chapter 8

Conclusion and Future Work

201

estimating/confirming the performance and design process of cyclone


separators.

8.1.2 Pressure Drop

The experimental work investigations carried out in this study have proved
that an increase in inlet flow rate rapidly increases the cyclone pressure
drop. The investigations were aimed at studying and controlling the
instability of vortex flow, as well as examining the effect of outlet piping
(vortex finder) on cyclone performance. The experiments were conducted
by inserting varying sizes of flow straightener and vortex finder from
105mm I.D. to 130mm I.D. Aironly experimental investigations into
cyclone pressure drop have been carried out. The larger vortex finder
resulted in a significant fall in pressure drop over the cyclone separator.
Consequently, the vortex finder diameter does significantly affect the
overall cyclone pressure drop for a given inlet velocity.

As the tests were carried out with varying outlet geometry and downstream
connections (i.e. discharging to atmosphere or via a pipe and filter),
predictions of pressure drop over the test cyclone were made using various
models for these conditions. The actual flow rates and dimensions of the
test cyclone were used as data inputs for the models equations.

Chapter 8

Conclusion and Future Work

202

The measured values of the cyclone pressure drop were compared with
pressure drop predictions from a number of different models.

From the results provided in this study it can be seen that:

Many of the model predictions are significantly lower than the


experimental values;
The Jacob and Dhodapkar. (1979) and Mason et al. (1983) methods
agree well with each other and are the closest to (but consistently
lower than) the experimentally determined curves;

Also, the results of the new theoretical model for pressure drop prediction
across the cyclone based on the consideration of the dissipative loss of flow
in the cyclone system and the experimental data and other various models
can be summarized, as follows:

(a)

the models of EEUA (1997), and Zenz (1999) provide predictions


significantly lower than the experimental values;

(b)

the models of Jacob and Dhodapkar (1979) and Stairmand (1949)


provide prediction slightly lower than the experimental values;

(c)

the model of Barth (1956)(k1, k2) predicts slightly higher values than
the experimental results.

Chapter 8

Conclusion and Future Work

203

8.1.3 Maximum Mass Capacity and Gravity Flow Discharge

Using the blow tank feeder, the materials were conveyed from the feeding
bin to the cyclone separator unit, where the mechanism of particle
separation took place inside the cyclone under ambient conditions or under
certain cyclone operating conditions. The maximum capacity of the
materials being collected by the cyclone system was measured for each
material. This was achieved by increasing the proportion of air flow to the
blow tank feeder. Also, during the time of testing and by the assistance of
glass windows on the roof of the cyclone, the mechanism of dust
separation, building up the solids flow patterns inside the cyclone, and
choking flow

phenomena in the cyclone dust outlet, were visually

observed.

Furthermore, tests were carried out in the laboratory to estimate the


maximum gravity flow discharge rate of particulate solids and the actual
particle velocity from the test cyclone for three different products (plastic
pellets, corn and rape seed). Two types of gravity mass flow rate
configurations were considered: choked flow using the test cyclone with
= 11 and an outlet diameter of 82 mm, and three different flow conditions
using a cone. To investigate the suggestions/concepts of Zenz (1975), this
cone ( =30, outlet diameter = 73mm) was used to achieve three

Chapter 8

Conclusion and Future Work

204

conditions: part flow (before choking), full-bore flow (almost choking)


and choking flow. The results of the maximum gravity flow discharge were
compared with the four theoretical models of Beverloo et al. (1961), Brown
(1961), Zenz (1962) and Johanson (1965). By comparing the measured and
predicted values, it can be seen that:

the models of (Beverloo, Brown, and Zenz) predict flow rates much
lower than the experimental values (choking and before chokingflow).
the model of Zenz predicts flow rates much lower than the
experimental values of the cyclone separator (just before choking).
the predictions of Johanson are generally in much better agreement
with the experimental results.

For the gravity flow rate discharge work (cyclone choked-flow), the
difference between the mass flow predictions of (Beverloo et al., 1961;
Brown, 1961; Johanson, 1965; and Zenz, 1962), and the experimental
results are as follows:

The differences between the mass flow predictions of Beverloo et al.,


(1961) model and the experimental results were -45% to -59%.

Chapter 8

Conclusion and Future Work

205

The differences between the mass flow predictions of Brown (1961)


model and the experimental results were -15% to -35%.

For the Zenz (1962) model, the differences between the mass flow
predictions and the experimental results were -13% to -43%.

For the Johanson (1965) model, the differences are much lower, for
Plastic pellets about +3 %, for the Corn product about -10 % and for
the Rape seed about -18 %.

For the maximum capacity cyclone tests (just before choking), the
variations between the values predicted by Zenz and the experimental
results for the three different products were significant (viz. -11% to -35%).
However, the Johanson (1965) model again provided reasonable results,
with variations of about -3%, +4% and +17% for plastic pellets, corn and
rape seed, respectively. Therefore, it was found from the work and results
presented that the flowrates computed by the Johanson model provided a
good agreement with the experimental results for the three different sized
materials.

8.2

Future Work

More experimental work should be carried out on different cyclone


dimensions:

Chapter 8

Conclusion and Future Work

206

more testing of cyclone solids capacity and pressure drop by


changing the solids outlet geometry(e.g. outlet diameters, cone
angle);
more testing to be carried out with other organic products with
the aim of avoiding product deterioration.

In the present study, high speed camera measurements of particle velocity


were found to be useful (e.g. choking, comparison with velocity predictions
of various models). Additional work is suggested to evaluate and develop
the models further, as described below.

A glass /Parspex cyclone was considered in this study but this was not
pursued because of high cost and difficulty in manufacture. Two
observation windows instead were designed and constructed on the top of
the cyclone. As these widows were found to be restrictive, a glass/perspex
cyclone still should be pursued to study the nature of the flow patterns in
detail and to: capture images using the video camera; gain a visual
perspective the number of vortex spirals; and observe the saturation point
that the product reaches inside the cyclone body separator at the outlet (i.e.
confirmation of maximum capacity and choking).

Chapter 8

Conclusion and Future Work

207

As presented in this thesis, the prediction of maximum gravity discharge


rate utilized a hopper which was suggested by Zenz (1975), (i.e. cone angle
=30, outlet diameter = 73mm). While the presented results tend to agree
well with the theoretical model (e.g. Johanson, 1965), additional work
should be carried out by using various hopper configurations (different
hopper slopes and different hopper outlets). The results from such work
will be very useful in comparing/validating the considered and developed
models.

REFERENCES

ACGIH, (1992). Exhaust System Design Procedure. In: Industrial


Ventilation 21st Edition: A manual of Recommended Practice.
American Conference of Governmental Industrial Hygienists, Inc.,
pp. 5.1-5.54.

ACGIH, (2001). Exhaust System Design Procedure. In: Industrial


Ventilation 24th Edition: A Manual of Recommended Practice.
American Conference of Governmental Industrial Hygienists, Inc.,
pp. 5.1-5.71.

AIRAH, DA3 (1987). Duct Design Manual: No. DA3. Australian Institute
of Refrigeration, Air Conditioning and Heating and the Dept. of
Administrative Services, Construction Group. Aus. Govt.
Publishing Service.

Alden, J. L. (1938). Design of Industrial Exhaust System. Heating and


Ventilating. Vol. 35, pp. 48-53.
Alexander, R. M. (1949). Fundamentals of Cyclone Design and Operation.
Proceedings Australia. Inst. Mining metall. (Inc.). Vol. 152-153, pp.
203-228.

Antonia, G., Luis, M. R., and Cristbal, C. (2001). Cold flow model of a
PFBC cyclone. Powder Technology. Vol. 117, pp. 207-220.

208

References

209

ASHRAE HANDBOOK (1985). Duct Design. In: Duct and Pipe Sizing.
Atlanta, pp. 33.1-33.50.

Avci, A., and Karagoz, I. (2001). Theoretical Investigation of Pressure


Losses in Cyclone Separators. International communications in Heat
and Mass Transfer. Vol. 28, pp. 107- 117.

Avci, A., and Karagoz, I. (2003). Effects of flow and geometrical


Parameters on the Collection Efficiency in Cyclone Separators.
Journal of Aerosol Science. Vol. 34, pp. 937-955.

Barth, W. (1956). Design and Layout of the Cyclone Separator on the Basis
of New Investigations. Brennst Waerme Kraft. Vol. 8, pp. 1- 9.

Beeckmans, J. M., and Morin, B. (1987). The Effect of Particulate Solids


on Pressure Drop across a Cyclone. Powder Technology. Vol. 52, pp.
227-232.

Beverloo, W. A., Leniger, H. A. and Van de Velde, J. (1961). The Flow of


Granular Solids Through Orifices. Chemical Engineering Science.
Vol. 15, pp. 260-269.

Bohnet and Matthias (1995). Influence of Gas Temperature on the


Separation Efficiency of Aerocyclones. Chemical Engineering and
Processing. Vol. 34, pp.151-156.

References

210

Bosley, J., Schofield, C. and Shook, C. A. (1969). An Experimental Study


of Granular Discharge From Model Hopper. Transaction of the
Institution of Chemical Engineers. Vol. 47, pp. T147- T153.

Briggs, L. W. (1946). Effect of Dust Concentration on Cyclone


Performance. Transaction of American Institute of Chemical
Engineers. Vol. 42, pp. 511-526.

Brown, R. L., (1961). Minimum Energy Theorem for Flow of Dry


Granules through Apertures, Nature, Vol. 191, No. 4787, pp. 458461.

Brown, R. L., and Richards, J. C., (1970). Principles of Powder Mechanics.


Pergamon Press, London

Bryent, H. S., Silverman, R. W., and Zenz, F. A. (1983). How Dust in Gas
Affects Cyclone Pressure Drop. Hydrocarbon Processing. Vol. 62,
pp. 87-90.

Buffalo, N. Y. (1949). Flow in Elbows. In: Fan Engineering, An


Engineers Handbook on Air, its Movement and Distribution in Air
Conditioning, Combustion, Conveying and other Applications
Employing Fans. Buffalo Forge Company.

Buttner, H. (1999). Dimensionless Representation of Particle Separation


characteristic of Cyclones. Journal of Aerosol Science. Vol. 30, pp.
1291- 1302.

References

211

Caplan, K. J. (1977). Source Control by Centrifugal Force and Gravity. In:


Air pollution, Vol. 4, 3rd Edition, (edited by Arthur C. Stern).
Academic Press, New York, pp. 97-148.

Casal, J., and Martinez, B. JM. (1983). A Better Way to Calculate Cyclone
Pressure Drop. Chemical Engineering. Vol. 99-100, pp. 99-100.

Chen, W. H., and Chen, J. C. (2001). Combustion characteristics and


energy recovery of a small mass burn incinerator. International
Communications in Heat and Mass Transfer. Vol. 28, pp. 299-310.

Cheremisinoff, N. P., and Cheremisinoff, P. N. (1986). Particle Properties


and Characterization. In: Encyclopedia of Fluid Mechanics. Volume
4, Solids and Gas-Solids Flows, (edited by Cheremisinoff, N. P.).
Gulf Publishing Co., Houston, Texas, pp. 3-36.

Colin, H. (2001). Cyclone dust Collectors: an Underestimated Technology.


Filtration and Separation. December, pp. 20-21.

Comas, M., Comas, J., and Chetrit, C. (1991). Cyclone Pressure Drop and
Efficiency With and Without an Inlet Vane. Powder Technology.
Vol. 66, pp. 143-148.

Craig, D. A. (1996). An Investigation of Blending And Mixing In Bulk


Materials Handling And Process System. Ph.D. Thesis, University
Of Wollongong, Australia.

References

212

De, S., Lal, A. K., and Nag, P. K. (1999). An Experimental Investigation


on Pressure drop and Collection Efficiency of Simple Plate-Type
Impact Separator. Powder Technology. Vol. 106, pp. 192-198.

DeOtte, R. E., Jr. (1990). A Model for the Prediction of the Collection
Efficiency Characteristics of a small, Cylindrical Aerosol Sampling
Cyclone. Aerosol Science and Technology. Vol. 12, pp. 1055-1066.

Dirgo, J. and Leith, D. (1985). Cyclone Collection Efficiency: Comparison


of Experimental Results with Theoretical Predictions. Aerosol
Science and Technology. Vol. 4, pp.401-415.

EEUA (1997). Gas Solids Separations. In: Perrys Chemical Engineers


Handbook, 7th Edition (edited by Robert H. Perry, Don W. Green
and James O. Maloney). McGraw-Hill Company, USA, pp.19-44.

Fassani, F. L., and Goldstein, L. Jr. (2000). A study of the Effect of High
Inlet Solids Loading on a Cyclone Separator Pressure Drop and
Collection Efficiency. Powder Technology. Vol. 107, pp. 60-65.

Gabra, M., Salman, H., and Kjellstrm, B. (1998). Development of a sugar


cane residue feeding system for a cyclone gasifier. Biomass and Bio
energy. Vol. 15, pp. 143-153.

Gautam, M., and Sreenath, A. (1997). Performance of a Respirable MultiInlet Cyclone Sampler. Journal of Aerosol Science. Vol. 28, pp.
1265-1281.

References

213

Gauthier, T. A., Briens, C. L., Bergougnou, M. A., and Galtier, P. (1990).


Uniflow Cyclone Efficiency Study. Powder Technology. Vol. 62, pp.
217-225.

George, A. Davdidson, Jr. (2000). Cyclone Separation, Chemical and


Petroleum Engineering, Unit Operations Laboratories. On-Line Lab
Manual.

Gerchow, F. J. (1986). Dilute (Lean and Medium) Phase Pneumatic


Conveying. In: Encyclopedia of Fluid Mechanics. Volume 4. Solids
and Gas-Solids Flows, (edited by Cheremisinoff, N. P.). Gulf
Publishing Co., Houston, Texas, pp. 421-470.

Gil, A., Corts, C., Romeo, L. M., and Velilla, J. (2002). Gas-Particle Flow
Inside Cyclone Diplegs With Pneumatic Extraction. Powder
Technology. Vol. 4570, pp. 1-14.

Gradon, L., Luckner, H., Podgorski, A., and Wertejuk, Z. (1998).


Separation of Neutral and Charged Aerosol Particles in Cyclones
With External Electric field. Journal of Aerosol Science. Vol. 29, pp.
S927-S928.

Griffiths, A. J., Yazdabadi, p., and Syred, N. (1996). The Use of Centre
Bodies and De-Swirl Vanes in the Exhaust of Cyclone Dust
Separators. Proceeding of the institution of Mechanical Engineers.
Vol. 210, pp. 193-203. ProQuest Science Journals.

References

214

Gu, Z. H. (1991). Gravity Flowrate of Bulk Solids From Mass Flow Bins.
Ph.D. Thesis, University Of Wollongong, Australia.

Harold, F., Paul, Y., Robert, L., and David, F. (1973). Gas-Solid Systems.
In: Chemical Engineers Handbook, 5th Edition, (edited by Perry, R.
H., Chilton, C. H.). McGraw-Hill, New York, pp.1-121.

Hejma, I. J. (1971). Influence of Turbulence on the Separation Process in a


Cyclone. Staub-Reinhalt.Luft. Vol. 31, pp. 22-29.

Hoffmann, A. C., and Stein, L. E. (2002). Gas Cyclones and Swirl Tubes:
Principles, Design and Operation. Springer-Verlag Berlin
Heidelberg, Germany.

Hoffmann, A. C., de Groot, M., Peng, W., Dries, H. W. A., and Kater, J.
(2001). Advantages and Risks in Increasing cyclone Separator
Length. American Institute of Chemical Engineers, AIChE Journal.
Vol. 47, pp. 2452-2460.

Hoffmann, A. C., Jonge, R. D., Arends, H., and Hanrats, C. (1995).


Evidence of the Natural Vortex Length and its effect on the
Separation Efficiency of Gas Cyclones. Filtration and Separation.
Vol. 32, pp. 799-804.

Hoffmann, A. C., Santen, A., Allen, R. W., and Clift, R. (1992). Effects of
Geometry and Solid Loading on the Performance of Gas Cyclones.
Powder Technology. Vol. 70, pp. 83-91.

References

215

Iozia, D. L., and Leith, D. (1989). Effect of Cyclone Dimensions on Gas


Flow Pattern and Collection Efficiency. Aerosol Science and
Technology. Vol. 10, pp. 491-500.

Ito, H. (1960). Pressure Losses in Smooth Pipe Bends. Journal of Basic


Engineering, Transactions of the ASME. Vol. D82, pp. 131-143.

Jacob, K., and Dhodapkar, S. (1979). Gas - Solid Separations. In:


Handbook of Separation Techniques for Chemical Engineers, 3rd
Edition, (edited by Schweitzer P. A.). McGraw-Hill, Inc, NY,
USA, pp. 3-34.

Jens, R., Arwed, S., and Joachim, W. (2000). Find the optimum cyclone
size with respect to the fines in pneumatic conveying systems.
Powder Technology. Vol. 112, pp. 251-255.

Johanson, J. R. (1965). Method of Calculating Rate of Discharge From


Hoppers and Bins. Transaction Society of Mining Engineers AIME,
Vol. 232, pp. 69-80.

Johanson, J. R., and Jenike, A. W. (1962). Stress and Velocity Fields in


Gravity Flow of Bulk Solids. Bulletin 116. Utah Engineering
Experiment Station, University of Utah, Vol. 53, No.21.

Joseph, H. M. (1979). Gas-Solid Separations. In: Handbook of Separation


Techniques for Chemical Engineers, 1st Edition, (edited by
Schweitzer P. A.). McGraw-Hill Book Co., USA, pp. 3-30.

References

216

Kang, S. K., Kwon, T. W., and Kim, S. D. (1989). Hydrodynamic


Characteristics of Cyclone Reactors. Powder Technology. Vol. 58,
pp. 211-220.

Kelsall, D. F. (1952). A study of The Motion of Solid Particles in a


Hydraulic Cyclone. Transaction of institution of Chemical
Engineers. Vol. 30, pp. 87-108.

kim, H. t., Lee, K. W., and Kuhlman, M. R. (2001). Exploratory design


modifications for Enhancing Cyclone Performance. Journal of
Aerosol Science. Vol. 32, pp. 1135-1146.

Kim, J. C., and Lee, K. W. (1990). Experimental Study of Particle


Collection by Small Cyclones. Aerosol Science and Technology.
Vol. 12, pp. 1003-1015.

Lee, K. W., Gieseke, J. A., and Piispanan, W. H. (1985). Evaluation of


Cyclone Performance in Different Gases. Atmospheric environment
(1967). Vol. 19, pp. 847-852.

Leith, D., and Jones, D. L. (1984). Cyclones. In: Handbook of powder


science and technology 1nd Edition, (edited by Muhammad E. Fayed
and Lambert Otten). Chapman and Hall, New York, NY, USA,
pp. 727-752.

Leith, D., and Jones, D. L. (1997). Cyclones. In: Handbook of Powder


Science and Technology, 2nd Edition, (edited by Muhammad E.

References

217

Fayed and Lambert Otten). Chapman and Hall, New York, NY,
USA, pp. 727-752.

Liden, G., and Gudmundsson, A. (1997). Semi-Empirical Modeling to


Generalise the Dependence of Cyclone Collection Efficiency on
Operating Conditions and Cyclone Design. Journal of Aerosol
Science. Vol. 28, pp. 853-874.

Lim, K. S., Kim, H. S., and Lee, K. W. (2003a). Characteristics of the


Collection Efficiency for a Cyclone With Different Vortex Finder
Shapes. Journal of Aerosol Science. Vol. 35, pp. 743-754.

Lim, K. S., Kim, H. S., and Lee, K. W. (2003b). Comparative


Performances of Conventional Cyclones and a Double Cyclone With
and Without an Electric Field. Journal of Aerosol Science. Vol. 35,
pp. 103-116.

Lim, K. S., Kwon, S. B., and Lee, K. W. (2003c). Characteristics of the


Collection Efficiency for a Double Inlet Cyclone with Clean Air.
Journal of Aerosol Science. Vol. 34, pp. 1085-1095.

Madhumita, B. R., Hoffmann, A. C., and Postma, R. S. (2000).


Performance of Different Analytical Methods in Evaluating Grade
Efficiency of Centrifugal Separators. Journal of Aerosol Science.
Vol. 31, pp. 563-581.

References

218

Madhumita, B. R., Luning, P. E., Hoffmann, A. C., Plomp, A., and


Beumer, M. I. L. (1998). Improving the Removal Efficiency of
Industry-Scale Cyclones for particles Smaller than Five Micrometer.
International Journal of Mineral Processing. Vol. 53, pp. 39-47.

Marcus, R. D., Leung, L. S., Klinzing, G. E., and Rizk, F. (1990). Feeding
of Pneumatic Conveying Systems. In: Pneumatic Conveying of
Solids: A theoretical and practical approach. Thomson Press
(India) Ltd, New Delhi, pp. 238-308.

Mason, J. S., Mills, D., Reed, A. R., & Woodcock, C. R. (1983). Cyclone
Separators. In: Pneumatic Handling of Bulk Materials. School of
Mechanical Engineering, UK, pp.7-14.

Matthias, B. (1995). Influence of the Gas Temperature on the Separation


Efficiency of Aero cyclones. Chemical Engineering and Processing.
Vol. 34, pp. 151-156.

McQuiston, F. C. and Parker, J. D. (1988). Fans and Building Air


Distribution. In: Heating, Ventilating, and Air Conditioning:
Analysis and Design. 3rd Edition. NewYork: John Wiley & Sons,
Inc., pp. 11.1-11.12.

McQuiston, F. C. and Parker, J. D., (1994). Fans and Building Air


Distribution. In: Heating, Ventilating, and Air Conditioning:
Analysis and Design. 4th Edition. NewYork: John Wiley & Sons,
Inc., pp. 461-542.

References

219

Meier, H. F., and Mori, M. (1998). Gas- Solid Flow in Cyclones: The
Eulerian-Eulerian Approach. Computer and Chemical Engineering.
Vol. 22, pp. S641-S644.

Moore, M. E., and McFarland, A. R. (1993). Performance Modeling of


Single-Inlet Aerosol Sampling Cyclones. Environmental Science and
Technology. Vol. 27, pp. 1842-1848.

Moore, M. E., and McFarland, A. R. (1996). Design Methodology for


Multiple Inlet Cyclones. Environmental Science and Technology.
Vol. 30, pp. 271-276.

Mothes, H., and Loffler, F. (1985). Motion and Deposition of Particles in


Cyclones. Chemical Engineering. Vol. 8, pp. 223-233.

Obermair, S., and Staudinger, G. (2001). The Effect of the Dust Outlet
Geometry on the Separation Efficiency of a Gas Cyclone. Chemical
Engineering Technology. Vol. 24, pp. 1259-1263.

Pant, K., Crowe, C. T., and Irving, P. (2002). On the design of miniature
Cyclones for the collection of Bioaerosols. Powder Technology. Vol.
125, pp. 260-265.

Parker, R., Jain, R., Calvert, S., Drehmel, D., and Abbott, J. (1981).
Particle Collection in Cyclones at High Temperature and High
Pressure. American Chemical Society. Vol. 15, pp. 451-458.

220

References

Peukert, W., and Wadenpohl, C. (2001). Industrial Separation of Fine


Particles with Difficult Dust Properties. Powder Technology. Vol.
118, pp. 136-148.

Plomp, A., Beumer, m. I. L., and Hoffmann, A. (1996). Post Cyclone


(PoC) An Approach to a Better Efficiency of Dust Cyclones. Journal
of Aerosol Science. Vol. 27, pp. S631-S632.

Plucinski, J., Gradon, L., and Nowicki, J. (1989). Collection of Aerosol


Particles in a Cyclone with an External Electric Field. Journal of
Aerosol Science. Vol. 20, pp. 695-700.

Ranz, W. E. (1985). Wall Flows in a Cyclone Separator: A Description of


Internal Phenomena. Aerosol Science and Technology. Vol. 4, pp.
417-432.

Rhodes, M. (1998). Separation of Particles From a Gas: Gas Cyclones. In:


Introduction to Particle Technology. John Wiley and Sons Ltd,
England, pp.175-192.

Saltzman, B. E., and Hochstrasser, J. M. (1983). Design and Performance


of

Miniature

Cyclones

for

Respirable

Aerosol

sampling.

Environmental Science and Technology. Vol. 17, pp. 418-424.

Shepherd, C. B., and Lapple, C. E. (1939). Flow Pattern and Pressure Drop
in Cyclone Dust Collectors, Industrial and Engineering Chemistry.
Vol. 31, pp. 972-984.

References

221

Shepherd, C. B., and Lapple, C. E. (1940). Flow Pattern and Pressure Drop
in Cyclone Dust Collectors. Industrial and Engineering Chemistry.
Vol. 32, pp. 1246-1248.

Shin, M. S., Kim, H. S., Jang, D. S., Chung, J. D., and Bohnet, M. (2005).
A numerical and Experimental Study on a High Efficiency Cyclone
Dust Separator for High Temperature and Pressurized Environments.
Applied Thermal Engineering. Vol. 25, pp. 1821- 1835.

Smith, H. L. (1962). An Experimental Study of the Vortex in the Cyclone


Separator. Transaction of the ASME. December, pp. 602-608.

Smith, W. B., Cushing, K. M., and Wilson, R. R. (1982). Cyclone Samplers


for Measuring the Concentration of Invaluable Particles in Process
Streams. Journal of Aerosol Science. Vol. 13, pp. 259-267.

Stairmand, C. J. (1949). Pressure Drop in Cyclone Separators. Engineering.


Vol. 168, pp. 409 412.

Stairmand, C. J. (1952). The Design and Performance of Modern Gas


Cleaning Equipment. Journal of Institute of Fuel. Vol. 29, pp. 58-76.

Stepanoff, A. J. (1969). Gravity Flow of Bulk Solids. In: Gravity Flow of


Bulk Solids and Transportation of Solids in Suspension. John
Wiley &Sons, Inc. USA, pp. 1-11.

References

222

Stern, A. C., Caplan, K. J., and Bush, P. D. (1955). Cyclone Dust


Collectors. American Petroleum Institute. New York, NY, pp. 167.

Swift, P. (1969). Dust Control in Industry-2. Steam and Heating Ingineer.


Vol. 38, pp. 10-16.
Tangbergen, H. J. (1955). Influence of the Size of the Dust Outlet on the
Efficiency of Cyclones. In: Problems and Control of Air Pollution,
(edited by F.S. Mallette). Reinhold, NewYork, pp. 225-235.

Ter Linden, A. J. (1949). Investigations Into Cyclone Dust Collector.


Proceedings Institute of Mechanical Engineers, IMechE
Conference. Vol. 160, pp. 233-500.

Timothy, J. and Wayne, T. (2000). Control of Particulate matter. In: Air


Pollution Engineering Manual, 2nd Edition, (edited by Wayne T. D.).
John Wiley & Sons, Inc., Canada, pp. 66-73.

USACE, (1988). Cyclones and Multi cyclones. In: Design: Air Pollution
Control Systems for Boilers and Incinerators. U.S. Army Corps of
Engineers, USA: UFC, pp. 1-16.
Utts, J. M. (2004). Standard Deviation. In: Mind on Statistics, 2nd Edition.
Brooks/Col-Thomson Learning, Canada, pp. 394-398.

References

223

Walton, W. H. (1974). Cyclone Dust Separators. In: Dust Control and Air
Cleaning, Vol. 9, (edited by Dorman R. G.). Pergamon Press Ltd.,
USA, pp. 236-279.

Wang, D. (1995). Analysis and Design of Air-Jet pumps for Pneumatic


Transportation of Bulk Solids in Pipelines. Ph.D. Thesis, University
Of Wollongong, Australia.

Wey, M. Y., Yu, L. J., and Jou, S. I. (1998). The influence of heavy metals
on the formation of organics and HCI during incinerating of PVCcontaining waste. Journal of Hazardous Materials. Vol. 60, pp. 259270.

Wypych, P. W. (2005). MECH372 Solids Handling and Process


Engineering, A Collection of Notes and Papers, Spring Session.
Faculty of Engineering, University of Wollongong.

Xiang, R., Park, S. H., and Lee, K. W. (2001). Effects of Cone Dimension
on Cyclone Performance. Journal of Aerosol Science. Vol. 32, pp.
549-561.

Yazdabadi, P. A., Griffith, A. J., and Syred, N. (1994). Characterization of


the PVC Phenomena in the Exhaust of A Cyclone Dust Separator.
Experiments in Fluids. Vol. 17, pp. 84-95.

Youngmin, J., Tien, C., and Madhumita, B. R. (2000). Development of a


Post Cyclone to Improve the Efficiency of Reverse Flow Cyclones.
Powder Technology. Vol. 113, pp. 97-108.

References

224

Yuu, S., Jotaki, T., Tomita, Y., and Yoshida, K. (1978). The Reduction of
Pressure Drop Due To Dust Loading in a Conventional Cyclone.
Chemical Engineering Science. Vol. 33, pp. 1573-1580.

Zenz, F. A. (1962). Use Fluid Data For Solids Flow Rates. Hydrocarbon
Processing and Petroleum Refiner. Vol. 41, pp. 159-168.

Zenz, F. A. (1975). Cyclone Separators. In: Manual on disposal of refinery


wastes: Volume on atmospheric emissions. American Petroleum
Institute Publication. Vol. 931, pp.1-55.

Zenz, F. A. (1999). Cyclone Design. In: Fluidization Solids Handling and


Processing: Industrial Applications, (edited by Wen-Ching Y.).
Noyes Publications, USA, pp. 773-816.

Zhu, Y., Lee, K. W. (1999). Experimental Study on Small Cyclones


Operating at High Flowrates. Journal of Aerosol Science. Vol. 30,
pp. 1303-1315.

APPENDICES

Appendix A

A. 1

CYCLONE PRESSURE DROP EXPERIMENTAL WORK

A. 1. 1 With Proper Inside Diameter of Vortex Finder 105mm I.D.


Flow Straightener Discharged Directly to Atmosphere
And the cyclone Separator Connected to Drum.

225

226

Appendices

Table A.1.1: Data spreadsheet (105mm I.D.) to atmosphere


CYCLONE EXPERIMENTAL WORK NO. 1
4 INCHES
PROPER INSIDE DIAMETER OF VORTEX FINDER 105MM.
FLOW STRAIGHT TO ATMOSPHERE.
CONNECTED TO DRUM.

Inlet diameter
iInlet area (cyc)
Inlet area (tub)
Air supply annubar
Room temperature
Room pressure
Room humidity
Air flow temperature

Air mass flow

0.1
0.008064
0.008659
1.5
20
100
70
20

m
m2
m2
inch
C
kpa
%
C

mf (kg/s)

Inlet Air
Density
kg/m3

Annubar Air
Pressure
(kpa g)

Cyclone
Inlet Velocity
Vi (m/s)

Velocity
Pressure
Vpi (Pa g)

Air Supply
Dp Meter
full scale
(in H2O)

Dp Meter
Reading
%

Cyclone
Manometer
(mm H2O)

Inlet Static
Pressure
Spi (Pa g)

Total
Pressure
Tpi (Pa g)

0.0599
0.0769
0.1022
0.1279
0.1492
0.1702
0.1810
0.1983
0.2115
0.2282
0.2460
0.2597

1.191
1.194
1.201
1.209
1.218
1.228
1.233
1.243
1.251
1.262
1.274
1.284

600
600
600
600
590
585
585
580
570
550
540
525

6.23
7.99
10.55
13.12
15.19
17.18
18.21
19.79
20.96
22.42
23.95
25.08

23.127
38.114
66.809
104.008
140.532
181.322
204.348
243.342
274.880
317.221
365.262
403.889

6
6
6
30
30
30
30
30
30
30
30
30

23
38
67
21
29
38
43
52
60
72
85
97

29
55
111
179
262
340
388
468
542
634
737
824

284.41
539.39
1088.59
1755.47
2569.46
3334.41
3805.15
4589.72
5315.44
6217.69
7227.82
8081.04

307.533
577.504
1155.396
1859.476
2709.989
3515.732
4009.498
4833.058
5590.321
6534.914
7593.085
8484.928

0.0883
0.1235
0.1616
0.1768
0.1945
0.2097
0.2226
0.2431
0.2600

1.197
1.208
1.223
1.231
1.241
1.249
1.258
1.273
1.289

630
600
590
585
580
570
555
540
520

9.15
12.68
16.38
17.81
19.43
20.82
21.94
23.68
25.01

50.081
97.154
164.088
195.160
234.359
270.732
302.862
356.948
403.218

6
6
30
30
30
30
30
30
30

48
98
34
41
50
59
68
83
98

83
169
304
368
457
522
601
725
866

813.99
1657.40
2981.35
3609.01
4481.84
5119.30
5894.06
7110.14
8492.94

864.069
1754.552
3145.442
3804.168
4716.198
5390.031
6196.921
7467.086
8896.155

0.1137
0.1528
0.1810
0.2032
0.2234
0.2366
0.2469
0.2587

1.204
1.219
1.232
1.247
1.259
1.267
1.277
1.285

600
600
585
575
550
545
530
520

11.71
15.55
18.22
20.21
22.00
23.16
23.98
24.96

82.557
147.364
204.514
254.668
304.728
339.667
367.150
400.346

6
30
30
30
30
30
30
30

83
30
43
55
69
78
87
97

139
268
379
505
613
683
760
832

1363.19
2628.30
3716.89
4952.58
6011.74
6698.24
7453.39
8159.50

1445.742
2775.663
3921.400
5207.247
6316.472
7037.907
7820.536
8559.842

0.1309
0.1603
0.1852
0.2040
0.2226
0.2416
0.2512
0.2600

1.211
1.222
1.236
1.247
1.259
1.271
1.279
1.288

600
600
585
580
555
540
530
520

13.40
16.27
18.58
20.28
21.93
23.58
24.35
25.03

108.781
161.702
213.334
256.555
302.621
353.202
379.217
403.531

30
30
30
30
30
30
30
30

22
33
45
55
68
82
90
98

194
294
412
503
608
713
782
864

1902.57
2883.28
4040.52
4932.96
5962.71
6992.45
7669.14
8473.32

2011.355
3044.986
4253.854
5189.519
6265.330
7345.655
8048.359
8876.854

Appendices

A. 1. 2 With Proper Inside Diameter of Vortex Finder 130mm I.D.


Flow Straightener Discharged Directly to Atmosphere
And the cyclone Separator Connected to Drum.

227

228

Appendices

Table A.1.2: Data spreadsheet (130mm I.D.) to atmosphere


CYCLONE EXPERIMENTAL WORK NO. 1
5 INCHES
PROPER INSIDE DIAMETER OF VORTEX FINDER 130MM
FLOW STRAIGHT TO ATMOSPHERE
CONNECTED TO DRUM.

Inlet diameter
Inlet area (cyc)
Inlet area (tub)
Air supply annubar
Room temperature
Room pressure
Room humidity
Air flow temperature

0.1
0.008064
0.008659
1.5
20
100
72
20.5

m
m2
m2
inch
C
kpa
%
C

Air mass flow

Inlet Air

Annubar Air

Cyclone

Velocity

Air Supply

Dp Meter

Cyclone

Inlet Static

Total

mf (kg/s)

Density
kg/m3

Pressure
(kpa g)

Inlet Velocity
Vi (m/s)

Pressure
VPi
(Pa g)

Dp Meter
Full Scale
(in H2O)

Reading
(%)

Manometer
(Cm H2O)

Pressure
SPi (Pa g)

Pressure
TPi (Pa)

0.0623
0.0882
0.0998
0.1115
0.1278
0.1449
0.1626
0.1741
0.1912
0.1972
0.2105
0.2349
0.2448
0.2591
0.2700

1.191
1.193
1.195
1.197
1.201
1.204
1.209
1.212
1.217
1.219
1.224
1.234
1.239
1.246
1.254

600
600
600
600
600
600
600
600
600
600
600
600
550
550
525

6.492
9.165
10.352
11.554
13.193
14.922
16.676
17.816
19.478
20.057
21.327
23.610
24.499
25.788
26.697

25.095
50.106
64.029
79.902
104.522
134.051
168.107
192.351
230.856
245.188
278.373
343.935
371.839
414.298
446.896

6
6
6
6
30
30
30
30
30
30
30
30
30
30
30

25
50
64
80
21
27
34
39
47
50
57
71
83
93
105

2.1
4.4
5.95
7.6
10.75
13.6
17.6
20.35
25.1
26.9
30.8
39.4
44.1
49.7
56.4

206
431.5
584
745
1059
1334
1726
2001
2462
2638
3021
3864
4325
4874
5531

231.095
481.606
648.029
824.902
1163.522
1468.051
1894.107
2193.351
2692.856
2883.188
3299.373
4207.935
4696.839
5288.298
5977.896

0.0925
0.1183
0.1422
0.1807
0.2011
0.2178
0.2333
0.2373
0.2477
0.2635

1.194
1.198
1.204
1.215
1.221
1.227
1.233
1.238
1.243
1.251

600
600
600
600
600
600
600
550
550
525

9.604
12.245
14.643
18.443
20.421
22.009
23.462
23.769
24.713
26.117

55.071
89.815
129.086
206.636
254.578
297.179
339.366
349.722
379.574
426.636

6
6
30
30
30
30
30
30
30
30

55
90
26
42
52
61
70
78
85
100

5.5
9.2
13.85
22.8
28.55
33.6
38.6
42.8
46.9
54.4

539
902
1363
2236
2805
3295
3786
4197
4600
5335

594.071
991.815
1492.086
2442.636
3059.578
3592.179
4125.366
4546.722
4979.574
5761.636

0.1013
0.1366
0.1673
0.1912
0.2213
0.2373
0.2492
0.2635

1.196
1.202
1.211
1.218
1.228
1.238
1.243
1.251

600
600
600
600
600
550
550
525

10.504
14.092
17.131
19.462
22.349
23.769
24.858
26.117

65.974
119.355
177.702
230.666
306.673
349.722
384.039
426.636

6
30
30
30
30
30
30
30

66
24
36
47
63
78
86
100

6.7
12.2
19.6
26.2
34.6
42.75
47.4
54.05

657
1196
1922
2569
3393
4197
4648
5306

722.974
1315.355
2099.702
2799.666
3699.673
4546.722
5032.039
5732.636

0.1065
0.1308
0.1475
0.1626
0.1870
0.1972
0.2160
0.2403
0.2492
0.2581

1.197
1.201
1.205
1.209
1.216
1.219
1.227
1.238
1.242
1.249

600
600
600
600
600
600
600
550
550
525

11.037
13.504
15.184
16.676
19.075
20.057
21.828
24.072
24.878
25.630

72.911
109.500
138.901
168.107
221.214
245.188
292.307
358.689
384.349
410.227

6
30
30
30
30
30
30
30
30
30

73
22
28
34
45
50
60
80
86
96

7.5
11.3
14.6
17.85
24.2
27.15
33.3
43.2
45.9
52.55

735
1108
1432
1755
2373
2668
3266
4237
4501
5159

807.911
1217.500
1570.901
1923.107
2594.214
2913.188
3558.307
4595.689
4885.349
5569.227

0.1094
0.1366
0.1602
0.1763
0.1932
0.2030
0.2178
0.2327
0.2433
0.2621

1.197
1.202
1.208
1.213
1.219
1.222
1.227
1.235
1.239
1.248

600
600
600
600
600
600
600
550
550
525

11.336
14.092
16.443
18.028
19.652
20.599
22.009
23.364
24.351
26.048

76.906
119.355
163.298
197.121
235.381
259.261
297.179
337.088
367.359
423.385

6
30
30
30
30
30
30
30
30
30

77
24
33
40
48
53
61
75
82
99

8
12.2
17.65
21.25
26.35
29.05
33.75
40.8
43.9
51.7

784.6
1196
1736
2089
2589
2854
3315
3962
4305
5070

861.506
1315.355
1899.298
2286.121
2824.381
3113.261
3612.179
4299.088
4672.359
5493.385

Appendices

A. 1. 3

With Proper Inside Diameter of Vortex Finder 105mm I.D.


Flow Straightener Connected Via A Pipe to A Filter
And the cyclone Separator Connected to Drum.

229

230

Appendices

Table A. 1.3: Data spreadsheet (105mm I.D.) connected to filter


CYCLONE EXPERIMENTAL WORK NO. 2
4 INCHES
PROPER INSIDE DIAMETER OF VORTEX FINDER 105MM
FLOW STRAIGHT TO THE FILTER
CONNECTED TO DRUM.

Inlet pipe diameter


Inlet area (cyc)
Inlet area (tub)
Air supply annubar
Room temperature
Room pressure
Room humidity
Air flow temperature

0.1
0.008064
0.008659
1.5
20
100
73
20

*Manometer(1) SP, connected to (inlet&exit)


*Manometer(2) SP1, connected to (inlet&atm.)

m
m2
m2
inch
C
kpa
%
C
CYCLONE INLET

CYCLONE OUTLET

CYCLONE ITSELF

Air mass flow

Inlet air

Annubar

Air supply

Dp Meter

Difference

Static

Cyclone Inlet

Velocity

Total

Cyclone Static

Cyclone Outlet

Cyclone Outlet

Velocity

Total

Total

Total Pressure

mf (kg/s)

Density
(kg/m3)

air pressure
(kpa g)

Dp Meter
full scale
(in H2O)

Reading
%

static pressure
DSP
(mm H2O)

Pressure(1)
SP1
(pa g)
(mm H2O)

Velocuty(1)
V1 (m/s)

Pressure
VP1 (pa g)

Pressure(1)
TP1 (Pa g)

Pressure(2)
SP2
(mmH2O)
(Pa g)

Air Density
(kg/m3)

Velocity(2)
V2 (m/s)

Pressure(2)
VP2 (Pa g)

Pressure(2)
TP2 (pa g)

Static pressure
DSP (Pa g)

DTP (Pa g)

0.0834
0.1088
0.1338
0.1579
0.1696
0.1906
0.2006
0.2099
0.2218

1.203
1.214
1.226
1.24
1.249
1.263
1.273
1.282
1.302

610
600
600
600
580
580
570
560
550

6
6
30
30
30
30
30
30
30

44
76
23
32
38
48
54
60
68

72
132.5
204
282
341
420
480
528
590

130
223
328
441
526
642
729
802
973

1274.86
2186.88
3216.58
4324.73
5158.30
6295.87
7149.05
7864.93
9541.87

8.594
11.114
13.537
15.787
16.834
18.710
19.544
20.303
21.120

44.427
74.972
112.334
154.526
176.972
221.066
243.117
264.230
290.393

1319.292
2261.855
3328.915
4479.259
5335.270
6516.935
7392.165
8129.163
9832.264

58
90.5
124
159
185
222
249
274
383

568.79
887.50
1216.02
1559.26
1814.23
2177.08
2441.86
2687.02
3755.95

1.195
1.199
1.203
1.207
1.21
1.214
1.217
1.22
1.233

8.057
10.479
12.848
15.104
16.183
18.128
19.038
19.869
20.770

38.789
65.836
99.289
137.683
158.434
199.468
220.556
240.811
265.950

607.575
953.338
1315.314
1696.941
1972.664
2376.544
2662.412
2927.833
4021.897

706.08
1299.38
2000.56
2765.48
3344.07
4118.79
4707.19
5177.91
5785.92

711.717
1308.517
2013.601
2782.318
3362.606
4140.391
4729.753
5201.330
5810.367

0.1109
0.1395
0.1603
0.1724
0.1865
0.1995
0.2099
0.2234
0.2341

1.215
1.231
1.242
1.251
1.262
1.272
1.28
1.292
1.301

600
600
600
585
580
575
560
550
540

6
30
30
30
30
30
30
30
30

79
25
33
39
46
53
60
69
77

135
229
297
350
417
479
527
598
652

232
366
463
538
632
718
788
892
970

2275.14
3589.23
4540.48
5275.98
6197.80
7041.17
7727.64
8747.53
9512.45

11.321
14.056
16.006
17.089
18.331
19.449
20.335
21.440
22.318

77.867
121.606
159.098
182.672
212.023
240.585
264.643
296.944
324.018

2353.010
3710.840
4699.577
5458.650
6409.826
7281.759
7992.283
9044.476
9836.468

97
137
166
188
215
239
261
294
318

951.25
1343.51
1627.90
1843.65
2108.43
2343.79
2559.54
2883.16
3118.51

1.199
1.204
1.208
1.21
1.213
1.216
1.219
1.222
1.225

10.684
13.384
15.326
16.454
17.761
18.947
19.885
21.110
22.074

68.435
107.833
141.868
163.799
191.314
218.266
241.008
272.290
298.453

1019.680
1451.344
1769.772
2007.449
2299.744
2562.056
2800.544
3155.445
3416.967

1323.90
2245.72
2912.58
3432.33
4089.37
4697.39
5168.10
5864.38
6393.94

1333.330
2259.496
2929.805
3451.201
4110.082
4719.704
5191.739
5889.031
6419.501

0.1029
0.1229
0.1528
0.1696
0.1804
0.1906
0.2051
0.2150
0.2234
0.2335

1.211
1.223
1.237
1.248
1.256
1.264
1.274
1.282
1.292
1.301

600
600
600
580
580
580
575
570
550
545

6
6
30
30
30
30
30
30
30
30

68
97
30
38
43
48
56
62
69
76

121.5
185
269
336
385
430
492
543.5
601
653

197
300
417
514
584
647
735
805
891
968

1931.91
2942.00
4089.37
5040.62
5727.08
6344.90
7207.89
7894.35
8737.73
9492.84

10.538
12.463
15.323
16.847
17.807
18.695
19.961
20.794
21.440
22.259

67.246
94.984
145.219
177.114
199.142
220.891
253.803
277.175
296.944
322.308

1999.156
3036.979
4234.592
5217.732
5926.226
6565.794
7461.691
8171.528
9034.670
9815.145

75.5
115
148
178
199
217
243
261.5
290
315

740.40
1127.76
1451.38
1745.58
1951.52
2128.04
2383.02
2564.44
2843.93
3089.09

1.197
1.202
1.205
1.209
1.211
1.213
1.216
1.219
1.222
1.225

9.929
11.809
14.649
16.196
17.200
18.143
19.476
20.366
21.110
22.016

59.004
83.818
129.292
158.565
179.132
199.632
230.621
252.815
272.290
296.878

799.406
1211.583
1580.677
1904.149
2130.656
2327.675
2613.637
2817.254
3116.219
3385.973

1191.51
1814.23
2637.99
3295.03
3775.56
4216.86
4824.87
5329.91
5893.80
6403.74

1199.750
1825.396
2653.916
3313.584
3795.570
4238.119
4848.054
5354.274
5918.451
6429.173

0.0789
0.1044
0.1239
0.1492
0.1679
0.1824
0.1945
0.2051
0.2151
0.2266

1.202
1.212
1.224
1.236
1.246
1.256
1.266
1.273
1.282
1.293

600
600
590
590
585
580
580
575
560
550

6
6
6
30
30
30
30
30
30
30

40
70
100
29
37
44
50
56
63
71

62
119
190
264
328
390
450
497
549
615

120
206
307
410
497
584
664
731
804
899

1176.80
2020.17
3010.64
4020.73
4873.91
5727.08
6511.62
7168.66
7884.55
8816.18

8.143
10.683
12.553
14.970
16.712
18.013
19.051
19.977
20.804
21.732

39.853
69.167
96.443
138.485
174.000
203.773
229.731
254.003
277.442
305.315

1216.651
2089.337
3107.085
4159.212
5047.905
5930.857
6741.347
7422.664
8161.988
9121.493

58
87
117
146
169
194
214
234
255
284

568.79
853.18
1147.38
1431.77
1657.32
1902.49
2098.62
2294.76
2500.70
2785.09

1.195
1.198
1.202
1.205
1.208
1.211
1.213
1.215
1.218
1.221

7.628
10.066
11.905
14.300
16.053
17.399
18.517
19.492
20.393
21.432

34.766
60.689
85.176
123.197
155.656
183.298
207.950
230.811
253.267
280.412

603.552
913.868
1232.554
1554.968
1812.980
2085.788
2306.573
2525.567
2753.962
3065.501

608.01
1166.99
1863.26
2588.96
3216.58
3824.59
4412.99
4873.91
5383.85
6031.09

613.099
1175.469
1874.531
2604.244
3234.925
3845.069
4434.774
4897.097
5408.026
6055.993

Appendices

A. 1. 4

With Proper Inside Diameter of Vortex Finder 130mm I.D.


Flow Straightener Connected Via A Pipe to A Filter
And the cyclone Separator Connected to Drum.

231

232

Appendices

Table A. 1.4: Data spreadsheet (130mm I.D.) connected to filter


CYCLONE EXPERIMENTAL WORK NO. 2
*Manometer(1) SP, connected to (inlet&exit)
*Manometer(2) SP1, connected to (inlet&atm.)

5 INCHES
PROPER INSIDE DIAMETER OF VORTEX FINDER 130MM
FLOW STRAIGHT TO THE FILTER
CONNECTED TO DRUM
Inlet diameter
Inlet area (cyc)
Inlet area (tub)
Air supply annubar
Room temperature
Room pressure
Room humidity
Air flow temperature

0.1
0.008064
0.008659
1.5
20
100
73
20

m
m2
m2
inch
C
kpa
%
C
CYCLONE INLET

CYCLONE OUTLET

CYCLONE ITSELF

Air mass

Inlet air

Annubar air

Air supply

Dp Meter

Cyclone

Static

Inlet

Velocity

Total

Static

Outlet

Outlet

Velocity

Total

Total

Total

flow

density

pressure

Dp Meter

Reading

Difference static

Pressure (1)

Velocuty(1)

Pressure

Pressure(1)

Pressure(2)

Air Density

Velocity (2)

Pressure(2)

Pressure(2)

Static pressure

Pressure

mf (kg/s)

(kg/m3)

(kpa g)

full scale

Pressure

SP1

V1 (m/s)

VP1

TP1 (Pa g)

SP2

(kg/m3)

V2 (m/s)

VP2

TP2

DSP

DTP

(Pa g)

(pa g)

(Pa g)

(Pa g)

(in H2O)

DSP (mm H2O)

(mm H2O)

(Pa g)

(Pa g)

(mmH2O)

(Pa g)

0.0837
0.1116
0.1450
0.1627
0.1783
0.1976
0.2116
0.2298
0.2402
0.2515

1.201
1.21
1.223
1.231
1.241
1.251
1.261
1.273
1.282
1.291

600
600
600
600
580
575
560
550
540
525

6
6
30
30
30
30
30
30
30
30

45
80
27
34
42
52
61
73
81
91

46
89
153
196
243
300
346
409
452
491

112
186
297
370
450
542
624
729
806
885

1098.34
1824.04
2912.58
3628.46
4412.99
5315.20
6119.35
7149.05
7904.16
8678.89

8.644
11.440
14.703
16.392
17.812
19.588
20.813
22.382
23.230
24.162

44.872
79.179
132.194
165.384
196.862
240.008
273.108
318.847
345.901
376.852

1143.217
1903.215
3044.769
3793.845
4609.854
5555.212
6392.457
7467.895
8250.061
9055.737

66
97
144
174
207
242
278
320
354
394

647.24
951.25
1412.16
1706.36
2029.98
2373.21
2726.25
3138.13
3471.55
3863.82

1.196
1.199
1.205
1.208
1.212
1.216
1.221
1.225
1.229
1.234

8.08
10.75
13.90
15.56
16.98
18.77
20.02
21.66
22.57
23.54

39.08
69.30
116.36
146.17
174.82
214.15
244.62
287.37
312.93
341.94

686.32
1020.55
1528.52
1852.52
2204.80
2587.36
2970.87
3425.50
3784.49
4205.76

451.11
872.79
1500.42
1922.10
2383.02
2942.00
3393.10
4010.92
4432.61
4815.07

456.90
882.67
1516.25
1941.32
2405.06
2967.85
3421.58
4042.40
4465.57
4849.98

0.0856
0.1197
0.1503
0.1656
0.1845
0.2035
0.2168
0.2311
0.2492

1.201
1.212
1.225
1.233
1.244
1.255
1.264
1.276
1.288

600
600
600
585
580
565
550
540
500

6
6
30
30
30
30
30
30
30

47
92
29
36
45
56
65
75
93

47
101
164
208
264
320
365
417
479

112
208
316
388
483
573
652
750
854

1098.34
2039.78
3098.90
3804.98
4736.61
5619.21
6393.94
7354.99
8374.88

8.834
12.248
15.213
16.659
18.393
20.112
21.270
22.458
23.989

46.866
90.905
141.754
171.082
210.415
253.829
285.927
321.785
370.590

1145.211
2130.688
3240.655
3976.063
4947.027
5873.039
6679.863
7676.772
8745.469

65
107
152
180
219
253
287
333
375

637.43
1049.31
1490.61
1765.20
2147.66
2481.08
2814.51
3265.61
3677.49

1.196
1.201
1.206
1.209
1.214
1.218
1.222
1.227
1.232

8.26
11.51
14.39
15.82
17.55
19.30
20.49
21.75
23.36

40.82
79.56
124.88
151.32
187.00
226.83
256.51
290.23
336.02

678.25
1128.87
1615.49
1916.52
2334.66
2707.91
3071.01
3555.84
4013.51

460.91
990.47
1608.29
2039.78
2588.96
3138.13
3579.43
4089.37
4697.39

466.96
1001.81
1625.17
2059.54
2612.37
3165.13
3608.85
4120.93
4731.96

0.1244
0.1503
0.1746
0.1945
0.2132
0.2266
0.2381
0.2512

1.214
1.225
1.239
1.252
1.262
1.272
1.28
1.283

610
600
585
580
570
550
545
530

6
30
30
30
30
30
30
30

98
29
40
50
61
71
79
90

112
165
234
292
350
402
441
493

225
317
436
544
633
723
792
896

2206.50
3108.71
4275.70
5334.82
6207.61
7090.21
7766.87
8786.76

12.710
15.213
17.475
19.264
20.953
22.090
23.067
24.275

98.055
141.754
189.171
232.300
277.026
310.356
340.527
378.035

2304.552
3250.462
4464.870
5567.118
6484.635
7400.564
8107.394
9164.793

113
152
202
252
283
321
351
403

1108.15
1490.61
1980.94
2471.28
2775.28
3147.93
3442.13
3952.08

1.201
1.206
1.212
1.218
1.221
1.226
1.229
1.235

11.96
14.39
16.64
18.44
20.17
21.34
22.37
23.49

85.96
124.88
167.72
207.10
248.33
279.27
307.59
340.61

1194.11
1615.49
2148.66
2678.37
3023.61
3427.20
3749.73
4292.69

1098.34
1618.10
2294.76
2863.54
3432.33
3942.27
4324.73
4834.68

1110.44
1634.97
2316.21
2888.75
3461.02
3973.36
4357.67
4872.10

0.1088
0.1395
0.1580
0.1824
0.1957
0.2115
0.2266

1.207
1.22
1.23
1.241
1.25
1.26
1.272

600
600
580
580
575
570
550

6
30
30
30
30
30
30

76
25
33
44
51
60
71

78
140
189
244
289
341
401

164
274
360
456
532
620
719

1608.29
2687.02
3530.39
4471.83
5217.14
6080.12
7050.98

11.178
14.183
15.930
18.231
19.415
20.813
22.090

75.407
122.702
156.060
206.236
235.580
272.917
310.356

1683.697
2809.725
3686.454
4678.069
5452.718
6353.040
7361.337

86
134
171
212
243
279
318

843.37
1314.09
1676.94
2079.01
2383.02
2736.06
3118.51

1.198
1.204
1.208
1.213
1.216
1.221
1.225

10.49
13.38
15.11
17.37
18.59
20.00
21.36

65.89
107.83
137.82
183.00
210.03
244.26
279.50

909.26
1421.92
1814.75
2262.01
2593.05
2980.31
3398.01

764.92
1372.93
1853.46
2392.82
2834.12
3344.07
3932.47

774.43
1387.80
1871.70
2416.06
2859.67
3372.73
3963.33

Appendices

A. 2 ANNUBAR CONVERSION GRAPH


A. 2. 1

GNT - 10
(1.5 inches 6.00 inch H2O)

Figure A. 1: Annubar conversion chart (6 inch H2O)

233

Appendices

A. 2. 2

GNT - 10
(1.5 inches 30.00 inch H2O)

Figure A. 2: Annubar conversion chart (30 inch H2O)

234

235

Appendices

APPENDIX B
FRICTION LOSS CALCULATIONS
B. 1 105mm I.D-Vortex Finder
Exit, Discharge to a Filter from a 90 Elbow, Round

D
r

B. 2 130mm I.D-Vortex Finder and Transition. Round to Round


Exit, Discharge to a Filter from a 90 Elbow

D
r

236

Appendices

Table B. 1: Data spreadsheet local loss coefficient


Local Loss Coefficients, TRANSITIONS for:
Ashrae,1985
McQuiston, 1988
A1/Ao(4") =
1.53 A1/Ao(4") =
1.53
A1/Ao(5") =
1.53 A1/Ao(5") =
1.53
the local loss coefficient, Co for all =
Inlet Gas Density =
Barrel I.D =
Gas Exit I.D(4'') =
Gas Exit I.D(5'') =

1.2
400
105
130

McQuiston, 1994
Ao/A1(4") =
0.652
Ao/A1(5") =
0.652
Ao (4'') =
A1 (5'') =

0.26
Kg/m3 =
mm
mm
mm

8659.01 mm2
13273.2 mm2

105mm, gas exit pipe


130mm, gas exit pipe

0.0749 lbm/ft3
DP= Co p V/2

EEUA. Model
Vortex Finder to Atmosphere:
4 Inch

Jacob et al. Model


Vortex Finder to Atomspher:
4 Inch

Rhodes Model
Vortex Finder to Atmosphere:
4 Inch

Mason et al. Model


Vortex Finder To Atmosphere:
4 Inch

Zenz Model
Vortex Finder to Atmosphere:
4 Inch

Inlet Velocity
m/s
5
10
15
20
25
5 Inch
Inlet Velocity
m/s
5
10
15
20
25

ft/s
16.404
32.808
49.213
65.617
82.021

contraction
Pa
2.6207
10.4827
23.5861
41.9308
65.5169

Inlet Velocity
contraction
m/s
ft/s
Pa
5
16.404
2.6207
10
32.808 10.4827
15
49.213 23.5861
20
65.617 41.9308
25
82.021 65.5169
5 Inch

Inlet Velocity
m/s (ft/s)
5
16.404
10
32.808
15
49.213
20
65.617
25
82.021
5 Inch

contraction
Pa
2.6207
10.4827
23.5861
41.9308
65.5169

Inlet Velocity
m/s (ft/s)
5
16.404
10
32.808
15
49.213
20
65.617
25
82.021
5 Inch

contraction
Pa
2.6207
10.4827
23.5861
41.9308
65.5169

Inlet Velocity
m/s
ft/s
5
16.404
10
32.808
15
49.213
20
65.617
25
82.021
5 Inch

contraction
Pa
2.6207
10.4827
23.5861
41.9308
65.5169

ft/s
16.404
32.808
49.213
65.617
82.021

contraction
Pa
2.6207
10.4827
23.5861
41.9308
65.5169

Inlet Velocity
m/s
5
10
15
20
25

Inlet Velocity
m/s
(ft/s)
5
16.404
10
32.808
15
49.213
20
65.617
25
82.021

contraction
Pa
2.6207
10.4827
23.5861
41.9308
65.5169

Inlet Velocity
m/s
(ft/s)
5
16.404
10
32.808
15
49.213
20
65.617
25
82.021

contraction
Pa
2.6207
10.4827
23.5861
41.9308
65.5169

Inlet Velocity
m/s
ft/s
5
16.404
10
32.808
15
49.213
20
65.617
25
82.021

contraction
Pa
2.6207
10.4827
23.5861
41.9308
65.5169

ft/s
16.404
32.808
49.213
65.617
82.021

contraction
Pa
2.6207
10.4827
23.5861
41.9308
65.5169

Vortex Finder Connected to Filter:


4 Inch

Vortex Finder Connected to Filter:


4 Inch

Vortex Finder Connected to Filter:


4 Inch

Vortex Finder Connected to Filter:


4 Inch

Vortex Finder Connected to Filter:


4 Inch

Inlet Velocity
m/s
8
10
15
20
22
5 Inch
Inlet Velocity
m/s
9
10
15
20
24

ft/s
26.247
32.808
49.213
65.617
72.178

contraction
Pa
6.7089
10.4827
23.5861
41.9308
50.7363

Inlet Velocity
m/s
8
10
15
20
22
5 Inch

ft/s
29.528
32.808
49.213
65.617
78.74

contraction
Pa
8.4910
10.4827
23.5861
41.9308
60.3804

Inlet Velocity
m/s
9
10
15
20
24

ft/s
26.247
32.808
49.213
65.617
72.178

contraction
Pa
6.7089
10.4827
23.5861
41.9308
50.7363

Inlet Velocity
m/s
(ft/s)
8
26.247
10
32.808
15
49.213
20
65.617
22
72.178
5 Inch

contraction
Pa
6.7089
10.4827
23.5861
41.9308
50.7363

Inlet Velocity
m/s
(ft/s)
8
26.247
10
32.808
15
49.213
20
65.617
22
72.178
5 Inch

contraction
Pa
6.7089
10.4827
23.5861
41.9308
50.7363

Inlet Velocity
m/s
ft/s
8
26.247
10
32.808
15
49.213
20
65.617
22
72.178
5 Inch

contraction
Pa
6.7089
10.4827
23.5861
41.9308
50.7363

ft/s
29.528
32.808
49.213
65.617
78.74

contraction
Pa
8.4910
10.4827
23.5861
41.9308
60.3804

Inlet Velocity
m/s
(ft/s)
9
29.528
10
32.808
15
49.213
20
65.617
24
78.74

contraction
Pa
8.4910
10.4827
23.5861
41.9308
60.3804

Inlet Velocity
m/s
(ft/s)
9
29.528
10
32.808
15
49.213
20
65.617
24
78.740

contraction
Pa
8.4910
10.4827
23.5861
41.9308
60.3804

Inlet Velocity
m/s
ft/s
9
29.528
10
32.808
15
49.213
20
65.617
24
78.740

contraction
Pa
8.4910
10.4827
23.5861
41.9308
60.3804

237

Appendices

B. 3 Total Pressure Loss Coefficients for Transitions:


Transitions, Round to Round

127mm

A1

A0
Q0

A0/A1< or >1

B. 4 Local Loss Coefficients, ENTRIES


Duct Mounted in Wall (Hood, Non-Enclosing, Flanged and Unflanged)

Di

E
D

V0
L

t =3mm

D
C

238

Appendices

Table B. 2: Data spreadsheet local loss coefficient


Local Loss Coefficients, TRANSITIONS for:
Ashrae,1985
McQuiston, 1988
McQuiston, 1994
A1/Ao(4") = 14.512 A1/Ao(4") = 14.51247166 Ao/A1(4") = 0.06891
A1/Ao(5") = 9.4675 A1/Ao(5") = 9.467455621 Ao/A1(5") = 0.10563
Ao (4'') =8659.0148 mm2
Ao (5'') =13273.229 mm2
A1 =
125663.71 mm2

the local loss coefficient, Co(4"&5") for all

0.43

Inlet Gas Density =


Barrel I.D =
Gas Exit I.D(4'') =
Gas Exit I.D(5'') =

0.07491 lbm/ft3

1.2
400
105
130

Kg/m3 =
mm
mm
mm

Inlet Velocity
m/s
5
10
15
20
25

Inlet Velocity
m/s
9
10
15
20
24

Barrel Area

Jacob Model
Vortex Finder to Atomspher:
4 Inch

ft/s
16.404
32.808
49.213
65.617
82.021

C' - D
contraction
Pa
4.334
17.337
39.008
69.347
108.355

ft/s
16.404
32.808
49.213
65.617
82.021

C' - D
contraction
Pa
4.334
17.337
39.008
69.347
108.355

D' - E
6.451
25.804
58.058
103.214
161.272

D' - E
6.85408
27.4163
61.6867
109.665
171.352

Vortex Finder Connected to a Filter:


4 Inch
Inlet Velocity
m/s
8
10
15
20
22
5 Inch

gas exit area

ft/s
26.247
32.808
49.213
65.617
72.178

C' - D
contraction
Pa
11.096
17.337
39.008
69.347
83.910

ft/s
29.528
32.808
49.213
65.617
78.74

C' - D
contraction
Pa
14.043
17.337
39.008
69.347
99.860

Ashrae,1985

DP= Co rV2/2

EEUA. Model
Vortex Finder to Atmosphere:
4 Inch
Inlet Velocity
m/s
5
10
15
20
25
5 Inch

gas exit area

Local Loss Coefficients, ENTRIES


Ind. Ventilation, 21st Edt.
Ind. Ventilation, 24th Edt.
105mm&130mm I.D - Vortex finder
t/D (4") =
0.029 mm
t/D (5") =
0.023 mm
L/D(4"&5") >
1
C o (4") =
0.64
C o (5") =
0.68

D' - E
16.5143
25.8036
58.0581
103.214
124.889

D' - E
22.2072
27.4163
61.6867
109.665
157.918

Inlet Velocity
m/s
ft/s
5
16.404
10
32.808
15
49.213
20
65.617
25
82.021
5 Inch
Inlet Velocity
m/s
ft/s
5
16.404
10
32.808
15
49.213
20
65.617
25
82.021

D' - E
6.451
25.80
58.06
103.21
161.27

D-E
6.854
27.416
61.687
109.665
171.352

Rhodes Model
Vortex Finder to Atmosphere:
4 Inch

Mason Model
Vortex Finder To Atmosphere:
4 Inch

Inlet Velocity
m/s
(ft/s)
5
16.4042
10
32.8084
15
49.2126
20
65.6168
25
82.021
5 Inch

Inlet Velocity
m/s
(ft/s)
5
16.404
10
32.808
15
49.213
20
65.617
25
82.021
5 Inch

Inlet Velocity
m/s
(ft/s)
5
16.4042
10
32.8084
15
49.2126
20
65.6168
25
82.021

D' - E
6.451
25.804
58.058
103.214
161.272

D-E
6.854
27.416
61.687
109.665
171.352

Inlet Velocity
m/s
(ft/s)
5
16.404
10
32.808
15
49.213
20
65.617
25
82.021

D' - E
6.451
25.804
58.058
103.214
161.272

D' - E
6.854
27.416
61.687
109.665
171.352

Vortex Finder Connected to Filter:


4 Inch

Vortex Finder Connected to Filter:


4 Inch

Vortex Finder Connected to Filter:


4 Inch

Inlet Velocity
m/s
ft/s
8
26.247
10
32.808
15
49.213
20
65.617
22
72.178
5 Inch

Inlet Velocity
m/s
(ft/s)
8
26.2467
10
32.8084
15
49.2126
20
65.6168
22
72.1785
5 Inch

Inlet Velocity
m/s
(ft/s)
8
26.247
10
32.808
15
49.213
20
65.617
22
72.178
5 Inch

Inlet Velocity
m/s
ft/s
9
29.528
10
32.808
15
49.213
20
65.617
24
78.74

D' - E
16.514
25.804
58.058
103.214
124.889

D' - E
22.207
27.416
61.687
109.665
157.918

Inlet Velocity
m/s
(ft/s)
9
29.5276
10
32.8084
15
49.2126
20
65.6168
24
78.7402

D' - E
16.514
25.804
58.058
103.214
124.889

D' - E
22.207
27.416
61.687
109.665
157.918

Inlet Velocity
m/s
(ft/s)
9
29.528
10
32.808
15
49.213
20
65.617
24
78.74

D' - E
16.514
25.804
58.058
103.214
124.889

D' - E
22.207
27.416
61.687
109.665
157.918

Zenz Model
Vortex Finder to Atmosphere:
4 Inch
C' - D
Inlet Velocity
contraction
m/s
ft/s
Pa
5
16.404
4.334
10
32.808
17.337
15
49.213
39.008
20
65.617
69.347
25
82.021
108.355
5 Inch
C' - D
Inlet Velocity
contraction
m/s
ft/s
Pa
5
16.404
4.334
10
32.808
17.337
15
49.213
39.008
20
65.617
69.347
25
82.021
108.355
Vortex Finder Connected to Filter:
4 Inch
C' - D
Inlet Velocity
contraction
m/s
ft/s
Pa
8
26.247
11.096
10
32.808
17.337
15
49.213
39.008
20
65.617
69.347
22
72.178
83.910
5 Inch
C' - D
Inlet Velocity
contraction
m/s
ft/s
Pa
9
29.528
14.043
10
32.808
17.337
15
49.213
39.008
20
65.617
69.347
24
78.74
99.860

D' - E
6.451
25.804
58.058
103.214
161.272

D-E
6.854
27.416
61.687
109.665
171.352

D' - E
16.5143
25.8036
58.0581
103.214
124.889

D' - E
22.2072
27.4163
61.6867
109.665
157.918

239

Appendices

Number of spirales, Ns

7
6
5
4
3
2

y = -0.000000000525x6 + 0.000000135574x5 - 0.000013863380x4 +


0.000725626710x3 - 0.021911545905x2 + 0.448018309720x
R2 = 0.999933426587

1
0
0

20

40

60

80

Inlet Velocity, Vi (m/s)


Figure B.1: show the number of vortex spirals against the inlet air velocity

0.6
0.5

0.4
0.3
2

y = -0.1855x - 0.2998x + 0.5


2
R = 0.9991

0.2
0.1
0
0

0.1

0.2

0.3

0.4

0.5

Area Ratio, (A0/Ai)

Figure B.2: Shows the function (K) verses area ratio (De/Dc) 2

Appendices

Please see print copy for Figure B.3

Figure B.3: Pressure loss of (EEUA model) via inlet velocity

240

Appendices

Please see print copy for Figure B.4

Figure B.4: Pressure loss of (Jacob et al. model) via inlet velocity

241

Appendices

Please see print copy for Figure B.5

Figure B.5: Pressure loss of (Rhodes model) via inlet velocity

242

Appendices

Please see print copy for Figure B.6

Figure B.6: Pressure loss of (Mason et al. model) via inlet velocity

243

Appendices

Please see print copy for Figure B.7

Figure B.7: Pressure loss of (Zenz model) via inlet velocity

244

Appendices

Please see print copy for Figure B.8

Figure B.8: Bend loss coefficient in 90 Round Elbows

245

Appendices

246

APPENDIX C

FLOW STRAIGHTENER PRESSURE DROP EXPERIMENTAL


WORK AND LOSS COEFFICIENT

C.1

4 inch or (105 mm I.D. Flow Straightner)

C.2

5 inch or (130 mm I.D. Flow Straightner)

247

Appendices

Table C.1: Data spreadsheet (105mm I.D.) F.S. pressure drop


FLOW STRAIGHTENER PRESSURE DROP
4 INCH
105mm I.D FLOW STRAIGHTENER
Air supply annubar =
Air Density =
Room temperature =
Room pressure =
Room humidity =
Inlet Area =
Distance between (tap1 & tap2) DL =
The proper diameter of flow straightener =
Thickness t =
The proper length of flow straightener =
Flow Cofficient =

Air Flow
mf (kg/s)
0
0.1428
0.1663
0.2011
0.2524

Air Flow
mf (kg/s)
0
0.1428
0.1663
0.1991
0.2524

2.5
1.2
24
101
62
0.0083
860
103
5
150
0.6026

inch =
kg/m3
C
Kpa
%
m2
mm
mm
mm
mm

62.713 mm

FLOW STRAIGHTENER LOSS COEFFICIENT

Inlet velocity
Vi (m/s)
0
14.28
16.63
20.12
25.25

WITH FLOW STRAIGHTENER


Annubar Air Pressure
Air supply Dp Meter
full scall (in H2O)
Kpa g
0
0
600
6
590
6
580
6
540
6

Inlet velocity
Vi (m/s)
0
14.28
16.63
19.91
25.25

WITHOUT FLOW STRAIGHTENER


Annubar Air Pressure
Air supply Dp Meter Dp Meter Reading
(Kpa g)
full scall (in H2O)
%
0
0
0
600
6
24
590
6
33
580
6
48
540
6
82

Dp Meter Reading
%
0
24
33
49
82

K=

0.182

Flow Straightener Pressure, DP (pa)


flow straightener difference Velocity Pressure Experimented
Predicted
pressure, SPD (Pa)
VP
(pa g)
(pa)
(pa)
0
0
0
0
45
122.43
25
22.28
62
165.93
32
30.20
86
242.82
44
44.19
132
382.49
67
69.61

flow straightner difference


pressure, SPD (Pa)
0
20
30
42
65

error
%
0
0.122
0.060
0.004
0.037

248

Appendices

Table C.2: Data spreadsheet (130mm I.D.) F.S. pressure drop


FLOW STRAIGHTENER PRESSURE DROP
5 INCH
130mm I.D FLOW STRAIGHTENER
Air supply annubar =
Air Density =
Room temperature =
Room pressure =
Room humidity =
Inlet Area =
Distance between (tap1 &tap2),DL =
The proper diameter of flow straightener =
Thickness t =
The proper length of flow straightener =
Flow Cofficient =

Air Flow
mf (kg/s)

Inlet velocity
Vi (m/s)

0.1400
0.1672
0.1994
0.2434

8.9295
10.658
12.715
15.518

Air Flow
mf (kg/s)

Inlet velocity
Vi (m/s)

0.1400
0.1672
0.1994
0.2434

8.93
10.66
12.72
15.52

2.5
1.2
23
101
62
0.0131
1040
129
6
129
0.6026

inch =
kg/m3
C
Kpa
%
m2
mm
mm
mm
mm

62.713 mm

FLOW STRAIGHTENER LOSS COEFFICIENT

WITH FLOW STRAIGHTENER


Annubar Air Pressure
Air supply Dp Meter
Kpa g
full scall (in H2O)
600
595
540
440

6
6
6
6

WITHOUT FLOW STRAIGHTENER


Annubar Air Pressure
Air supply Dp Meter
full scall (in H2O)
Kpa g
600
590
570
530

6
6
6
6

Dp Meter Reading
%
23
33
51
90

Dp Meter Reading
%
24
34
50
84

K=

0.182

flow straightner difference Velocity Pressure


pressure, SPD (Pa)
VP (pa g)
0
20
47.841
29
68.152
40
97.003
60
144.476

Flow straightner difference


pressure, SPD (Pa)
10
15
21
33

Flow Straightener Pressure, DP (pa)


Experimented
Predicted
error
(pa)
(pa)
%
0
0
0
10
8.71
15
14
12.40
13
19
17.65
8
27
26.29
3

249

Appendices

F.S. Pressure Drop, (Pa)

80

y = 0.182x

70
60
50
40
30
20
10
0
0

100

200

300

400

Velocity Pressure, VP (Pa)

Figure C.1: Flow Straightener Loss Coefficient


(105mm I.D. and 130mm I.D.)

500

Appendices

APPENDIX D
PRESSURE DROP PREDICTION FOR FIVE MODELS

Table D.1: Pressure drop prediction of (EEUA Model)

Please see print copy for Table D.1

250

Appendices

Table D.2: Pressure drop prediction of (Jacob et al. Model)

Please see print copy for Table D.2

251

Appendices

Table D.3: Pressure drop prediction of (Rhodes Model)

Please see print copy for Table D.3

252

Appendices

Table D.4: Pressure drop prediction of (Mason et al. Model)

Please see print copy for Table D.4

253

Appendices

254

Table D.5: Pressure drop prediction of (Zenz Model)

Please see print copy for Table D.5

Appendices

APPENDIX E

NEW PRESSURE DROP MODEL

255

Appendices

256

Table E.1: Data spreadsheet - new Pressure drop model of (EEUA Model)

Please see print copy for Table E.1

Appendices

257

Table E.2: Data spreadsheet - new Pressure drop model of (Jacob et al. Model)

Please see print copy for Table E.2

Appendices

258

Table E.3: Data spreadsheet - new Pressure drop model of (Rhodes Model)

Please see print copy for Table E.3

Appendices

259

Table E.4: Data spreadsheet - new Pressure drop model of (Mason et al. Model)

Please see print copy for Table E.4

Appendices

260

Table E.5: Data spreadsheet - new Pressure drop model of (Zenz Model)

Please see print copy for Table E.5

Appendices

261

Table E.6: Data spreadsheet - new Pressure drop model of (Stairmand Model)

Please see print copy for Table E.6

Appendices

262

Table E.7: Data spreadsheet - new Pressure drop model of (Barth Model)

Please see print copy for Table E.7

Appendices

APPENDIX F

PARTICLE VELOCITY EXPERIMENTAL WORK

F. 1. Particle Velocity at the Cyclone Outlet Opening for Different


proportion of air flow to the blow tank feeder (before choking).
F. 2.

Particle Velocity at the Cyclone Outlet Opening - Gravity Flow


Conditions (choked flow)

263

264

Appendices

Tables F.1a: Data spreadsheet of Particles Velocity for different proportion


of air flow to the blow tank (0.25 to 0.04)

Experimental Results
Test No. Cam.20
Particle Velocity After the Cyclone
Material

white plastic pellets

Particle density
loose poured density
Air mass flow rate, mf (in)
Blow tank
Time interval per frame, t
1/1000 =
from the test rig scale:
1 pixel =
Frame
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Image

1809
1808
1807
1806
1805
1804
1803
1802
1801
1800
1799
1798
1797
1796
1795
x
D
Number of frames
=
Number of time steps =
t =
Particle Velocity Vp =
Total velocity
Vt =

920.5
501
0.25
0.04
0.001
0.00037

Particle no. (1)


X
Y
502
161
496
159
490
158
484
156
478
155
475
154
468
153
463
152
458
150
452
149
448
147
441
146
436
145
430
144
423
142
79
19
0.029
0.0070
15
15
14
14
0.0140
0.0140
2.09
0.50
2.150
m/s

kg/m3
kg/m3
kg/s
sec.
m

or

1000

Frame
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

pixels
m
sec
sec
m/s

PPS

Image

384
383
382
381
380
379
378
377
376
375
374
373
372
371
370
x
D
Number of frames
=
Number of time steps =
t =
Particle Velocity
Vp =
Total velocity
Vt =

Particle no. (2)


X
Y
486
161
481
159
473
158
469
156
465
155
459
154
452
153
447
152
443
149
437
147
430
146
425
145
420
144
415
143
407
140
79
21
pixels
0.029 0.0078
m
15
15
sec
14
14
0.0140 0.0140 sec
2.09
0.56
m/s
2.163
m/s

265

Appendices

Tables F.1b: Data spreadsheet of Particles Velocity for different proportion


of air flow to the blow tank (0.25 to 0.02)

Experimental Results
Test No. Cam. 31
Particle Velocity After the Cyclone
Material

white plastic pellets

Particle density
920.5
loose poured density
501
0.25
Air mass flow rate, mf (in)
0.02
Blow tank
Time interval per frame, t 1/1000 = 0.001
from the test rig scale:
1 pixel = 0.000341
Frame
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Image

1570
1569
1568
1567
1566
1565
1564
1563
1562
1561
1560
1559
1558
1557
1556
x
D
Number of frames
=
Number of time steps =
t =
Particle Velocity Vp =
Total velocity
Vt =

Particle no. (1)


X
Y
506
162
501
161
495
159
488
157
481
156
476
154
471
152
464
150
457
148
451
146
446
144
439
143
432
140
426
138
419
137
87
25
0.030
0.0085
15
15
14
14
0.0140 0.0140
2.12
0.61
2.205
m/s

kg/m3
kg/m3
kg/s
sec.
m

or

PPS

Frame
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

pixels
m
sec
sec
m/s

Image

496
495
494
493
492
491
490
489
488
487
486
485
484
483
482
x
D
Number of frames
=
Number of time steps =
t =
Particle Velocity
Vp =
Total velocity
Vt =

Particle no. (2)


X
Y
502
162
495
161
491
160
484
158
479
157
474
156
469
154
462
152
456
151
451
149
446
148
440
146
434
144
427
142
422
141
80
21
pixels
0.027 0.0072
m
15
15
sec
14
14
0.0140 0.0140 sec
1.95
0.51
m/s
2.015
m/s

266

Appendices

Tables F.1c: Data spreadsheet of Particles Velocity for different proportion


of air flow to the blow tank (0.2 to 0.02)

Experimental Results
Test No. Cam. 34
Particle Velocity After the Cyclone
Material

white plastic pellets

Particle density
loose poured density
Air mass flow rate, mf (in)
Blow tank
Time interval per frame, t 1/1000 =
from the test rig scale:
1 pixel =
Frame
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Image

1406
1405
1404
1403
1402
1401
1400
1399
1398
1397
1396
1395
1394
1393
1392
x
D
Number of frames
=
Number of time steps =
t =
Particle Velocity Vp =
Total velocity
Vt =

920.5
501
0.2
0.02
0.001
0.00034

kg/m3
kg/m3
kg/s
sec.
m

Particle no. (1)


X
Y
500
162
494
161
489
159
481
157
476
155
470
153
464
152
457
151
451
148
445
147
439
145
432
143
425
140
419
138
414
137
86
25
pixels
0.029 0.0085
m
15
15
sec
14
14
0.0140 0.0140 sec
2.09
0.61
m/s
2.181
m/s

or

PPS

Frame
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Image

325
324
323
322
321
320
319
318
317
316
315
314
313
312
311
x
D
Number of frames
=
Number of time steps =
t =
Particle Velocity Vp =
Total velocity
Vt =

Particle no. (2)


X
Y
485
162
479
159
471
158
466
157
459
154
453
153
447
151
441
149
436
148
429
146
422
144
416
143
408
142
405
141
398
139
87
23 pixels
0.030 0.0078 m
15
15
sec
14
14
0.0140 0.0140 sec
2.12
0.56
m/s
2.192
m/s

267

Appendices

Tables F.1d: Data spreadsheet of Particles Velocity for different proportion


of air flow to the blow tank (0.15 to 0.02)

Experimental Results
Test No. Cam. 36
Particle Velocity After the Cyclone
Material

white plastic pellets

Particle density
920.5
loose poured density
501
0.15
Air mass flow rate, mf (in)
0.02
Blow tank
Time interval per frame, t
1/1000 = 0.001
from the test rig scale:
1 pixel =
0.000341
Frame
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Image

1503
1502
1501
1500
1499
1498
1497
1496
1495
1494
1493
1492
1491
1490
1489
x
D
Number of frames
=
Number of time steps =
t =
Particle Velocity
Vp =
Total velocity
Vt =

Particle no. (1)


X
Y
490
162
483
160
478
158
471
157
466
155
460
154
454
153
449
151
444
150
437
148
430
146
425
145
419
143
415
141
407
140
83
22
0.028
0.0075
15
15
14
14
0.0140
0.0140
2.02
0.54
2.091
m/s

kg/m3
kg/m3
kg/s
sec.
m

or

PPS

Frame
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

pixels
m
sec
sec
m/s

Image

478
477
476
475
474
473
472
471
470
469
468
467
466
465
464
x
D
Number of frames
=
Number of time steps =
t =
Particle Velocity
Vp =
Total velocity
Vt =

Particle no. (2)


X
Y
488
162
482
160
477
158
472
156
466
154
461
153
455
151
450
149
445
148
438
147
432
145
426
143
419
141
415
140
407
138
81
24
pixels
0.028 0.0082
m
15
15
sec
14
14
0.0140 0.0140 sec
1.97
0.58
m/s
2.058
m/s

268

Appendices

Table F.2a: Data spreadsheet of (Plastic pellets) velocity at the cyclone


outlet (choked flow)

Experimental Results
Test No. PPD1
Particle Velocity After the Cyclone - (choked flow)
Material :

white plastic pellets

Particle density
loose poured density
Time interval per frame, t
from the test rig scale:

Frame
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Image

1824
1823
1822
1821
1820
1819
1818
1817
1816
1815
1814
1813
1812
1811
1810
x
D
Number of frames
=
Number of time steps =
t
Vp
Particle Velocity

1/1000 =
1 pixel =

Particle no. (1)


X
Y
342
151
341
152
337
152
335
151
331
152
328
151
325
152
322
151
319
151
317
151
314
151
312
151
308
150
305
151
302
150
40
pixels
0.020
m
15
sec
14
0.01400
sec
1.43 m/s

920.5
531
0.001
0.0005

kg/m3
kg/m3
sec.
m

or

Frame
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

1000

Image

1192
1191
1190
1189
1188
1187
1186
1185
1184
1183
1182
1181
1180
1179
1178
x
D
Number of frames
=
Number of time steps =
t
Vp
Particle Velocity

PPS

Particle no. (2)


X
Y
349
133
346
133
344
133
341
133
338
133
335
133
333
133
330
133
327
133
324
133
322
133
319
133
316
133
313
133
310
133
39
pixels
0.020
m
15
sec
14
0.01400
sec
1.39 m/s

269

Appendices

Table F.2b: Data spreadsheet of (Corn) velocity at the cyclone outlet


(choked flow)

Experimental Results
Test No. CD1
Particle Velocity After the Cyclone - Gravity Flow Conditions
Material :

Corn

Particle density
loose poured density
Time interval per frame, t
from the test rig scale:

Frame
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Image

1/1000 =
1 pixel =

Particle no. (1)


X
Y
1887
168
112
1886
169
113
1885
172
113
1884
176
113
1883
178
113
1882
181
113
1881
184
113
1880
187
113
1879
189
113
1878
182
113
1877
195
113
1876
199
114
1875
202
115
1874
203
115
1873
206
114
x
38
pixels
D
0.019
m
Number of frames
=
15
sec
Number of time steps =
14
t
0.01400
sec
Velocity
Vp
1.36 m/s

1380
670
0.001
0.0005

kg/m3
kg/m3
sec.
m

or

Frame
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

1000

Image

PPS

Particle no. (2)


X
Y
1867
166
107
1866
169
107
1865
172
108
1864
175
108
1863
178
108
1862
181
108
1861
183
108
1860
186
108
1859
189
108
1858
192
108
1857
195
109
1856
199
109
1855
201
109
1854
204
109
1853
207
109
x
41
pixels
D
0.021
m
Number of frames
=
15
sec
Number of time steps =
14
t
0.01400
sec
Velocity
Vp
1.46 m/s

270

Appendices

Table F.2c: Data spreadsheet of (Rape seed) velocity at the cyclone outlet
(choked flow)

Particle Velocity After the Cyclone - Gravity Flow


Material :

Rape Seed

Particle density
loose poured density
Time interval per frame, t
from the test rig scale:

Frame
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Image

1139.6
621.1
1/1000 = 0.001
1 pixel =
0.0005

Particle no. (1)


X
Y
1938
144
147
1937
147
147
1936
150
147
1935
152
147
1934
155
147
1933
158
147
1932
161
147
1931
164
147
1930
167
147
1929
170
147
1928
173
147
1927
176
147
1926
179
147
1925
182
147
1924
185
147
pixels
x
41
m
D
0.0205
sec
Number of frames
=
15
Number of time steps =
14
sec
t
0.01400
Velocity
Vp
1.46 m/s

kg/m3
kg/m3
sec.
m

or

Frame
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

1000

Image

PPS

Particle no. (2)


X
Y
1839
144
141
1838
147
141
1837
150
141
1836
153
141
1835
155
141
1834
158
141
1833
161
141
1832
164
141
1831
167
141
1830
169
141
1829
172
141
1828
173
141
1827
177
141
1826
181
141
1825
184
141
pixels
x
40
m
D
0.020
sec
Number of frames
=
15
Number of time steps =
14
sec
t
0.01400
Velocity
Vp
1.43 m/s

Appendices

271

APPENDIX G

G. 1. Mass flow rate prediction for Plastic pellets, Corn, and Rape seed
(canola) with different cone angles of cyclone separator , using
four theoretical models of Beverloo et al. (1961), Brown (1961),
Zenz (1962) and Johansone (1965).

Table G.1a: Data spreadsheet of maximum mass flow rate prediction


(Plastic pellets)

Please see print copy for Table G.1a

Appendices

Table G.1b: Data spreadsheet of maximum mass flow rate prediction


(Rape seed)

Please see print copy for Table G.1b

272

Appendices

273

Table G.1c: Data spreadsheet of maximum mass flow rate prediction (Corn)

Please see print copy for Table G.1c

Appendices

274

APPENDIX H

BULK MATERIAL PROPERTIES

The materials that were used in this thesis are:


1. Plastic Pellets
2. Corn
3. Rape Seed (Canola)

H.1. Particle Size Distribution

The three different materials used in this work were updated and their
physical properties were found to be quite different. Tables H.1 and H.2
illustrate the particle sizes of the old and new materials tested.

275

Appendices

Table H.1: Data spreadsheet of particle size distributions (old materials)


Rape seed

test 1
sieve Size Range

weight empty

weight after sieving

2.36
2
1.4
1
pan

427.02
412.53
369.15
352.17
241.89

427.02
412.9
399.98
352.75
242.85

test 2
sieve Size Range
2.36
2
1.4
1
pan

weight empty
427.02
412.53
369.15
352.17
241.89

weight after sieving


427.02
412.88
398.71
352.72
242.92

test 1
sieve Size Range

weight empty

5.6
4.75
4
3.35
2.8
pan

Sample weight
gm
0
0.37
30.83
0.58
0.96
32.74

% of sample

Sum

0
1.1301
94.1662
1.7715
2.9322
100

0
1.1301
95.296
97.068
100

Sample weight
0
0.35
29.56
0.55
1.03
31.49

% of sample
0
1.1115
93.8711
1.7466
3.2709
100

Sum
0
1.1115
94.9825
96.7291
100

Sample weight
gm
0
1.27
31.65
0.94
0.19
0.15
34.2

% of sample

Sum

514.39
432.46
420.42
488.56
428.26
241.89

weight after sieving


sieving
514.39
433.73
452.07
489.5
428.45
242.04

0
3.7135
92.5439
2.7485
0.5556
0.4386
100

0
3.71345
96.2573
99.0058
99.5614
100

test 2
sieve Size Range
5.6
4.75
4
3.35
2.8
pan

weight empty
514.39
432.47
420.42
488.56
428.25
241.92

weight after sieving


514.39
433.79
450.23
489.66
428.33
242.1

Sample weight
0
1.32
29.81
1.1
0.08
0.18
32.49

% of sample
0
4.0628
91.7513
3.3857
0.2462
0.5540
100

Sum
0
4.06279
95.8141
99.1998
99.446
100

test 1
sieve Size Range

weight empty

weight after sieving

Sample weight

% of sample

Sum

9.5
8
6.3
5.6
4.75
3.35
pan

444.82
489.06
533.78
514.41
432.49
488.62
241.93

446.68
575.18
664.78
530.66
439.01
494.79
242.83

1.86
86.12
131
16.25
6.52
6.17
0.9
248.82

0.7475
34.6114
52.6485
6.5308
2.6204
2.4797
0.3617
100

0.748
35.3594
88.0079
94.5387
97.1591
99.6388
100

test 2
sieve Size Range
9.5
8
6.3
5.6
4.75
3.35
pan

weight empty
444.82
489.07
533.86
514.42
432.5
488.62
241.97

weight after sieving


444.82
567.73
674.97
528.81
439.02
495.53
244.67

Sample weight
0
78.66
141.11
14.39
6.52
6.91
2.7
250.29

% of sample
0
31.4275
56.3786
5.7493
2.6050
2.7608
1.0787
100

Sum
0
31.4275
87.8061
93.5555
96.1605
98.9213
100

Plastic Pellets

Corn

276

Appendices

mass>d (%)

Cumulative Distribution

100

50

0
1

10

d(mm)
d50=1.6mm

Figure H.1: Particle size distribution (old rape seed)

mass>d (%)

Cumulative Distribution

100

50

0
1

d (mm)
d50= (4.3mm)

Figure H.2: Particle size distribution (old plastic pellets)

10

277

Appendices

mass>d (%)

Cumulative Distribution

100

50

0
1

10
d(mm)
d50=7.4mm

Figure H.3: Particle size distribution (old corn)

100

278

Appendices

Table H.2: Data spreadsheet of particle size distributions (new materials)

Rape Seed

test 1
sieve Size Range

weight empty

weight after sieving

Sample weight

% of sample

Sum

2.36
1.4
1
pan

393.75
369.19
352.19
239.63

393.76
416.06
353.66
239.67

0.01
46.87
1.47
0.04

0.02
96.86
3.04
0.08

0.02
96.88
99.92
100

48.39

100

test 2
sieve Size Range

weight empty

weight after sieving

Sample weight

% of sample

Sum

2.36
1.4
1
pan

393.75
369.19
352.19
239.63

393.77
426.16
353.99
239.68

0.02
56.97
1.8
0.05

0.034
96.822
3.059
0.085

0.034
96.86
99.92
100

58.84

100

Corn

test 1
sieve Size Range

weight empty

weight after sieving

Sample weight

% of sample

9.5
8
5.6
4
pan

444.7
488.93
451.49
420.49
328.48

444.7
505.27
545.09
423.77
330.03

0
16.34
93.6
3.28
1.55

0
14.24
81.55
2.86
1.35

114.77

100

Sample weight
0
14.18
86.57
2.68
0.69

% of sample
0
13.62
83.14
2.57
0.66

104.12

100

test 2
sieve Size Range
9.5
8
5.6
4
pan

weight empty
444.7
488.93
451.49
420.49
328.48

weight after sieving


444.7
503.11
538.06
423.17
329.17

Sum
0
14.24
95.79
98.65
100

Sum
0
13.62
96.76
99.34
100

279

Appendices

mass>d (%)

Cumulative Distribution

100

50

0
1

d (mm)
d50= (4.3mm)

10

Figure H.4: Particle size distribution (new plastic pellets)

Mass>d(%)

Cumulative Distribution

100

50

10
d (mm)
d50=(1.9mm)

Figure H.5: Particle size distribution (new rape seed)

280

Appendices

Mass>d(%)

Cumulative Distribution

100

50

0
1

10
d(mm)
d50=(7mm)

Figure H.6: Particle size distribution (new corn)

100

Appendices

281

H.2. Instantaneous Yield Loci (IYL) and Wall Yield Loci (WYL)
Measured for All Test Materials

The flow functions of effective angle of internal friction, instantaneous


yield loci, and wall friction angle are given. Also, Kinematics angle of wall
friction with (mild steel and stainless steel), and the wall yield loci of the
above materials are provided in this section.

Figure H. 7: Effective Angle of Internal Friction (Plastic Pellets)

Appendices

Figure H. 8: Instantaneous Yield Loci (Plastic Pellets)

Figure H. 9: Wall Friction Angle (plastic pellets)

282

Appendices

Figure H. 10: Wall Yield Loci (Plastic Pellets)

Figure H. 11: Wall Friction Angle (plastic pellets)

283

Appendices

Figure H. 12: Wall Yield Loci (Plastic Pellets)

Figure H. 13: Effective Angle of Internal Friction (Corn)

284

Appendices

Figure H. 14: Instantaneous yield Loci (Corn)

Figure H. 15: Wall Friction Angle (Corn)

285

Appendices

Figure H. 16: Wall Yield Loci (Corn)

Figure H. 17: Wall Friction Angle (Corn)

286

Appendices

Figure H. 18: Wall Yield Loci (Corn)

Figure H. 19: Effective Angle of Internal Friction (Rape Seed)

287

Appendices

Figure H. 20: Instantaneous yield Loci (Rape Seed)

Figure H. 21: Wall Friction Angle (Rape Seed)

288

Appendices

Figure H. 22: Wall Yield Loci (Rape Seed)

Figure H. 23: Wall Friction Angle (Rape Seed)

289

Appendices

Figure H. 24: Wall Yield Loci (Rape Seed)

290

Appendices

291

APPENDIX I

PUBLICATIONS DURING Ph.D. CANDIDATURE

[1]

Saad, M. S. and Wypych, P. W., Effect of Cyclone Vortex Finder


Diameter on Pressure Drop, submitted to the 8th International
Conference on Bulk Materials Storage, Handling and Transportation,
5 - 8 July, 2004, Wollongong, Australia.