bulk

solicls Volume 1..1 • Number 2 • ApnL'June 1994 ·:.

tiandlins
_._ �
·

Advances in the Design of

Mechanical conveyors

A.W. ;Roberts, Australia

Summary quently, greatly improved design methodologies. It is important,

herefore, that engineers and those involved in the design and
operation of handling plant and transportation systems maintain
Recent developments in the technology of mechanical convey­
an awareness of these developments.
ing of bulk solids are reviewed. An overvie of technical and ec­
onomic factors involved in conveyor design is presen ed. Be­ When comparisons are made between continuous modes of
cause of their wide use in bulk solids handling operations, the transport such as slurry pipelines, belt conveyors, screw con­
paper focuses on belt conveyors indicating research develop­ veyors and pneumatic systems, and discontinuous modes such
ments and design trends. Atten ion is given to long overtand as ship or barge, road and rail, the variations in costs may differ
conveyors where the influence o economic and echnical con­ by several orders of magnitude. Even when one mode of con­
siderations in he design methodology are indica ed. Aspects of veying, such as belt conveying, is examined for a particular in­
conveyor dynamics are discussed, and research and develop­ s allation within the range of possible combinations of conveyor
ment work concerned with specific subjects including drive size, speed and plant layout, there can be considerable varia­
drum friction, rolling resistance, boos er drives and bulk solid tions in the overall costs. For these reasons it is relevant that the
and conveyor belt interaction are highlig ted. Mention is made conditions for optimum performance of particular types of con­
of other types of belt conveyors notably he Cable Be system. veyors and handling equipment be established.
high lift and pipe conveyors. Other types o mechanical convey­
Various models have been proposed for comparing different
ing equipment are briefly revie eel and, by ay o example,
modes of conveying and transportation. JoNKERS [1] introduced
some research developments into he performance o sere
the concept of the •1oss factor in transport" in which compari­
conveyors and bucket elevators are presented.
sons are made by considering the energy loss per unit distance
for a specific conveying or transportation task. Loss factors
were computed by J KERS for both continuous and discontin­
1. Introduction uous modes of transport, the results being as indicated in Fig. 1.
As shown, the modes of transport involving fluid friction, such
The handling of materials in bulk form is of major concern o a as slurry pipelines for the continuous modes and ship or barge
vast number and variety of industries hroughout the wortd. In­
dustries such as those concerned with mining, mineral and Fig. : Companson of Joss fact()(S in transport
chemical processing, agriculture, and food processing all rely or different modes of conveying (1]
heavily on a need to handle and convey bulk ma erials in one
form or another. In view of the substantial cos s of handling op­ DlSCONTINUOUS
CONTINUOUS
erations, it is most important that hese operations be per­
formed as economically, reliably and efficien ly as possible. 10
Even small incremental gains in efficiency can yield substan ial
cost reductions.
In the past the design of bulk materials handling systems has
often been treated very empirically. Yet over recen years signif­
icant advances have been made in the development of heories
and associated analytical procedures o describe the behaviour
of bulk solids under the variety of states that are encoun ered in Aeroplane
bulk materials handling operations. The same is true of convey­ �-----...-10 -1
ing, where much research has been conducted leading o a bet­
Truck
ter understanding of the mechanism of conveying, and, conse-
Train
2
-------10-
Pro essor Or. AW. Roberts. Director. Ins
Research, Facu ty of Engineen , The U · Ship
Drive. Callaghan, .S.W. 2308. Aus raia..
Tel.: +6 9 21 6067; Fax: +61 9 21 602 . -3
-------+-10
Details about the a

For more ormation on REI 255

circle Reader Serviee Card o. 8
 _.,).: bulk
Mechanical conveyor Design _'$.'l<
;:,:,ig;-.
Volume 14 • Number 2 • April/June 1994 solids
handlin9
�Jt5-

for the discontinuous modes, have the lowest energy loss. Then cal handling equipment are widely used. Notable are bucket el­
come the modes involving rolling resistance, followed by sliding evators, screw conveyors, drag chain conveyors and vibratory
or Coulomb friction and finally pneumatic systems involving air conveyors which have been the subject of considerable re­
and body friction at higher velocities. search and development [64-77].
The 'Loss Factor' data of JoNKERS is useful for comparing the The subject of mechanical conveying is indeed a fascinating one
energy costs of transport, but there are many other factors to be and one which lends great encouragement to innovative design.
taken into consideration. In general, the selection of the most The subject is also very broad. Therefore, it is beyond the scope
appropriate conveying system is dependent on its suitability to of this paper to cover this subject in great detail. Rather the ap­
the task, rather like the old adage 'horses for courses'. Energy proach taken is to focus mainly on belt conveying and including
requirements 1may have a low priority in the overall selection pro­ some discussion of screw conveyor and bucket elevator perfor­
cedure with such factors as effectiveness, maintenance and re­ mance. The paper draws heavily on the research undertaken at
liability assuming greater importance. Furthermore, any com­ the University of Newcastle, Australia.
parison between conveying systems should be based on both
technical and economic grounds, the economic considerations
being correctly formulated using life cycle cost models.
2. The Bulk Solid Conveying Problem -
ROBERTS et al. [2-5] have shown that the energy costs and other Basic Considerations
operating costs cannot be considered in isolation, and design
and performance evaluations of conveyor systems are only The manner in which the strength and flow properties of bulk
meaningful when complete economic studies, based on the solids influence the performance of bulk storage and discharge
total life of the plant, are made. In this work an optimum design equipment is well documented. In the area of conveying and
methodology has been developed in which cost functions are transportation, it is the frictional characteristics of the bulk solid
derived which take into account the energy costs and annual that have a major influence.
equivalent costs computed on a life cycle basis. Once a partic­
ular type of conveying system has been chosen, optimum, min­ Unlike fluids which can be transported by pumping through
imum cost solutions to the conveyor design problem are sought pipelines, bulk solids in their natural single phase state cannot
taking into account constraints imposed by design, manufactur­ be transported in this way. It is well known that bulk solids, due
ing and operational limitations. The optimum solutions may be to their internal and boundary or wall friction characteristics,
implemented directly or used as a 'yardstick' against which the generate shear stresses under static conditions whereas fluids
actual conveyor performance can be measured. do not. The shear stresses generated within a deforming bulk
solid are essentially independent of the rate of deformation;
Of the various modes of continuous conveying of bulk solids, fluids, on the other hand, generate shear stresses at a boun­
belt conveyors are of considerable importance in view of their dary, such as a pipe wall, which are a function of the velocity.
widespread use and proven reliability. Conventional troughed
The properties of bulk solids in relation to the storage and con­
belt conveyors have been used extensively in industry over a
veying problems will now be considered.
long period of time. While their usage is largely associated with
in-plant movement of materials, their application to long dis­
tance overland transportation is now widely established. This 2.1 Janssen Pressure Distribution under
has been made possible through the development of steel cord
reinforced belts, better quality control in the belt manufacture,
Static or Storage Conditions
improved reliability and life through the application of belt condi­ Consider a column of a bulk solid at rest in a vertical cylindrical
tion monitoring and greatly improved design methodologies. container as illustrated in Fig. 2(a). At depth z from the surface
There is a wealth of published literature on the subject of belt the pressure in the vertical direction is cr2 and the radial or lateral
conveying; a selection of papers that illustrate the developments pressure is crr. Owing to the frictional properties of the bulk solid
that have taken place are given in [6-63]. the lateral pressure crr at the section considered is less than the
Notable developments in conveyor design include analyses of vertical pressure. It is assumed that:
belt conveyor dynamics during starting and stopping, belt vibra­ crr
tion analysis and improved models to predict belt tracking under =K (1 )
various loading and conveyor curvature conditions [6-21]. Com­ CTz
binations of horizontal and vertical curves which permit convey­ The factor K depends on the internal frictional properties of the
ors to be effectively integrated into the terrain over which they bulk solid.
traverse are now a proven reality [22-25]. Modern installations
include single flight lengths of 10 to 15 km, increased belt dcr2 4µKcr2
speeds over the range 5 to 9 mis and increased tonnages typi­ -+--=y
dz D
(2)
cally 2,000 to 6,000 t/h with one known installation in Germany
having a capacity of 37,000 t/h. Developments in belt materials
Solution of this equation yields the well-known JANSSEN equa­
include Aramid fibre reinforced belts which offer significant ad­
tion:
vantages due to weight reduction [39-43]. The application of
booster drives to reduce tensions in long overland conveyors is yD _ 4µKz]
a matter of some interest [44-47]. 02 = -- [ 1 -e D (3)
4µK
The versatility of belt type conveyors for long distance transpor­
tation is also being demonstrated by recent installations and where 'Y p g = bulk specific weight
current developments in Cable Belt systems [49-51]. These
p bulk density
conveyors offer single flight lengths up to 30 km and speeds of
6 mis and higher. The versatility of belt type conveyors is further µ coefficient of friction at the boundary wall.
demonstrated through developments in high lift sandwich type
conveyors [52-53], pipe conveyors [54-57], enclosed sus­ The variation of cr2 , crr and 't, as a function of depth z is illustrated
pended type belt conveyors [58], 'Flexowell' conveyors [59-61], in Fig. 2(b). As indicated, the pressure cr2 approaches, asymp-
pocket-belt conveyors [62] and air supported belt conveyors yD
totically, a limiting value of as the depth z increases.
[63]. In process plants, other types of conveyors and mechani- 4µK

256
 bulk , --
solids
�n��
Volume 14 • Number 2 • ApriVJune 1994 ·_ Mechanical conveyor Design

,� D

,: f',
: Hydrostat,ic
; Pressure

2
yrr, D dz
W=
4
, CT •
'tw r f1°z
z
Fig. 2: Properties o pressure is ribution curve. a) Column of

This is in sharp contrast o the hydros a ic pressure which in­ consequently provides a positive feedback effect. That is, the
creases linearly as shown. In the case of the bulk solid, signifi­ greater the vertical pressure o z, the greater the lateral pressure
cant support is provided by the all friction. The hydrostatic ar acting against the walls and the greater the frictional drag be­
pressure is tangential to the J curve at z = 0. comes. Hence, in turn, the greater oz must become. For equilib­
rium the following differential equation applies:

2.2 Forced Flow ·n Vertical Cylindrical Pipe doz K

- -4µ-0z = y (4)
dz D
While the foregoing analysis shows that e frictional properties
of bulk solids offer advantages in the storage of materials par­
ticularly in view of the support given by the walls of he bin or Solving Eq. (4) yields:
silo, these properties have an adverse effec hen bulk materi­
als are to be elevated. These disadvantages are particularly em­ = -yD [e 4 µK b -1] (5)
phasised when a comparison is made with he pumping of 4µK
fluids; here the pressure to be overcome is hat due to the hy­
drostatic head plus he frictional losses due to the pipe walls The conveying force FH for a column of height H is:
and bends, the latter losses being depen ent on he fluid veloc­ 3
ity. FH = µK e o 1 (6)
[. 4µK� ]
-
Consider the problems of forcing a column of bulk solid up a cir­ 1y,r.0
6
cular tube as illustrated in Fig. 3. In this case he shear stress a
the wall acts downward in a direction o posing he motion and Thus the force FH increases exponentially with column height H.
By way of illustration, Fig. 4 shows the force required to initiate
Fig. 3: Analysis of fOf'ced flow conditions
the movement of a column of millet seed up a tube 40 mm inter­
nal diameter. The full nne shows the theoretical curve computed
using Eq. (6); the experimental results are shown as plotted
points. It should be noted that for the longer column lengths,
once flow commences, the force for a particular column length
increases beyond the value indicated in Flg. 4.
For comparison purposes the hydrostatic force is also shown in
Rg. 4. The results 'for the bulk solid clearly indicate that bulk sol­
ids cannot be 'pumped' in a single phase mode in the same
sense as a liquid can. The exponentially increasing force with
column length indicates that for the vertical conveying of bulk
solids the column length must be limited. In practice this means
that continuous conveyors such as the cable and disk conveyor
or the enclosed screw or auger conveyor shown respectively in
Figs. S(a) and (b) employ a series of short columns equivalent to
the pitch of the disks or screw. The conveying force becomes a
linear function of the total conveyor length as indicated in Fig. 6.
By way of illustration the cable and disk conveyor consists of a
series of circular disks attached at equal spacing on an endless
F cable; the cable and disks move through a circular tube carrying
ma erial as indicated in Fig. 5(a).

257
 ,,;,;· bulk
Mechanical conveyor Design ....�·· Volume 14 • Number 2 • April/June 1994 solids
handlin9
• ;_ 'f

350

······································-�--�-
300

250 ! ! ;
Theoretical -------£:JI
Z 200

e
Q)
: i i
& 150 ; : : 1
• Experimental """" :

100 - +VT~3or:a1. . . . . . . . . . .
Fig. 6: Comparison of forces to convey a bulk material through a vertical tube
..
50 veying where the bulk solid is conveyed in the two phase mode.
For the single phase mode it is necessary to transport the bulk
solid by means of a carrying vehicle which has low mechanical
0 friction. Rail or road transport for the discontinuous case and
100 200 300 400 500
belt conveyors for the continuous case meet this objective.
0
Column Length - mm

Fig. 4: Force to convey millet seed vertically through a 40 mm tube

3. Belt Conveyor Research and
The force to move the material on one disk is, from Eq. (6): Development - An Overview

o3 [ 4 h ] In view of the importance of belt conveying to a wide range of in­

!1F = � e µKo - 1 (7) dustries, it is not surprising that the subject of belt conveying
16µK has received considerable attention in terms of research and
development. There has been wide acceptance by industry of
where h = height of material on one disk. the advances being made and this is reflected in the new con­
The total conveying force is: veyor installations particularly in regard to those associated with
long overland transportation. Examples include the 15 km Selby
H coal drift conveyor in the U.K., the 11.2 km nickel ore conveyor
Fe = !1F­ (8)
p with four horizontal cuNes in New Caledonia, the Channar iron
ore conveyor in Australia comprising two conveyors of 10.4 km
where H height of lift
and 10.1 km respectively and with one horizontal cuNe, and the
p pitch. 13.1 km coal conveyor of PT Kaltim Prima Coal, Indonesia.
The actual conveying force will, of course, be larger than that The following list highlights some recent and current advances
given by Eq. (7) in order to overcome such losses as churning in in conveyor research and development:
the conveyor intake.
1. Conveyor dynamics - starting and stopping characteristics
The forced flow characteristics of bulk solids can be greatly im­ and dynamic belt stresses.
proved by reducing both the internal friction of the material and 2. Transverse vibrations of conveyor belts and the associated
wall friction. This is the basis of hydraulic and pneumatic con- influence of idler spacing and troughing configurations on
such vibrations.
3. Conveyor belt idler resistance taking into account belt rub­
Fig. 5: Cable and disk and screw conveyors

ber hardness, sag, troughing configuration, idler/belt inden­

tation and ambient operating temperature.
4. Conveyor transition geometry.
Fig. 7: Conveyor design model

I
(a) Cable and Disk (b) Screw Conveyor
Conveyor

258
 bulk
solids
han�lins
Voiun1e 1..: • Number 2 • April June 1994
T
nlcal conveyor Design

5. Stability of bulk solid on conveyor be s during mo ion on 3.1.2 Belt Resistance and Power
horizontal and vertical curves.
6. Conveyor be drive drum friction t ·ng 1n o account rub­ F=Lli (1 1)
ber hardness, surface roughness rap angle. •=1

7. Steel cord splice design and analysis. here � are the various resistances such as empty belt frictional
resistance, load resistance, slope resistance, special and local­
8. Conveyor belt monitoring as applied o s eel cord and fab- ised resistances such as those due to skirtplates and belt clean­
ric belts ers.
9. Belt rip detection employing ultra sonic devices. As a first order approximation, it may be assumed that the fric­
tion coe cients in the idlers and drive components are constant
10. Belt tension monitoring during operation.
and inde ndent of the velocity. It is then apparent that the total
1 1. Belt cleaning including carry back measurement and devel­ resistance to motion in functional form is:
opment of improved cleaning e 1ciencies.
F = f1 (8 L p (1 2)
12. Economic analysis applied to conveyor design.
where B = belt width
13. Booster drives to reduce belt tensions and permit longer in­
dividual flight length L = total belt length.

14. Design of horizontal curves The conveyor power is given by:

15. Improved quality control in be and component manufac- Fv

P=- ( 1 3)
ture. T\
Recent research has shown that belts ranging in hs from
800 to 1,200 mm are the most suitable. In this way, bet ten­ where Tl = drive efficiency.
sions are kept o acceptable limits allo ·ng longer, individual
conveyor lengths to be realised for he given range of bel SR 3.1.3 Peak Belt Tension
values commercially available. For large tonnages, the use of
belts within the above-men ioned range of · hs, but running For a given arrangement, the tight side tension F1 based on the
at higher speeds provides the best solution. ile these trends simplified tension distribution around the drive drum(s) is given
by
are supported on the grounds of economics, they are also sup­
ported on technical grounds. ith regard o belt speeds he
economic evaluations clearty highlight he advan ages o em­ F1 = [�] F (14)
ploying speeds greater than 4 mis here large tonnage e µe - 1
throughputs are involved.
This assumes the well-known tension relationship
The foregoing observations may be demonstrated as folio
F,
- = e µa ( 1 5)
3.1 Basic Design Considerations hi
where F2 is ·stack• side tension.
The application of •steady-state· design heory provides a use­
The m imum belt tension under steady-state operation will be
ful basis for the study of he firs order in uences of such factors
the larger of that given by Eq. (14) and that required to limit the
as belt width, strength rating, factor-of-safety and cos s for par­
m ·mum belt sag in the low tension zone of the belt. When dy­
ticular conveying tas s. While the design equations and heory namic effects are taken into account, the maximum belt tension
are well known and documented, for the purpose of he present will also include the influences of transverse and longitudinal belt
discussion, it is useful to brie revie he basic procedures. vibrations. For a •first-order- approximation, the maximum ten­
Referring to Fig. 7, the following basic design equations ap sion in functional form is expressed by

3.1 .1 Throughput Q (1 6)

Q =pA V (9 This equation indicates that the maximum belt tension for a
p = bulk dens· of bulk solid given length conveyor is directly related to the belt width. The
where
narrower he belt, the lower the tension. However, the interrela­
v = belt ve1oc· tion of belt width and conveyor speed needs to be taken into
account when establishing the most appropriate parameters to
•N represents the cross-sectional area of he bulk solid on he
meet the throughput requirements.
belt and is given by:

A =U b
2
(10 3.1.4 Beft Rating and Width
For a computed maximum belt tension, the required belt width
where u non-dimensional cross-sectional area is given by:
shape factor
= contact or • etted· perime er. Fma fs
B=-
b - (1 7)
SR
The belt width B > b to allow for edge e ects.
Shape factors for various idler troughing configura ions are here f5 factor-of-safety
given in [3]. SR = maximum strength rating of belt (in kN/m).

259
 >
�.·.
';
bulk
Mechanical conveyor Design Volume 1 4 • Number 2 • April/June 1 994 solids
u handlins

3.2.1 Belt Width and Velocity Considerations

Figs. 8 and 9 show the annual equivalent cost per unit length for
l= 5 l<m l � various throughputs for a conveying distance of 5 km. In Fig. 8
the costs as a function of belt width are presented with belt ve­
p =0-85 trn

locities curves superimposed. The graphs show that the lowest

costs occur in the range of belt widths 0.6 to 1 . 2 m with 1 m
being a 'good ' average value. In Fig. 9 the cost data are plotted
as a function of belt velocity with belt width curves superim­
posed. As indicated , the annual equivalent cost decreases sig­
nificantly at low belt velocities but, as the velocities increase, the
cost is substantially constant or increases very slightly.

100 3.2.2 Belt Length Considerations

The potential of belt conveyors for long distance transportation
0 -1---..------.--.--.--.....--..----+ is illustrated in Fig. 1 0. Here the annual equivalent cost per unit
04 0-6 o-a 1-0 1-2 1·4 1-6 1-a �o length is plotted as a function of conveyor length for various belt
Belt Wid th B (ml widths. The corresponding conveying velocities are indicated
and the belt SR curves are superimposed.
For conveyor lengths beyond 1 km, the cost per unit length in­
Fig. 8 : Annual equivalent cost per unit length a s function o f belt width.
L = 5 km, p = 0.85 t/m3 , f5 = 7 : 1
creases as shown, this being due to the need to employ belts
with higher SR values. For this reason it would be nominally
more economical to use several shorter conveyors in series than
fewer longer conveyors for conveying over long distances. This
could well be the case when wide belts are used. However, the
L= S km 3
= 085 t/m
cost per unit length advantages associated with shorter belts
....."' would be offset by the cost of the transfer stations. Hence it is
recommended that, in general, longer individual belts be used
and this reinforces the need to employ narrower belts.

3.3 Dynamic Considerations

The two most important dynamic considerations in the design

100 of belt conveyors are:
(i) Reduction of starting and stopping stresses in the belt.
2 4 6 8 W
Belt Velocity v (f!l/s)
U . � � (ii) Running stability or resonance-free operation.

3.3.1 Dynamic Belt Stresses -

Fig. 9: Annual equivalent cost per unit length as function of belt velocity. Starting and Stopping of Conveyors
L = 5 km, p = 0.85 t/m3 , (5 = 7 : 1
Numerous methods have been devised for starting belt convey­
ors, including eddy current drives, switched resistance starts,
3.2 Conveyor Economic Analysis
fluid and fluid-scoop coupling drives and controlled starting for
Economic studies based on life-cycle cost analysis have been AC and DC motors. Most of these methods have discontinuities
conducted to identify the most cost effective operating condi­ in the drive acceleration at the instant the torque is applied to
tions for conveyors [2-5] . Relevant aspects of these studies are the belt, leading to travelling elastic waves in the belt [4]. Meas­
highlighted below. While the cost data are based on 1 985 val­ urement of starting tensions in belts containing shock waves in­
ues, the relativity with respect to current costs would be ex­ dicates that tensions in the belt may exceed ten times the high­
pected to be the same. est operating tension, and this has been the reason for the high
safety factors that have been adopted in the past. Large safety
Fig. 1 0: Annual equivalent cost per unit length for horizontal steel cord belt con­
factors increase the cost of belting manufacture.
veyors. Q = 1 ,000 t/h, p = 0.85 t/m3 The control of belt speed during the start-up and running period
is most important in higher speed belts if shock is to be avoided
and if lower belt safety factor "sf' is to be employed to reduce
installation and driving costs. During the braking period, similar
700 ___,___.__..___.___.___.___.__..,____.__-+

considerations apply; however, the concept of emergency stop­

ping must be reviewed when belt speed is increased . Naturally,
all safety requirements must be much more closely observed
near the running belt and at the loading and discharge stations.
The criteria for minimising transient stresses in conveyor belts
during starting and stopping particularly applies to fast belts.
The dynamics of starting and stopping have been examined in
some detail by HARRISON [8]. He showed that an optimal "stop
start" is given by the belt start-up velocity.

3 4 5 6
Conveyor lel)gth {Kilometre�)
7 8
v (t) = V; [1 - *] for < T

= vb for t > T (1 8)

260
 ��
solids Volume i ..; • Number 2 • April/June 1994 hanlcal conveyor Design
handling
,•

where vb = belt running velocity

T = run-up period.
The form of Eq. { 8) is shown in Fig. 1 1 . rth modem computer
control sys ems, it is possible o genera e the desired s arting
characteristic.
A general formula for the factor-of-safe has been proposed. = Vb ; tor t � T

f.o.s. ;, 3 ( 1 19)
T

where Td is the non-oscillatory dynamic belt ension during ran­ Fig. 1 1 : Ideal belt velocity characteris IC to minimise transient tensions at start
sient periods, I.T8 are additional ensions due to ransitions. im­ stop
pact and vertical curves. The peak ension T along the bett is
determined from conventional design me hods such as DI CO NVENTIONAL QEStQN
221 01 or ISO 5048. PROGRAM
Flud O d e s i gn

It needs to be noted hat he above e ression IS somewhat id­ IN PUTS

C a p a c i t y, Ma teria l, Lenglh,
A

Route, Idler size;

ealised inasmuch as many other factors need to be taken into • C E M A. . GOODYEAR,
----'----'-
s r a r t lng/S topplng 11 e
I DUNLOP, CIOUTH.
1
account. These factors include splice e ciency, belt quali and
,-- _,
01 22 1 0 1 , 1505048 HANDBOOK

tracking characteristics.
Tensions, power.
I ..----;;;,.j DESIGN
velocity, sf I
size,
L _ _ _J
belt

3.4 Conveyor Design Program IP ROOAAM

A dynamic analysis program shown in Fig. 2 uses the resul s of I !!

conventional handbook procedures and modifies the design in (Robert• e 1al}
I (rat.37)
the light of computed dynamic stresses. The value of Td critically
depends on the starting or stopping characteris ic of the mo or I COSTING I
drive or brake. An over-rated belt is progressively de-rated in L _ _ _J
this programme un ii a suitable . o.s. IS achieved. Typically a
,c-
f.o.s. � 3.3 is obtained and he be s reng h and hence cos is fc; - 7
STAESS
significantly reduced. In the case where a s eel cord belt is se­ I PAOGRA
lected, the weight is reduced in he selec ion process, leading to £
further reductions in tension and po er. CALC.

The design cost is computed each time he sta ic design is

modified in the programme loop. Though he flow-chart of Fig.
12 does not show a different pa h for calculating .o.s. for start­ I
ing or stopping, the largest average acceleration in he transient
IS DESIGN
YES VEA -RATED
period should also give the largest dynamic accelera ,on in the L _ _ _ _ _ � _ _j
belt. As a result, the greatest rate-o -change of be ve oc· D E S IGN AND

must be employed in f.o.s. calcula ions. BELT DETAILS

- ngth (SA-Steel cord)
TO PAOGA .Q ( k g/m w i d t h )
-Weight
DY A IC AN ALYSIS PROG
3.5 Resonance Free Design
Elimination of belt resonances by the correc design of idler
Dynamic analysis flow chart
spacing leads to increased idler life and reduced bel failure. To
Ftg. 1 2: or evaluating the starting and stopping
safe factors 1n conveyor belts, in conjunction with conventional design
illustrate the influence of idler spacing on belt vibration, a simpli­
fied, one-dimensional version of the more comple vibra ion where m = mode number for the direction along the
problem is considered. Fig. 1 3 depic s the belt vibration. belt (the transverse mode 'n' is assumed
For a flexible, flat belt of low s iffness in relation o the ension in zero in this case)
the belt, an approxima e solu ion to he ransverse vibra ion d, = idler diameter
along the belt is given by the s ring equa ion. v belt speed
The resonan frequencies are given by K5 = static sag ratio = ylam
y static mid span deflection.
m = 1 . 2 . 3 . . . . ( Hz) (20
Fig. 1 3: Belt Vlbrabon problem

Assuming the idlers have a single eccentricity lobe, e e citing

m = 1 (fundamental}
frequency generated by the idler is V

2 (2nd harmonic)
f1 = - Hz) 21 _ .._ (� !-
d,
Hence for resonance, f. = fm ; the critical idler spacing am is
therefore,
3 . 03m di2
2

am =
---­
2
m = 1 . 2. 3, . . . (22
v Ks

261
 'r.../Jt. bulk
Mechanical conveyor Design 1·, : Volume 1 4 • Number 2 • Apnl/June 1 994 solids
�'!- handlin9

This shows that the critical idler spacing a m varies directly with
6 -r----------------, di2 and inversely with v2 and Ks. Fig. 14 indicates the variation of
j BELT VELOCITY = 3 mis j
idler spacing a m versus Ks for di = 200 mm, belt velocity v = 3
- s m/s and for m = 1 and m = 2. It is important that the actual idler
s
spacing is chosen to avoid the zone of, say, ± 15 % of the criti­
cal spacing given by Eq. ( 23).
(!:) 4
z
� 3 The actual belt vibration problem is more complex than the sim­
plified case described above. The transverse modes of vibration
0..
Cl)

are two-dimensional. Furthermore, the stiffness of the belt can­

a: 2
w
not be ignored and the belt properties are orthotropic. A further
complicating factor is the troughing configuration of the belt. A
0 -+----.----"""T""---..--------1 detailed analysis of the transverse vibration problem has been
0.01 0.02 0.03 0.04 0.05 given by HARRISON [8] who derived expressions for the frequen­
SAG RATIO Ks cies fmn ' where m and n are integers with the values 1,2,3... and
refer to the vibration modes along and across the belt respec­
Fig. 1 4: Critical idler spacing for resonance based on simplified analysis tively. Plate mechanics are used to determine these frequen­
cies. Some of the modes of vibration are illustrated in Fig. 1 5 .
The running belt tension profile is required in order to design
idler diameter d, idler spacing a, belt width b and belt speed v,
so that the excitation frequency due to the idlers � is less than
the natural frequencies fmn · The excitation frequency � will de­
pend on the profile of the idlers which indicates whether single
or multiple eccentricity lobes are present. Usually only single
lobe idler eccentricities occur in manufacture.
Conventional static design, as distinct from dynamic design,
only uses loading, tension and belt sag ( including take-up coun­
terweight masses) to obtain idler spacing. In dynamic design,
(a) First two transverse modes - Flat return
the belt width is selected and then fm n and � are calculated
[8, 11 ]. The belt speed and idler diameter, together with belt
width and throughput, are adjusted and the dynamic design is
repeated until � < f1 1 and the design is complete. The belt fre­
quencies are determined for both loaded and unloaded condi­
tions to obtain design ranges. Generally, for very high-speed
designs near 15 m/s, belt width is b - 0.8 m, a - 2 m and idler
diameters are d - 0.20 m. Fig. 16 shows how transverse belt
resonances may be eliminated by idler spacing considerations
between loaded and unloaded conditions.
n=2

More recent research by HARRISON [26] has demonstrated the

(b) First two transverse modes - V - return

usefulness of Chaos Theory for analysing conveyor belt dynam­

Fig. 1 5: Some transverse vibration modes of belt conveyors ics.

Fig. 1 6: Diagram showing the removal of resonance zones by the application of

change of idler spacing in the resonance zone
4. Belt Idler Resistance and
Idler Bearing Performance
EOUAL IPbEH §PACfNQ 4.1 Idler Resistance
The determination of idler resistance has been based, tradition­
ally, on the application of "artificial friction" coefficients which are
quite empirical. More recently, research has been undertaken to
Zone 1 Zone 2 examine the nature of belt and idler rolling resistance from a
I• •I' more fundamental viewpoint. This is of particular importance in
"'"
S lable
the case of long overland conveyors where the cumulative influ­
R Honant
,t
ence of idler resistance demands an accurate determination of
the individual resistance contributions.
One study performed by HARRISON et al. [27, 28] has examined
the various factors influencing idler resistance. Factors taken
unIoaq1g into account are:
Zone 2 Zone 3
1• •I (i) Belt sag
S l able
( ii) Idler diameter
( iii) Indentation effects
( iv) Belt hardness
(v) Temperature
( vi) Idler configuration.
Zone
-
I• •1• Experiments using a special test rig to simulate the conveying
REDUCED IQLER §PAQINQ IN ZONE 1 ANQ 3 action showed that the idler resistance is somewhat insensitive
to idler diameter; all other parameters have an influence.

262
 bulk ·.
solids
11an.iun,
Vultu11f• 1 i • Number 2 • Apnl•June 1 994 ..- llanlcat conveyor Design
,_

The total resistance due to the cany idlers has been shown to
be Bearing Type LB-10 Life Test Failure Mean Power
Revolutions Mode Consumption rNJ

FRC = (µ, µ )N Fae (2 3 Seize Resistant 1 .40 1 o& Wear 28.0

where indentation c-oe cient depending on

(SKF 20205)

rubber hardness 0.09 X 1 08 Seizure 44.0

sag coefficient
= flexure coefficien 0. 1 4 1 08 Seizure 35.7

FBC bearing losses of carry side idlers

N normal load on idler set. 0.02 X 1 08 Seizure 61 .0

The components of the totaJ resistance are defined below:

1 1 910

0.01 1 3
µ, = 0 . 0 22 - -- H - 5 5 2
3.33 for roller bearings
=
32
(The load units for C and P are the same).
An appropriate linear expression for he hardness H as a func­ It is evident from Eq. (28) that, in the case of idler bearings for
tion of temperature is belt conveyors, the life is inversely proportional to the belt
speed. However as indicated by TALKS and KENNY [29] of the
H ( T = -0 . 5 T 80 25 former ational Coal Board (NCB) (now British Coal) U. K., the
where H = hardness - Shore A degrees limiting factor for idler life is the grease life rather than the 8-1 O
fatigue life. For instance, TALKS and KENNY estimated that the
T = temperature 0c. CB had around 1 1 x 1 06 idler bearings in operation at any one
The belt sag friction factors µs for s eel cord, fabric and PVC time wrth load ratios GIP ranging from 1 3.8 to 41 .9; the corre­
belts are 0.0033, 0.0066 and 0.005 respectively, for a bet sag sponding 8-1 0 lives range from twelve years to 360 years,
of 1 %. The sag fric ion along the whole return conveyor from based on continuous running at 400 rev/min. Yet, the average
head (x = 0) to tail (x = �. is computed using grease life for idler bearings is five years for operation at 50 °C
and o years at 70 °C.
* [ 1 - ff A 26 Much progress is being made to extend idler bearing life, and
he work reported by HERRAlY and BRAS (30] concerning a SKF
seize resistant bearing is worth noting. The seize resistant bear­
where ing was compared under dry abrasive test conditions, with

A = -; L - - a· · B= ( 1 -A ; = sag friction factor a x

hree other bearings, a standard deep groove ball bearing
(DGBB) a DGBB with increased clearance and a taper roller
bearing. The results are shown in Table 1 .
An average value of the sag for he whole Ieng h o the conveyor As indicated, the seize resistant bearings had a B- 1O life ten
is µs/2. Experimentally the combined fnct,on coe cient µ has imes that of the DGBB with increased clearance, 15.6 times
been measured, where that of the standard DGBB and 70 times that of the tapered
roller bearing. The mean power consumption during all tests
+ (2 7 ) showed the seize resistant bearing to have the lowest power,
being 64% of that for the standard DGBB, 78% of that for the
The flexure coefficient � has been measured indirectly by this DGBB with increased clearance and 46% of that for the tapered
approach. I is obtained by subtrac ing , + µJ from . Other bearing. The test failure mode for the seize resistant bearing
studies of idler friction conducted in he eld have generally was by eventual wear without seizure, whereas the other three
shown that artificiaJ fric ion coefficients as lo as 0.01 1 are no bearings all failed by seizure.
possible. In view of the cumulative resistance effec of idler fric­ A more recent paper by HERRA1Y and HAMBLIN [31] has reported
tion along the conveyor, it is particularly important to ensure hat on the service life caJculations of idler bearings. Continued test­
idler friction coefficients are as low as possible. ing of the seize resistant bearing has confirmed the superiority
of his e of bearing under arduous operating conditions in
dirty environments.
4.2 Idler Bearing Life
The rolling fatigue life of a property lubrica ed bearing is ex­
pressed in terms of the 8- 1 0 life rating. For a sample of nomi­ 5. Conveyor Belt and
nally identical bearings the 8-1 0 life, expressed in millions of Drive Drum Interaction
revolutions at constant speed or hours of operation, is the life
that 90% of the bearings in the sample · 11 attain or ceed be­ An important aspect of belt conveyor operation is the frictional
fore the first signs of fatigue are eviden . The B-1 O life is e - resistance generated between the belt and drive drum. The re­
pressed by: sistance controls the development of drive tensions. In accor­
10 �
0 dance with Fig. 1 7, the well-known tension relationship (referred
La - o � [ J (2 8 to in Section 3.1.3, which is widely used in design, is
6On P
F1
- = e µa (29)
where n rotational speed On rev/min} F2
C basic amic load rating (as gJVen in where F, = tight side tension
bearing catalogue) F2 = slack side tension
p actual bearing load = wrap angle
u 3.0 for ball bearings µ = coefficient of friction between belt and drum.

263
 ,j·- bulk
Mechanical conveyor Design ;...,, Volume 1 4 • Number 2 • April/June 1 994 solids
'-�-::.. ,,. handlins

F + dF

µ dN
� �
de
2

Fig. 1 7: Tensions generated on drive drum

Friction diagram Tension diagram

Some slip

Fig. 1 8: Typical belt friction and tension diagram for combinations of slip, creep and passive zones

It is usual to assume that µ = 0.25 to 0.3 for a steel drum sur­ 5.1 More Recent Research
face and µ = 0.35 for a lagged drum surface. The principal as­ A program of research being conducted at the University of
sumptions embodied in Eq . (30) are: Newcastle is aimed at providing a greater insight into belt and
(i) the frictional resistance is strictly of Coulomb type drive drum friction [32 , 33) . To date, the research has involved
experimental studies using two separate test set-ups.
(ii) the friction is developed around the contact surface of the
(i) Direct Shear Apparatus
drive drum
(ii) Drum Friction Apparatus (shown in Fig. 1 9).
(iii) slip (or pending slip) occurs around the contact surface
The Direct Shear Apparatus i s based on the Jenike Direct Shear
(iv) eccentricity of the surface defining cord location (and hence Tester. For a selection of belt and drum samples, records of
tensions) and drum surface is neglected.
Fig. 1 9: Drum friction apparatus
These assumptions are not strictly valid and values normally as­
signed to µ in Eq. (29) are not actual friction coefficients but,
rather, empirical values which are lower in magnitude than the
actual values. Although it is never stated, the reduced empirical
friction coefficients used in Eq. (30) provide for a conservative
design; they compensate for the fact that unless slip occurs
around the contact surface, the friction may not be fully devel­
oped.
It is quite clear that the eccentricity, ec , in Fig. 1 7 causes inter­
nal shear to be developed in the rubber as a result of the mo­
ment generated. It is also known that the angle of wrap may be
divided into combinations of active slip, creep and passive
zones as depicted in Fig. 1 8 . This causes variations in the ten­
sions as indicated.

264
 bulk ·__·-.; .
solids Volume 1 4 • Number 2 • ApriVJune 1 994 Machanlcal conveyor Design
handling 1-:.
_,

i
The influence of wrap angle has been examined using the drum
x - k2 x I friction apparatus shown in Fig. 19. Preliminary results indicate
s hat the tension relationship does not follow the exponential
- lc.
S(x) = S0( 1 - e ·, )( - e
form of Eq. (30)· a linear relationship of the form given below has
B 1 (94' Shore A) been indicated.

F
-=1 ca (32)
Fi
here C = constant.
On he basis o he above relationship, the equivalent friction de­
creases with wrap angle indicat,ing that the friction coefficient is
inversely proportional to the tension. However, these findings
are somewhat tentative and further tests are necessary before
any firm conclusions can be obtained. More recent experiments
carried out over a wider range of wrap angles confirmed the
g.eneral exponential form of Eq. (30) but with significantly higher
Shear + Slip values of the friction coefficient µ than those values of 0.25 to
Adhesion 0.35 currently used in design. These lower design values are
really friction factors, not friction coefficients; they permit Eq.
F,g. 20: Shear force versus shear de ormation curves (30) to be used with an in-built safety factor which allows rea­
sonable estimates to be made of peak belt tensions. In effect,
the lo er design values compensate for the fact that the friction
shear load and shear deformation were ob ai ed or a range of
is normally not fully developed around the total wrap angle of the
normal1 loads. By way of examp e, he fully developed friction for
belt on the drive drum.
steel on rubber has been shown to depend on the surface
roughness of both the belt and steel surfaces and, in particular, Current research is concentrating on finite element analyses of
on the rubber hardness. The shear force versus deforma ion he s ress conditions in belts wrapped around drive drums; the
curves are characterised by rubber hardness as indicated by results are providing greater insight into the nature of belt and
Fig. 20. The results relate to hree belt rubber hardness values drum fric ion.
examined, namely 66° , 75° and 94° Shore A.
Tests have shown hat the fully developed friction is given by
6. Bulk Solid and Conveyor Belt
30 Interactions
It is important that. during the design stage, careful considera-
where ion be given to the interaction between the conveyor belt and
the bulk solid being conveyed. Interaction effects relate to such
31 aspects of conveyor operation as the stability of the bulk solid
during transportation and the discharge characteristics. The lat-
H0 = 66 C0 = 1 .0 er concerns the design of the discharge chute and the belt
c1 = ·o c2 =-2.9 cleaning system.
C3 = 4.37 C =-0.8
R8 = roughness of steel drum. 6.1 Stability of Bulk Solid on
Belt During Conveying
The influence of rubber hardness is illus ra ed in Fig. 2 1 which
shows the comparison be een he measured and predic ed The loading of bulk solids onto conveyor belts is governed by
values of the friction coefficien he need to ensure effective transport of a maximum payload

Fig. 2 : Compariso be precfic ed and exl)E!flmE!ntal

::1. 1 .8 -,------;_-:._-_
1- 1 .6
-
I
_-:....-:....-:....-:....-_-:....-:....-_-:....-:....-:....-_
fl o □ Experimental Values I
-_-:._-_-:....-_
---.
_

.z
5 1 .4 Steel
Surface
Roughness
Ra (µm)
U:: 1 .2 -
u.
8
w
1 .0
S1
S2
OAS
5
0.75
0.8 S3 1 .03
i= 0.6
o
a:
u. 0.4
60 70 80 90 1 00
RUBBER HARDN ESS (Shore A deg.)

265
 _,i'l bulk
Mechanical conveyor Design ;J1}_-· Volume 1 4 • Number 2 • Ap11I/Ju11e r n�J,I solids
h.11ndlln9
.. -�,;

(20, 30, 35 and 45 ° ), and 3 roll and 5 roll suspended trough­

ing idlers. Idler spacing is variable from 1 ,000 - 2 , 000 mm.
(iii) Simulated belt speeds ranging from 0 - 1 5 m/s are possible
via a 3 hp AC geared motor unit fed by a variable frequency
electronic drive.
(iv) Belt sag and/or belt tension is controlled through two hy­
draulic jacking cylinders. The cylinders allow the belt to be
adjusted to a present tension, or until a required percentage
sag (either for coasting or loaded) is achieved.
(v) Frequency and amplitude of belt vibration is to simulate ma­
terial on an actual belt passing from a position on top of an
idler set down to the maximum sag position (midway
between consecutive idler sets) and finally to the top of the
next idler set.
The cam drive which controls the amplitude of vibration is ad­
justed so that the loaded belt moves between the maximum sag
position to level with the belt clamping and tensioning brackets
(that is, the top of the next idler set). The frequency of vibration
is set to the idler passing frequency which is a function of belt
speed and idler spacing.

Fig. 22: Schematic isometric drawing of the conveyor simulation test rig {left half Although the rig does not currently allow the idler rotating fre­
shown, right half similar). Note: belt and variable speed geared motor quency to be superimposed , it is believed that whilst the fre­
omitted for clarity quencies are higher, the amplitudes are so small that the effect
of the rotating idlers will be negligible compared to the idler
passing motion described above.
without spi llage. This requirement becomes particularly impor­
tant in the case of long, overland conveyors which execute a 6.1 .2 Typical Test Results
combination of vertical and horizontal cuNes. In order to gain
some insight into the load stability during transportation , a con­ A range of tests have been conducted over a period of time to
veyor simulation test rig was designed and constructed at the simulate the motion of bulk materials, notably coal and iron ore.
University of Newcastle, Australia [34). The test rig, which is il­ In the case of the latter, one series of tests was concerned with
lustrated in Fig. 22, comprises three main components: the stability of iron ore on the horizontally cuNed section of an
(i) Conveyor belt clamping and tensioning mechanism overland conveyor. By way of illustration, a typical set of results
are shown in Fig. 23.
(ii) Cam driven troughing idler unit to simulate the belt and ma­
terial passing over consecutive idler sets; and The results relate to a 35 ° idler set, the conveyor rig being set at
a horizontal cu Ne super-elevation angle of S 0 • The iron ore was
(iii) Main support frame incorporating a two-way tilting base to
accommodate vertical inclines and/or the superelevation of at a moisture content 0.S% (wet basis) and of a particle size of
horizontally cuNed conveyor belts. 90% minus 1 6 mm. As the results show, there was very little
movement of the iron ore.
6.1 .1 General Specifications (i) Fig. 23(a) applies to horizontally cuNed conveyor at a super­
elevation angle of S 0 with symmetrically positioned belt on
The general specifications of the test rig are: 35 ° troughing idlers.
(i) All belt types with belt widths up to 1 ,050 mm; (ii) Fig. 23(b) applies to horizontally cuNed conveyor at a super­
(ii) Carry idlers to any diameter and specification. Idler types in­ elevation angle of S 0 with the belt offset 400 mm down the
clude flat carry idlers, 3 roll in-line and offset troughing idlers 35 ° troughing idlers.

Fig. 23 (a) and {b): Change in surcharge profile during simulated conveying motion

4 on-.-------------, 400
(a) As Loaded
(b)
/ After Vibration
200

0-1---...---.:�-.......----,-...c----.!r=_J 0-1-----.........--..3,,,----....-�---i=-:...:..:.:..�

0 200 400 600 800 0 200 400 600 800

Elevated Edge of Belt Elevated Edge of Belt

266
 -
bulk
,I'

solids Volume 1 4 • Number 2 • ApriVJune 1994 -_ Mechanlcal conveyor Design

mnalir,s

6.2 Bulk Solid Discharge and

o·
Carry- Back After scharge
6.2.1 Condition Occurring at Discharge
An indication of the discharge trajectories and possible levels of
bulk solid carry-over following discharge may be ascertained by
considering, in a simplistic way, the mechanics of separation of
bulk solids from he belt as it passes around the discharge pul­
ley [35, 36) . .Rg. 24 shows the condition at the conveyor dis­
charge.
The equilibrium condition for the position at which discharge
commences is given by

v
2

(3 3 )
Fig. 24: Conveyor bett discharge
- = COS8
rg pg h

here v = velocity of bulk solid leaving belt (mis)

p = radius of bulk solid measured from drum
centre (m)
g = gravitational acceleration
9 = angle at which discharge commences
Wol l F ri c t i on crO = adhesive stress (kPa)
An g l e p = bulk density (Vm3)
h = thickness of bulk solid stream (m).
The adhesive stress, depicted in Fig. 25, is obtained from the
00 Surface Yield Locus measured in the laboratory.
Tension Compre s s · o n
◄ ► Referring o Ag. 25, the average height of material on the belt is
given by
F,g. 25: Surface yield locus for conveyor belt
Q
h = -­ (3 4 )
pvb

ere Q = hroughput (t/h)

Fines (Hig Cohes on) b = width of contact of bulk solid on belt (m).

Be Discharge will commence at the line of intersection of the belt

and drive drum, that is, for 9 = O. The minimum belt velocity for
0 · e Orum
(D scharge)
this o occur follows directly from Eq. (33) and is given by:

Fig. 26: But material on belt a cfischarge (3 5 )

Frg. 27: echanics o discharge and adhesion

A m v2

( n ) B e l t D i s c h a rg e
Pul l ey
( c ) A dh e s i o,n

267
 ,;.·-·
,c;;- •� bulk
Mechanical conveyor Design �_i.,- Volume 14 • Number 2 • April/June 1994 solids
handllns
•·.-=

By way of example, consider the belt conveying of bauxite.

While the bulk density varies with the consolidation pressure, it
is assumed that p = 1.2 t/m3 corresponding to the likely consol­ era 1000
h = -
idation condition ( from flow property data). Measurements of
surface yield loci of bauxite indicate that adhesive stresses up to
p -r-
- +g
( 38)

1 kPa may occur. Assuming a value of h = 0.09 m and a radius R

of R = 0.6 m, then vmin = 3.38 mis. Hence, if the belt speed By way of illustration, values of cr0 as high as 1.4 kPa for iron ore
v > 3.38 mis, then discharge will commence at the line of inter­ at 1 1 % moisture content on conveyor belt surfaces have been
section of the belt with the drum. measured. However, it is usual for values lower than this to
occur.
6.2.2 Carry-Over After Discharge - Assuming, for example, that cr0 = 0.1 kPa and density p = 1 t/m3
Conveyor Belt Cleaning ( as for coal), then crip = 0. 1 . From the foregoing, it follows that
The mechanics of discharge may be studied by reference to Fig. for a belt travelling at 3 mis and assuming a 1 m diameter dis­
27. Assuming the bulk solid does not slide relative to the belt, charge pulley, bulk solid will adhere to the belt in thicknesses up
discharge takes place as discussed previously. For cohesive to h = 5.56 mm just before the primary scraper ( Eq. ( 37)). This is
bulk solids, sliding is unlikely in view of the very high friction equivalent to a potential bulk solid carry-over of 60 t/h. Without
angle that accompanies the low normal pressures between the a primary scraper in place, then h = 3.6 mm at the bottom of the
bulk solid and the belt. pulley ( Eq. ( 38)), the corresponding carry-over being 38.9 t/h.
Values such as this give an indication of the amount of bulk solid
After the point of discharge defined by Eq. ( 33), some bulk solid to be handled by a belt cleaner.
will adhere to the belt as carry-over. This is an adhesive condi­
tion that arises as a result of negative or tensile stresses crnt Effective belt cleaning depends on an appreciation, at the de­
which occur between the bulk solid and the belt surface. The sign stage, of the amount of likely carry-over of bulk material
limiting tensile stress is the adhesive stress cr0 . While adhesive after discharge. It is unlikely that a single cleaning device will be
stresses are very small, they can have a significant influence on sufficient and a multiple cleaner system is recommended. An ef­
the amount of carry-over that may occur. An indication of the fective belt cleaning system should reduce carry-over to at least
thickness h of the carry-over may be obtained as follows: 100 glm2, which is equivalent to h = 0. 1 mm for p = 1 t/m3, or
1 .08 t/h on a one metre wide belt travelling at 3 mis. It is pos­
For an arbitrary location defined by e > e l , where e l is the angle sible to achieve much better cleaning efficiencies down to even
at which discharge commences, it follows that the thickness h 1 0 glm2. Apart from the selection of the most appropriate
of the carry-over is given by cleaning device, it is important that sufficient attention be given
to the need to ensure effective removal of bulk material without
h =
00[ 1 000 ]
(36)
build-up on the cleaner components.
p i - g cose
where h = carry over on belt ( mm) 7. Other Research Developments into
cr0 = tensile stress ( kPa) Belt Conveying
p = bulk density (t/m )
3
While it is beyond the scope of this paper to discuss, in any de­
V = belt velocity (mis) tail, other areas of research into belt conveying, it is of interest to
R = pulley radius ( m) note areas in which research is presently being conducted.
Q = gravitational acceleration.
At the position of the primary belt scraper, e = 90° and Eq. ( 37 ) 7.1 Belt Tracking and Dynamic Tensions
becomes
Two of the more important parameters needed by the designer
h 1 0 0 0 cr0 R
(3 7 ) of belt systems are the dynamic tracking characteristics and dy­
namic belt tensions that occur on starting and stopping. Both of
= 2
these parameters are difficult to predict during the design
p V

At the bottom of the pulley before the secondary scraper, e= phase. There are no theories to predict belt tracking character­
180° , and Eq. (37) becomes istics based on belting construction, particularly with a bonded

Fig. 28: Monitoring system for fabric belts. a) Monitoring system; b) Simulated output for variety of defects

a b
1 . 8 Miz Proces sor
Di gi tal osci l l ator
".- H I

EDGE LOSS FOLDED

FABRIC
Bel t

268
 bulk
solids Volume 1 4 • Number 2 • ApnVJune 1 994 • lleehanlcal conveyor Design
llan41lln8
.._.

steel cord belt in which cord load sharing is no always uniform.

Research being conducted by HARRJSON [13) is addressing this Profile of Be Tensions

problem. A dynamic sag measuring transducer has been devel­ _ !i_ _ _ - /_ _,.I - - - - - - - - F;ed
oped and monitoring of belt tracking and sag dunng starting *==:: ....
- - - .,,,,
- -

/fi: - - - - - - -
I✓

and stopping has provided new ays of proving hat these pa­ jb.-f _ _ _ _ _ _ _ _ _ =<j-�j
rameters have significant importance. The tracking and dynamic
� ld Conveyor Drive
01scharg
charac eristics may be predicted in advance during bet manu­ (a) "Piggy- Back Drive

facture.

7.2 On Line Control! of Belt Dynamics

Using the sag transducer mentioned in Sec ion 7 .1 above. HAR­

RISO has provided a methodology or he on-line con rol of con­
(b) Tripper Onve

veyor belt dynamic characteristics [13). The anaJys,s takes in o Fig. 29: Belt conveyor booster rives
account the design and location of belt tens,ornng devices.

7 .3 Conveyor Belt Monitoring

Following the development of · he s eel cord be monitor [8, 37]

a fabric belt monitor has been developed by HAIR�� [38] . Th,s
device employs non-contac capacitive ransducers and can
readily detect cuts in the cover material, edge damage, repairs,
splices and a number of manufacturing defec s. The monitor is
shown schematically in Fig. 28. Bo h s eel cord and fabric belt
monitoring devices have scope for in reduction into manufac­
turing processes in order o provide improved quar assu­
rance.

7 .4 Use of Aramid Fibres in Conveyor Belts

Much research and development work has been undertaken in

recent years in the use of Aramid fibre (such as 'Kevlar') rein­
forced belts (39-43 ]. These belts have high s rength w· h a sig­
nificant weight reduction advan age. However they have the Fig. 30: General of Cable Belt conveyor
disadvantage of being more susceptible o impac damage han
corresponding steel cord reinforced be s. Research at the Uni­
versity of Newcastle by BETZ et al. [4 , 2) has concentra ed on
the improvement of splices in 'Kevlar' reinforced bel s. Various
splice configurations have been examined and it has been
shown that splice streng hs greater han the paren be are
possible. Two- and three-dimensional finite elemen m els
have been developed to analyse e s ress distributions in
splices.

7 .5 Booster Drives Fig. 31 : Sandwich conveyor for . h lift

For long overland conveying applica ions. e application of 7.6 The Belt Conveyor as a Dryer
intermediate booster drives has he advantage of con rolling
belt tensions allowing longer fl1gh lengths o be employed l An additional application of belt conveyors was demonstrated
47). There are several types of booster drives including 'piggy­ by ATTWOOD and S ITI-I who studied the effectiveness of a belt
back' drives and tripper drives, illus rated in Fig. 29 as well as conveyor to dry brown coal during transportation [48). The
motorised idlers, linear motors and edge drives. The 'pig study involved the assessment of air velocity profiles in conveyor
back' drive shown in Fig. 29(a} consis s o a short driving be lo­ galleries in relation to the drying effect.
cated underneath and in contact wi h the carry side of he main
belt. The tripper drive shown in Fig. 29(b) has he a antage of
reliable drive tension control, but has e disa an age of requir­
ing transfer of the bulk solid at the drive loca ion. The other
8. Cab e Belt Conveyors
types of drives have their limitations ich so far has restric ed Cable belt conveyors offer a viable alternative to conventional
their application to other than a fe specialised cases. belt conveyor systems [49-51). In the Cable Belt system, illus­
A new type of booster drive is the 'Ozomin' drive hich is being trat in Fig. 30, the load carrying function of the belt and drive
developed by Minenco, Australia [46). The system compnses a tension function of the cables are separated. The belt, which is
secondary drive on the return bet and 'squeeze' idlers be een reinforced with wire mesh, is supported on the cables being lo­
the carrying and return sides of he belt to transfer he load en­ cated in grooves formed in the rubber. In current belts, it is usual
sion. A test rig incorpora ing his drive system is loca ed a he for the belt to have a preformed CU1vature to promote improved
University of ewcastle. tracking and assist the carrying capacity. Since the belt carries

269
 '/ bulk
Mechanical conveyor Design --. · Volume 1 4 • Number 2 • April/June 1994 solids
handlin9
:.
<I'

no drive tension, the belt thickness is less than the equivalent

-E3- steel cord or fabric reinforced belt. Should any section of the
belt become damaged, it is an easy matter to splice in a re­
placement section. Belt sections are often joined by mechanical
fasteners, although in some cases cold vulcanising is employed.
Again, since the belt carries no drive tension, field splicing is a
much simpler process than in the case of conventional troughed
belt conveyors. In the latter case, field splicing requires on site
vulcanising under controlled atmospheric conditions to ensure
the necessary quality control for load tension.
The design features of Cable Belts render them suitable for long
overland conveying applications with individual flight lengths up
to 30 km being possible. For instance the Worsley Aluminium
conveyor in Western Australia is 5 0.5 km made up of two con­
veyors, one 30 km and the other 20.5 km [49]. The conveyor
carries bauxite at a rate of 2,040 t/h, the speed of transportation
being 6.1 m/s. Another installation is the Selby coal drift con­
veyor in the U.K. [50]. This conveyor is 14.93 km long with a lift
of 990 m; the belt speed is 7.62 m/s and capacity is 1,830 t/h.
-E3- I Amongst the more recent installations is the Cable Belt coal
--E3- H
conveyor at Charlestown, Virginia, USA This conveyor trans­
ports coal at 1,000 t/h over somewhat rugged terrain. The
Fig. 32 : Cross-sectional arrangements of typical tube or pipe conveyors [54)
length is 6 km and the belt speed is 4 m/s.
As with conventional troughed belt conveyors, the technology
of Cable Belt conveyors is being continually improved; with the
aid of comprehensive computer aided design methodologies,
conveyors are able to be designed to operate in almost any ter­
rain. Combinations of horizontal and vertical curves are easily
accommodated. In the case of horizontal curves, radii as low as
5 00 m are possible. The introduction of belts with preformed re­
inforcing to give the belt an initial curvature is an important de­
velopment.
Fig. 33: Cross­ A major break-through in Cable Belt technology has been the
section development of the 'Triton ' plastic coated cables. These cables,
which operate in steel sheaves, have significantly improved life
of the
SICON
belt over the previous system of all steel cables running in sheaves
fitted with polyurethane tyres.
Fig. 34: Enclosed screw or auger conveyor

9. Special Conveyors

9.1 High Angl.e Sandwich Conveyors

The sandwich conveyor system [52, 5 3], illustrated in Fig. 31,

comprises two belts, which allows the bulk solid being con­
veyed to be completely constrained during transport. In this way
high angles of conveying up to 90 ° are possible rendering the
sandwich conveyor particularly useful where bulk materials are
required to be elevated. They are very applicable where space is
limited such as in continuous ship unloaders and process
plants. An important application is in open pit mining where high
lift conveyors provide an economic alternative to the use of large
:I:
�
0
...J
dump trucks.
.E One high lift conveyor, used for transporting coal refuse at 454
t/h, has a lift of 1 75 m at an angle up to 41°, the conveyor being
'ijj
:I:
454 m long. The installation is in Virginia, USA The belt width is
914 mm and the speed is 2.34 m/s. The belt has two drives top
and bottom each with an installed power of 186 kW. Another in­
stallation at a copper mine in Yugoslavia, comprises a conveyor
of length 174 m and elevation angle 35.5 ° conveying copper ore
at a rate of 4,000 t/h . The belt width is 2,000 mm and speed is
2.67 m/s. The installed power is 45 0 kW at the top and 900 kW
at the bottom.

270
 bulk ··---:-·
solids Volume 1 4 • Number 2 • ApriVJune 1 994 echanlcal conveyor Design
tiandtlna

9.2 Tubul1 ar or Pipe Conveyors

,,
,,
2.5
The pipe or tube type belt conveyor represents a major break­
through in belt conveyor technology [54-57]. There are several b , ci,
1 5°
mum Theoretical CII

versions of these conveyors which depend mainly on the

a
E 2.0 0
00 0
Throu
z
method of tube closure, as illustra ed by Rg. 32.
5 ,
,
5 >
0 �
Major advantages of tube or pipe conveyors include protection Q.
:I: 1 .5 , w
of the bulk material from the elemen s and from contamina ion
,,
C)
erP LI.m
and good dust control. Tube conveyors to date include belt di­ a: , 75° �
ameters ranging from 100 to 500 mm, capacities ranging from � ,,
,,
1 .0
90
°
c5
40 to 1 ,800 m3/h and speeds from 1 to 8 mis. They can be op­ a:
0
,
z
"(

erated at high angles of lift up to he vertical and both horizon al tu

� 0.5
and vertical curves can be readily incorporat in their design. ,

1------.-------------------4
:::::,

,,
0
0
9.3 Other Types of Conveyors 0 200 400 600 800 1000 1 200 1400 1600 1 800

AUGER ROTATIONAL SPEED rev/min

Another variation of he enclosed be conveyor is the sus­ a10k8 l.engtt a 76mm; Pilch • Oia. • 38mm: Radlal Clearance CID • 0.0104

pended type, such as the SICO conveyor illustrated in Fig. 33

...... Qne>;ad . �

[58). Generally this conveyor is limited to lower tonnage

throughputs in the order of 400 , but it is capable of negotiat­
Fig. 35 : Out versus speed for vanous angles of elevation

ing very sharp curves. Conveyor speeds range from 1 to 5 s. conveyor which comprised a screw 38 mm diameter with a
Other special conveyors include the 'Re o ell' conveyor [59- pitch 38 mm and length of 61 O mm. For each angle of elevation
61] and the pocket belt conveyor [62]. the output increases with speed, but the rate of increase less­
ens as the speed becomes higher. The throughput is shown to
Returning to the more conventional belt con eyor, air supported approach a maximum value around 1 ,800 to 2,000 rev/min; be­
belts are an interesting development [631, but o da e have yond this speed the throughput remains substantially constant
somewhat limited application. but in certain cases may actually decrease. The improvement in
throughput that results from a decrease in angle of elevation is
clearly shown and this is attributed to two factors, better feed­
1 O. The Enclosed Screw or ing of the screw at the intake end and a lessening of the rota­
tional or vortex motion of the grain.
Auger Conveyor
The dotted line on Fig. 35 represents the maximum theoretical
The construction and relevant de ails of he enclosed screw or hroughput which would occur if the conveyor were running
auger conveyor are shown in Rg. 34. The power driven sere 1 00% full and the grain were moving in an axial direction without
flight is supported in bearings and revolves in a stationary tubu­ rotation. This situation is impossible in practice but it does pro­
lar casing; practical limitations require a liberal clearance vide a basis against which the actual performance can be
between the flight and the casing and this has been sho n o measured.
be beneficial rather than detrimental to performance. The sere
flight is allowed to project beyond the casing at the lo er or in­
1 0. 1 .2 Effect of Choke Length
take end, this projection being referred o as he ·cho e". The
screw must be immersed into the feed hopper a least o the Cho e length has a marked influence on the conveyor through­
level of the lower end of the casing o herwise the auger will no put. At very slow speeds, under near static conditions, a mini­
elevate the bulk material. mum choke length equal to one screw pitch is sufficient to trap
As has been well established (64, 651, he volumetric efficiency the m ·mum amount of grain. However, as the speed in­
is limited by the degree of fill of the conveyor and the vortex or creases, the vortex formed limits the amount of grain that can
rotary motion of the material as it is transported by he sere . be trapped. Also, the centrifugal pressure impairs the inward
As an elevating device, the enclosed sere conveyor becomes flow of grain in the choke, particularly at the higher angles of el­
more efficient at higher rotational speeds as a resu of the re­ evation where the rotary motion of the grain is more pro­
duced, rotational speed or vorte motion of th.e bulk ma erial. nounced. To compensate for this, larger choke lengths are nec­
However, this advantage is usually more han offset by the de­ essary at higher speeds.
crease in 'fullness' of the conveyor that accompanies higher op­ Even with choke lengths up to three screw pitches optimum
erating speeds. The volume ric efficiency for steeply inclined performance is only obtained at lower speeds. Choke lengths
and vertical conveyors is quite lo as a resu of these factors beyond three pitch lengths become impractical and alternative
and the power requirements per unit throughput is high rela · e measures need to be taken if an improvement in performance is
to other types of conveyors. sought. In this respect some possibilities which have been tried
The performance of screw conveyors has been studied by sev­ and sometimes used are
eral authors. A brief selection o resu s based on the ork of • double start screw at intake end.
ROBERTS et al. [64-7 1 ] are now revie ed. • reduced pitch at the intake end particular1y in the case of the
vertical or steeply inclined conveyor.
1 0. 1 Performance Characteristics • guide vanes in the feed hopper.

1 0.1 .1 Throughput
1 0.1 .3 Effect of Conveyor Diameter -
For a given bulk granular material, the throughput of a sere or Corresponding Speeds
auger conveyor is significantly influenced by the conve or
geometry speed and angle of elevation. For e ample, Fig. 35 Screw conveyors of large diameter attain their maximum output
shows a set of output curves ob ained for a scale model sere at lower speeds than conveyors of smaller diameter. For in-

27 1
 'i.(�: bulk
Mechanical conveyor Design :·1( · Volume 14 • Number 2 • Apnl/June 1994 solids
handHn9
·:-:-;:• . ,

stance, tests on a 1 52 mm diameter by 152 mm pitch conveyor

geometrically similar to the model conveyor showed that the
characteristic throughput curves have the same shape as those
for the model but the full scale conveyor' reached its maximum
throughput in the speed range 600 to 1 ,000 rev/min. This com­
pares with the corresponding range 1 ,200 to 2,000 rev/min for
the model as shown in Fig. 35. The model was one quarter the
scale of the full size auger. Corresponding speeds are given by
non-dimensional specific speed number Ns defined by

N5 = -
oht ( 3 9)
g

where w = angular velocity ( rad/sec)

R0 = outer radius of screw
Fig. 36: ' Fullness' of conveyor
g = gravitational acceleration.

1 0.1 .4 Effect of Clearance

80

'#
> 7
An important consideration is to ensure that clearances >-
I="

between the outside edge of the screw and the casing approx­
0
z 6
w
imately equal to the particle size are avoided. This is necessary 0
u::: z
to prevent damage to the particles which become wedged in u. 50 0
w 0 ° �
the clearance space between the outside of the screw and the 0
a:
1 5 <(
30° G'.j
inside of the casing. From a practical point of view, since clear­
40

ances are necessary for smooth running of the conveyor, clear­

:::E 4s° �
:::, ° 0
...J 30 90
ances greater than the particle size should be employed. Tests
6 d' �

have shown that there is no loss in throughput due to leakage in

°
75 �
2 v-t---r--..-----r--r--..------.-�-�---.__::""t"' 9rP <(

the clearance space in steeply inclined conveyors. In the cases 0 200 400 600 800 1000 1200 1400 1600 1 800 2000

of lower elevation angles, the throughput is significantly im­ AUGER ROTATIONAL SPEED rev/min

proved by the greater cross-sectional area of the casing when Choke length = 76mm; Pitch = Dia = 38mm: Radial Clearance C/D = 0.0 1 04

the clearance is taken into account.

Material Conveyed = Millet

1 0. 1 .5 Volumetric Efficiency
Fig. 37: Volumetric efficiency versus speed for various angles of elevation

Comparisons of the volumetric efficiencies under various oper­

ating conditions provide important information in the study of h
auger conveyor performance. The volumetric efficiency ll v is de­ ll F = - ( 4 3)
fined as p
actual conveyor output Q
ll v = ----------- ( 4 0) where h is the average height of material on the helical flight and
maximum theoretical output Q t p is the screw pitch.
where

(4 1 )
Fig. 38: Power versus speed for various angles of elevation
D = screw diameter
Dc = core or shaft diameter
p = pitch
t5 = blade thickness. °
ci,

i 10 30 �
450 z
Considering all geometrical proportions of the conveyor to be in z °
15 Q
fixed ratios of the auger diameter 0, Eq. (41) may be reduced to �
7.5

Ot rND ( 4 2)
3 > 6� u:J
= z
0 u.
8 7 5° �
where r is a constant dependent on the geometry of the con­ � 5
veyor.
°
a: 90 �
<(
w
The auger output and volumetric efficiency are controlled by two
factors; firstly, the amount of grain that can be trapped in the
choke and conveyed into the casting; secondly, the amount of
rotational or wasteful motion given to the grain as it passes
through the conveyor. The "fullness" measured by ROBERTS and 0 200 400 600 800 1 000 1 200 1 400 1600 1 800
WILLIS [64] for the model auger have been used, together with AUGER ROTATIONAL SPEED rev/min
the volumetric efficiency, to provide an indication of the amount Choke length = 76mm; Pitch = Dia = 38mm; Radial Clearance CID = 0.0833
of rotational motion. The "fullness" is the amount of grain held Material Conveyed = Mile!

on the screw flight as indicated in Fig. 36. It is defined as

272
 bulk -...:..
solids VolurnP 1 -1 • Number 2 • Apnl June 1994 nlcal conveyor Design
h3ncl1ins -,

Fig. 37 shows, for he model conveyor a typical set of graphs in­

dicating the variation of volumetric e c·ency with speed for dif­ 25
ferent angles of elevation. These graphs are shown by the full 'tt.
9d' 1s°so
lines. The dotted lines in Fig. 37 are he superimposed constan >- 20

"fullness• 11 F graphs. w 15
The maximum volumetric e cienc1es occur at the lower speeds,
w
decreasing at an almost uniform rate with increase in speed.
10

The reduction in volume ric efficiency ith increase in angle of

elevation is quite substantial.
0 -t--..-�--r--,----,.----.-----...------.---4
The percentage rotational motion can be determined indirectly
from the difference be een 1'\ F and T\v· For instance hen =
0 200 400 600 800 1 000 1 200 1 400 1 600 1 800 2000

90° and N = 400 rev/min, llv = 33% and llF = 73% indicating
AUGER ROTATIONAL SPEED rev/min

40% loss due to rotational mo ion. On e o her hand, hen 0 =

Length "' 76mm; P1lth = Dia. = 38mm; Racfial Clearance CID = 0.0833
...... Ccnleyed = Mlel
90° and N = 1 ,000 rev/min T\ = 32% and ri F = 55% indicating
only 23% loss due to rota ionaJ motion. This shows hat the
amount of wasteful rotational mo ion ac ually decreases with in­
Fig . 39: Overall efficiency speed for various angles of elevation

crease in speed. However, this is more than offset by the de­

1 0.2 Predicted Performance
crease in fullness which accounts for the decrease in ll v ith in­
crease in speed. 10.2.1 Dimensional Analysis

1 0.1 .6 Conveyor Power and Overall Efficiency In order o relate the perfomiance data from model tests to pre­
die the perfomiance of the prototype conveyors, the appropri­
(ij Power to Convey Grain ate dimensionless parameters for dynamic similarity need to be
established. By means of dimensional analysis, the following pa­
Fig. 38 shows the power versus speed characteristics for vari­ rameters were derived.
ous angles of elevation as obtained from he model conveyor
studies [64]. d
-, µ )=o (4 6 )
The results plotted in Fig. 38 are the po ers necessary o con­
vey grain only and are the differe ces be een he no-load and
full-load powers. For each angle of elevation he power is shown
where Q= throughput
to increase at an almost uniform rate · h speed. For each
speed, the power is shown to increase · h angle of eleva ion d= particle size
from 0° to a maximum at 30°-45° and he o decrease again. N= rotational speed
P= power
M Overall Efficiency
D= screw diameter
The variation in overall efficiency 110, with speed and angle of el� µ= internal friction
evation is shown in Fig. 39. The overall efficiency is de ned as
µ5 = screw surface friction
Theoretical Power µc = casing friction
11 0 ------- 44
0= angle of elevation.
=
Actual Power
The discharge parameter, referred to as the discharge coeffi­
The theoretical power is the power required to give the required cient and denoted by C0, is
output at the required height in the absence of friction and is
given by Q
C 0 = -3 (4 7 )
P = Q p g H kW) 45, ND
where Q = output (m3/s) The power parameter, referred to as the power coefficient and
denoted by Gp, is
g = 9.81 (m/s2)
p = bulk density ( m3) (4 8)
H = heigh of lift (m).
The overall efficiencies are generally quite lo , the m ·mum
values occurring at the higher angles of elevation ere he The acceleration parameter g!N2D takes into account the cen­
conveyor lengths for a given height are a minimum. The •curving tripetal acceleration of the granular bulk solid and provides a
down• of the 90°and 75° graphs in the low speed range is means for comparing the speeds of geometrically similar con­
mainly due to leakage back in the clearance space. veyors. Because g in the acceleration parameter is constant, it
will be taken as being unity. Also, for convenience the parame-
The screw or auger conveyor po er losses may be divided ·nto er is inverted and is referred to as the speed coefficient denoted
three groups: by C5,
• Power lost in overcoming friction as the grain moves ially
along, the conveyor. This is an inevitable loss; it includes the (4 9)
grain-on-casing and grain-on-screw surface friction losses. It is noted that Cs is an alternative form to Ns defined by Eq.
• Power lost in rotating the grain. (40). It is assumed that, for a given value of Cs, the discharge
coefficient and power coefficient, respectively, are theoretically
• Power lost in the cho e. the same for all geometrically similar conveyors, the other pa-

273
 � b�k
Mechanical conveyor Design 1
1-.,jl;,
,:.
Volume 14 • Number 2 • Apnl 'June 1994 solids
handlin9
.

rameters being constant. Then the corresponding speeds for

the model and prototype for similarity may be obtained by con­
sidering the equality of the speed coefficients:

NmDm = NP O P
2 2

(5 0 )
h

( Suffix 'm' refers to the model and 'p ' to the prototype. )
1 0.2.2 Predicting Performance

[tliliill!
Given the performance characteristics of any known screw con­
veyor referred to here as the model, m, the performance of any
geometrical similar prototype conveyor, p, of different size may
be predicted as follows:
(i) Corresponding speed of model Ai ___. � I
- this is given by Eq. (39) or Eq. ( 49) I Ro ..,
( ii) Volumetric throughput - from Eq. ( 48) (a) Conveying Action (b) View Showing Cut-Away Section

(5 1 )
Fig. 41 : Mechanics of conveying action in auger conveyor

( iii) Power - from Eq. (41 ), assuming equal LID ratios for the
model and prototype

PP = � ( NP J ( 0 P J Pm
3 s

(5 2 )
Pm Nm Dm

Fig. 40: Predicted throughput of different size conveyors at various angles of

elevation and speeds. p = D; 1 /p = 2; D/D = 0.333; CID = 0.0833;
d/0 = 0.05

500 -r-,�......,.-,,-,-....-r,...,..,.........,.,......-,-........,...,...,.,...,...,...,...,..,.,.....-,..__.........

400 -t--+--+--+-+--+----+----+--+---+---+...t
Fig. 42: Velocity diagram for granular particle
300 -t--+-t--+-+--+----+----+--t----+--,j�

200 -t--+--t--t-+-t----+---t--t---t,§'H-t'-ifi
On the assumption that the power is a linear function of length,
1 50 +--+--t-+--+-+----+---+--1--iW,C..I-J4-1 Eq. ( 44) may be expressed in alternative form as
.....�
100 -t---t---+--+-+--+---�- b'-' -------­
q, (5 3 )
c,l
50 -r--t--+-+-+-+--- ���W,'-1-lr---+---H
f.s .l -.fl'rrl't-T--M--t---H
";-..
�

1 0.2.3 Performance Equations and Charts

The combination of the above dimensional parameters and ex­
t---t--+---+---+--+---

3: 20 -r--t--+-+-+-+-- perimental data permit performance equations as shown in [64,

g0 1 5 -t---t--+---+--+-+-
a: Fig. 43: Helix angle of grain path at outer periphery
� 10
a:
w 50
� . 5 45
0
w
40
b0
w 35
� 2 �ffl<-+l--,1--l--t---�- A o (Deg.) 3 0
1 .5 -tY1'-7hf-i'--i,''-t--t---+---,---�f---+---+-1
z 25
0
1,-L-'--�.L.LJL...I..L.lr,U..LL.1..U..U.1,...LJU..J,..U..u.i_,u..u..,__a..;u....L� 9 0 � 20
75 w
15
�,-,.........._._.,.._._..........
.L...U,
...........a....L"'4,_._._....ul,,.._._.J.¥......,__.__,_...._q
1-"---'r-L....L..........,,...._..............u.-,..u.1..u...L�.........�.........,;........�--'---4 60 ill i
t"-..._.,:--'-....._.,-'\-'........�,...._............�......_..._..,l-'-"-'-'+'"""""t-'-1 45 lL CD 10
..._._...........-\--'__,_-'--\-_._._.........4-'-L..Lo..a.....a...�L.&..1���� 30 0 o, 30
i----�---�_._.i-a-...........---+--...........................�......_............+-'-'-w+I 15 � ! 0 10 20 40 50 60 70
_...__,__._..._._�......._.........��.........��
SP E CIFIC SP EED Ns = ro�R o
0 CJ
70 1 00 1 50_._._�2�
00...............�3UOO �
SCREW DIAMETERS (mm)

274
 bulk
solids
hant1lin9
V11lurnt' 1 .1 • Numbt1r 2 • Apnl-'June 1 994 · • hanlcal conveyor Design

70]. It is often more convenient o use performance charts such

as that for volumetric hroughput shown in Rg. 40. A similar
chart for conveyor drive power has also been obtained [64).
In determining the opera ing conditions for a particular con­
veyor, consideration must be given to the maximum economical
speed · speeds higher than his will no yield any gain in output
and in some cases, may cause a reduc ion in o put. The m
imum economical speed ·11 vary · h conveyor size becoming
less as the diame er increases. Based on he model conveyor
studies [64] , and using he speed coe cien Cs, the maximum
economical speed for any geometrically similar conveyor is
given by
386
'o
=
Ozti j'h
N ma - 54 )
\
where D = conveyor diameter (m).
elemen in vortex motion

Considering an element of material at radius r, it follows that for

1 0.3 Vortex Motion in Vertical Screw Conveyors equilibrium, the radial pressure o, must balance the centrifugal
pressure. That is
As indicated previously, the performance of enclosed sere or
auger conveyors is influenced by the ro a ional motion of he
bulk material. The rota ional or vortex motion has the grea est (5 5)
effect in vertical or steeply inclined conveyors. Vorte mo ion in dr r
vertical conveyors has been s udied in some de ail by ROBERTS The general relationship for vortex or rotational motion is
[65]. The results of this investigation have an important bearing
on conveyor design, particularly · respect to the selec ion o ½- r = C = co n stan t
0
(5 6)
screw pitch and he diameter of the central core or sha . The
subject is of relevance to the study of performance and ear in where n = vortex index
screw conveyors [71].
It can be shown that the variation in height h with radial direction
1 0.3.1 Mechani. c s of Conveying Action across the screw surface is

The conveying ac ion is depicted in Rg. 41 . As he sere ro­

tates a granular particle moves in a helical pa h of opposite (5 7)
hand to that of the screw. In the vetoc· diagram shown in Rg.
40(a), Vs is the tangen ial velocity of the screw at the radius con­ here
sidered, VR is the relative veroc· of he particle · h respec to
the screw surface and V is the absolute velocity of the particle.
2

Zo = -- = h o + ---
Kv Kv CV
The angle A defines the direc ion of he absolute veIoc· and
(5 8 )
'Y 2ng Ro
2n
hence the helix angle of the pa h fol o ed by the partic es a he
radius considered. The absolute velocity V is resolved into o
components, the 'useful' lifting componen VL and the 'was eful' ROBERTS (65] showed that the characteristic vortex formed was
rotating component Vr a function of conveyor speed, changing from a forced vortex
· h n = -1 at low auger speeds to a constant tangential speed
The velocity diagram depic ed in Fig. 1 (a) ,s shown in more de­ vo e with n = 0 at the higher auger speeds. That is,
tail in Fig. 42.
For low speed V1 = Cv r (59)
The helix angle of the sere increases from a0 at the o er pe­
riphery of the screw to a the core or sha . This means ha For high speed VT = Cv (60)
the helix angle A of the pa h of particles of bulk ma erial de­ It is considered that the full immersion of the auger choke in re­
creases from l0 at the outer periphery to a he core. As a ducing grain slip in the central region of the screw serves to re­
consequence of he vortex motion established within · he bulk s rict the vortex motion; this in effect enhances the perfor­
material internal shear be een concen ric layers occurs. In he mance. Under these conditions, it was shown that for Ri = 0.33
case of free flowing granular ma erials, he in emal shear is R0 there was no slip and the whole of the screw surface was ac­
readily generated · this is not he case with cohesive and s ringy tive [65).
materials which do not convey well in vertical or s eeply inclined
augers.
1 0.3.2 Vortex Performance Results
As reported by ROBERTS and 1ws (64), measurements of the
grain helix angle at the outer periphery o a model sere con­ In order to illustrate the various vortex parameters, consider the
veyor of pitch = diameter = 38 mm showed tha the angle l0 in­ case of a vertical enclosed screw conveyor elevating grain [71}.
creased with increase in ro ational speed. The measured in­ The conveyor is 0. 1 5 m diameter with a pitch also equal to 0. 15
crease ranged from appro ima ety 1 2° a 200 rev/min o O at m. The conveyor operates at 400 rev/min and it is assumed that
1,800 rev/min approaching a cons an value of appro imatety the vortex motion is characterised by a constant tangential
46°. These measurements applied to rrnlle seed. The resul s speed Vr, that is, n = 0. From Eq. (39), Ns = 13.4. Hence from
may be generalised for geometrically ·milar conveyors by plot­ Fig. 43, 1'.0= 34°.
ting 11. against the specific speed parame er Ns which is defined
Fig. 45 shows the variation of screw helix angle a and grain path
by Eq. (32). Values of ).. as a func ·on of N5 are plo ed in Rg. 43.
angle A with screw radius, while Fig. 46 shows the variation in
The cross-sectional profile o bulk material of bulk denS1ty p on radius of the various velocity components as well as the grain
a screw surface is illustrated in Fig. 4 height h.

275
 Mechanical conveyor Design
->t··
-.., ··
1
Volume 14 • Number 2 • Apnl/June 1994
bulk
solids
handlin9
·�,1

50 -1"---�-.......-�-�-----�--------�--------+- 5 -+--.......:.�����..........--..............-.......__......_--+- 0 .25

i -.- vs
�HI
-e- Ajpha
--- L,mbda j
4 ................... f.....a....-vR- ·t----<-------1,-....-••-··· . ···················· 0.2
40
en j � VT
13 0.15 -
E
�
.c
fill 30
5
1-
:2,. :I:
g 2 �-.-�-�--:s,;�-----=-­ 0. 1 �
w
� 20 > :I:

< 1 0 . 05
I

10 :Ro
0 -+-.....,...,_,...,........._--r-i....,.......,....,....;.-......,.....,....,.......,_,...,.....,........,...,....._--+- 0
0 . 02 4 0 . 05 0. 6 0 . 07 0.08
RADIUS r (m)
0
0 . 02 0 . 03 0 . 04 0.05 0.06 0.07 0.08
Fig. 46: Screw and particle velocity components and height h versus radius
RADIUS r (m)
00 = pi = 0. 1 5 (m); 01 = 0.05 (m); N = 400 rev/min; Grain; n = 0

Fig. 45: Screw helix angle a and particle path angle ). versus radius As an example, consider the previous case of the 0. 1 5 m diam­
00 = pi = 0. 1 5 (m); 01 = 0.05(m); N = 400 rev/min; Grain; n = 0
eter conveyor operating at 400 rev/min. The friction angles for
the screw and the casing surfaces are <l>s = <l>c = 20° and the
1 0.4 Wear of Screw Conveyors height of material at the outer periphery is assumed to be h0 =
0. 1 2 m. The pressure ratio Kv = 0.35 and the bulk density of the
1 0.4.1 Wear of Screw Surface granular material being conveyed is p = 0.85 t/m3. As previously
determined, VT = 1 .0 mis = constant and VAo = 1 .2 1 5 m/s. Then
As shown by ROBERTS [7 1 ], the abrasive wear rate on the sur­ from Eq. (64)
face of the screw in a vertical conveyor is given by

(:�J (6 1 )
WRc = 0.632
(:�J
From an analysis of the forces and pressures acting on the From Fig. 48, the average wear of the screw surface WRc(avJ =
screw and casing, as illustrated in Fig. 47, it can be shown that 0.97 (kN/ ms) which indicates that the average casing wear in
the normal pressure on the screw surface is given by this case is 65% of the average screw surface wear.

(6 2 ) 1 0.5 Critical Speeds of Screw Conveyors

where g = pg = bulk specific weight It is particularly important that auger or screw conveyors be de­
signed and operated so as to avoid critical vibration frequen­
and
cies. Apart from the need to protect the equipment against
sin a. tan (a + A) + cosa damage and failure, there is the need to prevent grain damage
C N = -------- (6 3 ) due to the transverse vibration or 'whipping' motion of the flight
1 - � tan ( a + A)
within the casing.
h = height given by Eq. (57)
The subject of screw conveyor vibrations and grain damage has
By way of illustration, Fig. 48 compares the wear rates for a ver­ been studied by ROBERTS and ARNOLD [66-68). As has been
tical screw conveyor operating at two different speeds, 1 00 shown, for a single screw flight supported in two self-aligning
rev/min and 400 rev/min.
(i) For N = 400 rev/min, VT = 1 .0 (constant) and A0 = 34 ° . Fig. 47: Force analysis of 'lumped' particle
(ii) For N = 1 00 rev/min, V r · 0- 6 = 2 . 54 and A = 1 2 °.
T 0

The predicted results show that the screw surface wear in­ •n•
creases almost linearly from the shaft to the outside. The signif­
icant increase in wear at the higher speed of operation is clearly
\
shown. \
1 0.4.2 Casing Wear
At the interface between the bulk material and the casing, parti­
cles in contact with the casing execute an elliptical path of ra­
dius R/cos a.0 , where Roe is the radius of curvature of an ellipse,
a.0 is the screw helix angle and VAo is the absolute velocity at the
outer periphery.
It may be shown that the average casinq wear is given by

(::J (6 4 ) .l\m g

276
 bulk --:
solids
mn1111na
Volume 1 4 • Number 2 • April/June 1 994 hanlcal conveyor Design

bearings as in Rg. 33 the natural frequencies for the transverse

vibration are given by
j N = 40Q rev/mi�
---1 J 7t
= -
,2

-
z
en
E 1 .5 -··---· .·
N = 1 � rev/mi❖ f
I 2 L�
(Hz ) 6 5)

where j =
1 , 2 3 . . ..
a:
! lb =
unsupported length, m
3: =
elastic modulus, Pa
W·
E
I =
second moment of area of auger flight
about transverse axis, m4
< 0 .5
a: g = acceleration due to gravity = 9.81 m/s2
w5 = distributed mass of the flight, kg/m.
W I

3: R 1· :,
Research has shown that the helical blade makes a negligible
con ribution to the auger stiffness. A satisfactory estimate of the
0.07 0 . 08 critical speeds can be obtained by basing the calculations en­
tirely on the central core tube or shaft alone. Eq. (65) can be
modmed to the following form
Fig. 48: Wear rate of screw surface as function o
2 3 .6j IE d� + d�
2
D0 = P, = 0. 1 5 (m); D, = 0.05 (m): K = 0.35; G . p = 0.85
For N = 100 rev/min, n = -0.65 C = 2. CR =
-- rev / min ) (6 6)
For N = 00 rev/min, n = 0 and C = 1 .0 L� Ps

where d0 = outside diameter of core tube or shaft, m

d1 = inside diameter of core tube, m
I
'I s = density of steel, kg/m3.

: - - - ,, - - - - 4, - 2nd
-� - - - -Harmonic
--�-- --r---- ----
Rg. 49 illustrates the critical speeds as a function of unsup­
ported length of a screw flight having a central tubular shaft d0 =
1

a l f

50 mm and d1 = 45 mm. Owing to the flexibility of the screw

t
I

z fligh , i is common for the lower natural frequencies to occur

Cl) 60 wen within the operating speed range of the conveyor. It is im­
w portant, therefore that the critical frequencies be determined
w and the conveyor operated at speeds that avoid these frequen­
a.
(/) cies. To achieve this, the operating speed should be outside the
band ± 1 5% of the relevant critical speed New
O 20
' t , ' '
- - - -� - - - � - - - - - - - ➔- - - - � - - - - --·-�----

i=
• t ' '
I
.
I

0 0 -+---+---+---+---+----t---t----.---;-,
I
1 1 . Bucket E levators
2 3 4 5 6 7 8 9 10
CONVEYOR LENGTH' L (m) Bue et eleva ors are widely used in industry to elevate bulk ma­
terials from one level to another. Because the bulk materials are
retained posi ively in the buckets during elevation, bucket eleva­
Fig. 49: Critical speeds of screw con tors operate effectively in the vertical position thereby occupying
r le floor space.
Fig. 50: Types of buc et eleva ors

(a) {b) {c)

277
 �lf .
bulk
Mechanical conveyor Design -�- " Volume 14 • Number 2 • April/June 1994 solids
;� handlin9

Bucket elevators may be classified in ac­ spaced buckets and discharge com­ buckets and lead the material to the
cordance with their principal mode of dis­ mences the return side of the elevator discharge chute without falling back in
charge. As depicted in Fig. 5 0, there are after the buckets have passed the Top the elevator.
three type of elevators: Dead Centre ( TDC) position. Some
sliding of the bulk material may occur
( i) Low Speed Gravity Discharge Eleva­ at first, but then the material moves in 1 1 .1 Mechanics of Discharge -
tors - Fig. 5 0( a). These have closely free flight towards the discharge Elevators Type (a) and (b) -
spaced buckets, with discharge com­ chute. It is essential that the bucket A Case Study
mencing by material sliding on the spacing and operating characteristics
inner surface of the carrying bucket, be such that the material can dis­ Several studies of the performance of
the final stage of discharge occurring charge freely without interference with bucket elevators have been carried out
as the material is directed to the dis­ the leading bucket. [72-74]. As is evident from this research,
charge chute by sliding over the outer the motion of bulk materials in a bucket
surface of the leading bucket. In effect, ( iii) High Speed, Centrifugal Discharge El­
elevator during discharge is quite com­
the outside surface of the leading evators - Fig. 5 0( c). In this case, dis­
plex. It is important that the mechanics of
charge commences before the buck­
bucket acts as a transfer chute. performance be fully understood if the full
ets reach the TDC position. To avoid
performance potential is to be appre­
( ii) Medium Speed, Centrifugal Discharge interference, the discharge trajectory
ciated. This was demonstrated in [73]
Elevators - Fig. 5 0(b). These have must pass over the top of the rotating
with a case study example of an industrial
bucket elevator in a fertiliser plant in
which the speed of the elevator was in­
Fig. 51 : Path of particle and relevant bucket positions during discharge. v8 = 0.61 m/s

creased with a view to improving the per­

formance. However, the increased speed
produced greater fall-back in the elevator
and prolonged the run-out time to com­
pletely empty the system. In effect, the
bulk material could not discharge from
the buckets fast enough to allow total dis­
charge by the time the buckets had
passed the discharge chute.
An appreciation of the behaviour during

t·
I

discharge may be obtained by studying

the motion of a single particle. This ap­
proach has some limitations when used
to describe the motion of a mass of parti­
cles in a full bucket during the total dis­
charge period. However, the single parti­
cle theory is quite relevant to the study of
the final stage of the discharge process
when small quantities of bulk material are
required to be emptied from each bucket.
For the case study example referred to
--1����===::�=�Po;;s;
i t�
i o;
n�o;fP.
Pa:i
rt�le above, the discharge was a combination
of (a) and (b) of Fig. 5 0.
a t Commencement of
F r ee Fl i ght
Discharge from a bucket has three
phases which occur in sequence:
(i) Sliding of particles relative to the
bucket surface
(ii) Free flight
( iii) Interference with the leading bucket,
that is, the bucket immediately ahead
of the bucket from which discharge is
I Bu e ke t Speed Vs = 0·6 1 m/s I occurring.
In this case the flow is re-directed by the
0·1 0·2 0·3 0·4 leading bucket. The manner in which the
flow is re-directed depends on the geom­
Sc a le [ m e tres] etry and pitch of the buckets and the
speed at which the buckets travel. In the
case of gravity discharge elevators, the
outside surface of the leading bucket acts
as a chute to guide the flowing bulk ma­
terial.
The dynamic models and corresponding
equations of motion for each phase are
y
presented in [73]. They were used to an­
alyse the performance of the industrial
---'o---�.r1 _..,._______ bucket elevator, the results of the study
being summarised here.

278
 bulk - -=�
solids Volume 1 4 • Number 2 • ApriVJune 1 994 hanlcal conveyor Design
handlf119 , -

The buckets of he elevator in question and he other for he increased bucket is one of the last to be discharged, then
are attached to chains, the pitch circle ra­ speed of 1. mis. In each case, the path the lower operating speed of 0.61 m/s
dius on the drive sprocket eels being o a particle discharging from he back virtually guarantees complete discharge
denoted by R8. The elevator details are: end of he bucke is traced. The relevant of the buckets during the final run-out
de ails are as folio s: phase.
R8 = 0.37 m
= 30° = friction angle of bulk
solid on bucke 1 1 .1 .1 Case 1. : 1 1 .1 .2 Case 2:
Bucket Speed v8 = 0.61 mis Bucket Speed v8 = 1 . 1 43 mis
R = 0.22 m = inner radius o bucke
path For this case sliding first commences at For this case 80 = 24° . As can be seen
the bucke angle O = 30° . As indica ed in from Fig. 52, the path followed by a parti­
Bucket top surface
Fig. 5 the path of a particle during dis­ cle during discharge shows the particle
slope angle = 43°
charge is such tha he particle is directed missing the intake of the discharge chute;
Bucket pitch = 0. 57 m. through the discharge chute. Since the this results in the particle falling back in­
Two elevator speeds are examined, one particle under examination travels from side the elevator without being dis­
for the design bucket speed o 0.61 mis he rear end of he bucket and, as such charged. If the opening of the discharge
chute could be lengthened in the down­
ward direction, then particle discharge
during the run-out phase should be pos­
Rg. 52: Pa h of particle and relevan e po5IIOllS sible at the 1 . 1 43 mis operating speed.

1 1 .2 High Speed Discharge -

Elevator Type (c)

In this case discharge commences be­

fore the buckets reach top dead centre

,
as illustrated in Flg. 50(c). The mechanics
of discharge have been presented in [3,
74) with the work of BEVERLv et al. [74)
,.#; also including experimental investigations
on a pilot scale model elevator. Using
layers of marked particles and high speed
photography, the discharge process was
-.
Position o f
Particle at studied, a typical set of results being
Commencemen t shown in Flg. 53. The bulk material flows
o f F ee Fli t
from the inner portion of the bucket. a
convexity developing on the surface as
discharge proceeds as shown. It is im­
portant for all the material to be thrown
clear of the ip path of the buckets and
reach the discharge chute. This is to
avoid interference and prevent fall back.

1 2. Other Types of
Mechanical
Conveyors
There are many other types of mechani­
cal conveyors such as vibratory convey­
ors and drag chain feeder conveyors [75-
77]. It is beyond the scope of this paper
to consider these.

1 3. Concluding Remarks
MI :Z 0� 0;4 The subject of mechanical conveying is
Scale ( e res}
I f!.

indeed very broad. The applications are

many and varied, ranging from in-plant

___'(__, t_
\; /
movement of materials to long distance
over1and transportation. There have been
significant research developments in all
areas of mechanical conveying and there
✓
______ are numerous examples of new types of
conveyors which reflect clever, innovative
design.

279
 )·:•·, bulk
Mechanical conveyor Design �-.;_:_ Volume 1 4 • Number 2 • Apnl/June 1 994 solids
��
�c

I
60 °

./

84 RPM
FROODE NO. 2 . 84
BUCD:'1' ANGLE 45•

MATERIAL

Fig. 53: Progressive stage of bucket discharge

In view of the vastness of this subject, in a single review paper References

such as this, it has not been possible to cover all types of me­
chanical conveying equipment. The approach taken has been [1 ] JONKERS, C.O. : Loss Factor in Transport; Fordern und Heben, Vol. 31 (1 981 )

to highlight the basic problems associated with the mechanical

No. 2 .

conveying of bulk solids and to review those more common

[2]. ROBERTS, A.W. , HAYES, J.W. and SCOTT, O.J.: Optimum Design of Continuous
Conveyors; bulk solids handling, Vol. 1 (1 981 ), No. 2, pp. 255-264.
types of conveyors, notably belt conveyors for short and long [3] ROBERTS, A.W. and HAYES, J.W.: Economic Analysis in the Optimum Design of
distance transportation and screw conveyors and bucket eleva­ Conveyors; TUNRA, The University of Newcastle, Australia, 2nd Ed. , 1 980,
tors as examples of the more widely used types of in-plant ISBN 0-7259-3406.
equipment used to move and elevate bulk materials. (4] HARRISON, A. , HAYES, J.W. and ROBERTS, A.W. : The Feasibility of High Speed
Narrow Belt Conveying for Bulk Solids Handling; Mech. Eng. Trans. IE(Aust.),
Of the various modes of continuous conveying of bulk solids, Vol. ME7, {1 982). No. 3.
belt conveyors are of considerable importance in view of their [5] ROBERTS, A.W. , HARRISON, A. and HAYES, J.W. : Economic Factors Relating to
widespread use, the tonnages that can be accommodated and
the Design of Belt Conveyors for Long Distance Transportation of Bulk Solids;
bulk solids handling, Vol. 5 (1 985) No. 6, pp. 1 1 43-1 1 49.
their proven reliability. The paper has focused on long distance [6] HARRISON, A. : Flexural Behaviour of Tensioned Conveyor Belts; Mech. Eng. ,
overland conveying indicating the influence of economic and Trans. IE(Aust.), Vol. MES (1 983), No. 3.
technical considerations in the design methodology. Aspects of [7] HARRISON, A. : Criteria for Minimising Transient Stress in Conveyor Belts;
conveyor dynamics have been discussed and research and de­ Mech. Eng. Trans IE(Aust) , Vol. MES (1 983), No. 3, pp. 1 29- 1 34.
velopment work concerned with specific subjects including [8] HARRISON, A. : Dynamic Measurement and Analysis of Steel Cord Conveyor
drive drum friction, rolling resistance, booster drives and bulk Belts; Ph. D. Thesis, The University of Newcastle, Australia, March, 1 984.

solid and conveyor belt interaction have been highlighted. Men­ [9) HARRISON, A. and ROBERTS, A.W.: Technical Requirements for Operating Con­
veyor Belts at High Speed; bulk solids handing, Vol. 4 (1 984), No. 1 , pp. 99-
tion has been made of other types of belt conveyors notably the 1 04.
Cable Belt system, high lift and pipe conveyors. (1 0] HARRISON, A. and ROBERTS, A.W.: Future Design of Belt Conveyors Using Dy­
Both screw conveyors and bucket elevators play an important
namic Analysis; Proc. Transportation Conference, 'Bulk Transport - Solid,
Liquid or Gas', lnstn. of Engrs. Aust. , 1 984, pp. 47-51 .
role for in-plant movement of bulk materials. Some salient per­ [1 1 ] HARRISON, A. and ROBERTS, A.W.: Modern Concepts in Belt Conveying and
formance characteristics of these two types of conveyors have Handling of Bulk Solids; TUNRA, The University of Newcastle, Australia,
been presented. 1 988.
[1 2] HARRISON, A. : Future Design of Belt Conveyors Using Dynamic Analysis; bulk
It is clear that significant research and design developments are solids handling, Vol. 7 (1 987) No. 3, pp. 375-379.
continuing to be made in the field of mechanical conveying. (1 3] HARRISON, A. : New Concepts for Evaluating Belt Tracking and Dynamic Ten­
There is wide industrial acceptance of these developments and sions; Third Intl. Cont. on Bulk Materials, Storage, Handling and Transporta­
this is particularly gratifying. The interaction between industry tion, IE(Aust}, Newcastle, June 1 989, pp. 1 1 4-1 1 8.

and research institutions is of great importance in guiding cur­ (1 4] HARRISON, A. and BARFOOT, G. : Modelling the Effect of Take-Up Location on
Conveyor Belt Performance; Third Intl. Cont. on Bulk Materials, Storage, Han­
rent research and indicating new research directions. dling and Transportation, IE(Aust) , Newcastle, June 1 989 pp. 1 1 9- 1 23.

280
 bulk ·-: ·•--:,,
solids
hancllln9
\'�' ,ime 1 ..: • Number 2 • Apnl June 1 994 •- : '
- llecllanlcal conveyor Design
s -�-

[ 1 5] GOl..osa , T.S. . Cf.n.E.wA W. and J.: lrnesbgations o Transien 145] SGa.. TH.: The Use o Intermediate Drive Units for Belt Conveyor Systems;
Stress Front Velocity in Conveyor Belts; solids handhng, Vol. 10 ( 1 990) bu handling, Vol. 2, ( 1 982) o. 1 , pp. 53-57.
No. 4, pp. 385-390. 4
(6} OzOUNS, I. and F . M.A.: The 'Ozomin' Mid-Conveyor Drive; Proc.
[ 1 6J MORRISON, W. R.B.: Computer Graphics Techn ues or VISUalizing Bel Stress Mech. 91 . Conference 5, Cost Effective Bulk Materials Handling, The lnstn. of
Waves; bulk solids handling, Vol. 8 ( 988) . 2, pp. 221 -227 Engrs. Aust., Sy ney. Aust. July 1 99 1 , pp. 79-83.
. (47] HAAAlsoN, A.: Power Oscillation Mechanisms in Booster Conveyors; Proc.
[ 1 7) GREuNE. A. and HAGER, F. R.: The Energy-Sa Design of Belt Corweyo(s;
bul solids handling, Vol. 1 0 (1990) . 3. pp. 245-250. Mech. 91 , Conference 5, Cos Effective Bulk Materials Handling, The lnstn. of
Engrs_ Aus .• Sydney, Aust. July 1 99 1 , pp. 89-95.
(1 8) HARPJSON. A. : on-Linear Processes Chaos in Belt Conveyor Sys ems;
bulk solids handling, Vol. 1 1 (1991) o. 1 , pp. 147- 1 55. [48] Amvooo, R. and SMm-t, B.D.: Assessment of AJ.r Velocity Profiles in Belt Con­
vgyor Galleries and their Drying Effect on Brown Coal During Transportation;
[1 9} HAfRsoN, A.: em Belt Take- ps and Their Dynamic ,on: bu
bul solids handling, Vol. 7 (1 987) No. 3, pp. 385-390.
handling, Vol. 1 2 (1 992) o. 4, pp. 581 -584.
(49) FARRY, K.P.: The Worfd's Longest Conveyors; Proc. Transportation Confer­
[20) HAGER, M. and Be<a, S.: Curved Belt Conveyor Decen ralised Frietion
ence, 'Bui Transport · Solid, Liquid or Gas', lnstn. of Engrs. Aust., 1 984, pp.
Wheel Drives; bu solids ing, Vol. 2, ( 992) • • pp. 547-555. 30-36.
[21 ) G ER, K.J. and . F.: The Design o Belt Conveyors Honzon al (50) THOMSON, I.M.: The Cable Belt Conveyor at Selby Mine; The South African
Curves; bulk solids hand , Vol. 1 2, ( 1 992) . . pp. 557-563. echanical Engineer, June, 1 983, pp. 1 34-1 38.
[22) BoccHETTI, H.M.: 1 980- 1 983 Three Prestigtous ComlTllssioned Projects In­ (51 ] G: , G.A.: Cable Belt Conveyors - Miscellaneous Papers; Private Com-
clude 28 ew World Records o Steelcord Belt Convey()(s; Proc. Transporta­ munication, Cable Belt (Australia) Pty. Ud., July, 1 993.
tion Conference. 'Bulk Transport - Solid, Liquid or Gas', lnstn. o Eng,s.
1 984-, pp. 37-42. (52] Dos SANTos. JA: High Angle Conveyors - HACs: A Decade of Progress
Yi a Rock Solid Futu e: bulk solids handling, Vol. 1 0, (1 990) No. 3, pp.
[23] NORDW., L.K.: The Channar 20 267-270.
Conveyor Technology; bul .
792. [53] Dos SANTos, JA: High Angle Conveyors - HACs from Mine to Prep. Plant
and Beyond; bu solicfs handling, Vol. 1 3 (1 993) No. 2, pp. 303-309.
[24) J , G.L: Cost E active Design o Long Distance Conveyors; Proc.
9 , Conference 5, Cos E active Bu enals (54) BAHl<E, EA: Oevelopmen s of Belt Conveyor Systems; bulk solids han-
Aust ., Sydney, Aust. July 991 , pp. 32-39. d . Vol. 1 1 , (1 99 ) No. 1 , pp. 1 35- 1 45.

(25] JAMES, G.L.: Design o a 1 3. 1 km Overland Conveyor; bu sol [55) RAPPEN. A: Breaking a Lance for the Hose Belt Conveyors; bulk solids han­
ng,
Vol. 1 2, (1 992) No. 4, pp. 539-545. dling, Vol. 1 0 ( 1 990) No. 3, pp. 297-302.
(56) P.t.TZJ<E, J. and KH.: The Rollgurt-Conveyor, An Alternative Conveying
[26 HARAssoN, A.: on-Linear Processes and Chaos in Bel Col1veyO( Sys ems;
bul solids handling, Vol. 1 1 (1 99 ) . 1 , pp. 7 - 55. Sys em; bu solids handling, Vol. 1 0 (1 990) No.. 3, pp. 305-308.
(571 WACHTER, F.A.: Innovative Handling of Tailings Using the Pipe Conveyor
(27) HARA&soo, A, Teo, L.H. and RoeERTS, W.: rement of Belt- I
Sys em; solids hancrng, vo1. 1 o ( 1 990) No. 3, pp. 31 1 -31 4.
actions and Material Flexure Coe s tor Design of Troughed Conveyor
Systems; bul solids hand. , Vol. 7 (1 987) . 3, pp. 367-371 . (58) Jotw.issoN. R.: The Conveying Technique of the 1 990s. An Enclosed Belt
Conveyor which can run Round Sharp Comers; Proc. Bulk 2000 Conference,
[28) HAARsoN. A.: Troughabifity u.rement o F
The lnstn. of . Engrs., London, 1 99 1 , pp. 1 79- 1 79.
Powder and Bu Handfi Indus ry;
pp. 381 -384. (59) P J.W. and EMERroN, R.C.: Vertical FLEXOWELL Conveyors with 330 ft
( 1 00 m) Li Carry over 3 Million Tons of Coal for Illinois Coal Mine; bulk solids
[29] TALKS, M.G. and , P.: The Evaluabon o Some Fire- Greases as
hng, Vol. 6, (1986) o. 2, pp. 31 5-31 8.
Lubricants tor Conveyor Idler Bearings; Proc. Tooology in inemJ Extracoon
· [601 PAS.KE, J. ., EMauoN. R.C. and W , J.A.: Wyoming Coal Mine Truck
War on Wear Coof., Sept. 1 , I. .E., U o o ingham.
Load-out Facihty Carries 65,000 tons per month on FLEXOWELL Steep Angle
(30] HERRAlY, A.G.• and BAAs, J.C. .: Ro(fing Beanngs Of' Hostile Enwonments; Conveyors; bu sor s hand ng, Vol. 6 ( 986) No. 2, pp. 321 -324.
Proc. Tribology in Mineral Extraction ar on ear Coo ., Sep . 1 984,
1.Mech.E., University o o 1ngham. (61 I PAB.J<E, J.W.: 25 Years o Experience in Steep Angle and Vertical Conveying;
bu solids ing. Vol. 1 0 (1 990) o. 3, pp. 273-280.
[31 ] HERRAlY, 8. and , R.F.· Ider Rollers - ObseJvatioos and PredlCtions;
AntWl/ear 88, The Royal Socie , London, 988. pp. 27. 1 -27 .8. [62j WUNS0-1, D. and ScH.Ec:Hr, 8.: Discharge of Pocket-Belt Conveyors, Theory
and Experiment; bul solids handling, Vol. 1 1 (1 99 1 ) No.4, pp. 91 3-920.
(32] ROBERTS, A.W., PAPAUSl<I, D. and t-lAAA:sCN. • The Fnct.ion and Tension
Characteristics on Dciving Drums o Conveyor Belts; Proceedings. 2th In I. (63) READ, G.: Ai.r Supported Bel Conveyors; bulk solids handling, Vol. 5 (1985}
Power and Bulk Solids Handling Con erence, Chicago, U.S • y 988. 0. 5. pp. 1 07 1 - 1 076.
[33] HARRISON. A. and ROSERTS, A.W.: Mechanisms of Force Transfer 00 Conveyor (64) Roeems. A.W. and Wws, A.H.: Performance of Grain Augers; Proc. lnstn. of
Belt Drive Drums: bul solids hand ng , Vol. 1 2 ( 992) . 4, pp. 529-537. . Engrs., Vol. 1 76 (8) (1 962), pp. 1 65- 1 94.

[34] BENNEn. D.J. and ROBERTS. A. .: So6d and Conveyor 8elt In (65) Roemrs, A.W.: An Investigation of Grain Vortex Motion with Relation to the
During Transporta ion; Third Intl. Con. on Perloonance within Vertical Grain Augers; Proc. Jnstn. of Mech. Engrs, Vol.
and Transportation, IE US1). ewcastle, June 1 78 (1 ) (1 963-64) o. 1 2.

, D.J.: [66) RoeeRTs. AW. and AANolD, P.C.: Do Augers Damage Grain?; Power Fann­
[35} ROBERTS, A.W., OoMs, . and
ing, Vol. 74 (1 965) No. 9.
action in Relation o Belt Cleaning; solids
pp. 355-362. (67] ROBERTS, A.W. and , P.C.: Transverse Vibrations of Auger Conveyors;
Jnl. Agne. Engng. Res., Vol. 1 0 (1 965) No. 3.
[36) ROBERTS, A.W. and • D.J.: Friction,
Belt Cleaning Operaltons; An l'YVear 88, The (68) ROBERTS, W. and ARNc:W, P.C.: Critical Speeds of Single Span Screw or
pp. 28. 1 -28.8. Auger Conveyors: Mechanical Handling, London (April 1 966).
[37] , A.: Detecting Faiure in Vulcanised Conveyor 8elt SpllC8S, sol- [69) ROBERTS, A.W., HAYES, J.W. and Scorr, O.J.: Optimum Design of Continuous
ids handling, VOi. 4, ( 984) o.4, pp. 865-868. Conveyors; bul sorrds handling, Vol. 1 (1 981 ) No. 2, pp. 255-264.
·
(38} HAARlsGI, A.: A ew Development in Textile Belt onog· bu han- [70) Roeems, A. W.: Screw or Auger Conveyors Revisited; Proc. Powders and
dling, Vol. 8 (1 988) o. 2. pp. 23 1 -233. Solids Conference, Chicago, USA. 1 991 .
(39J � CHER, 8.: Kevlar Aramid - New Flbfe · foroemen or Conveyor [7 1 ) ROBERTS, AW.: Aspects of Attrition and Wear in Enclosed Screw Conveyors;
Belting; The South African Engineer, Vol. 33 ( 1 983) .6. pp. 1 44- 1 49. Proc. Powder and Bui Solids Conference, Chicago, USA, 1 993.
(40) Kcv.cz. Z. R.P.: Kevlar Aramid as a ReinOC'Cll'l9 Fibre in Conveyor Belts; Intl. (72) RAclEMAa-ER. F.J.C.: on-Spill Discharge Characteristics of Bucket Elevators;
Conf. on Bu k Ma erials Storage, and Transportation, Ins . of Powder Technology, 22, pp. 21 5-24 1 .
Engrs. Aust., 1 983, pp. 70-73. [73) ROBERTS, A.W.. ics of Bucket Bevator Discharge During the Final
(41 ] BETz, E.: The Use o 'Kevlar' Atamld Rbtes in Conveyor Belts, Part : Pul-Out Run-Out Phase; Jnl. of Powder and Bulk Solids Technology, Vol. 1 2 (1 988)
Problem and Splice Design; solids Vol. 6 ( 1 986) . 2, pp. . 2, pp. 1 9-26.
355-367. (7 I Y, G.J., AOSERTs, A.W. and HAYES, J.W.: Mechanics of High Speed El-
[42J Rooerrs, AW., Baz, E.. GoH, 8. ., LA • S.C. and W • W.C.: The Use of evator Discharge· but solids handling Vol. 3 {1 983) No. 4, pp. 853-859.
'Kevlar' Aramid Fibres in Conveyor Belts. Part 11: Sta ic Strength o Splices: (75) RAoa.w:::HeR, F.J.C.: Feeders and Vibratory Conveyors; TUNRA Bulk Solids
bulk solids handling, Vol. 8 (1 988) o. 4, pp. 429-44 1 . H ·ng Research Associates, The University of Newcastle, Australia, 1 980
.
[43 HOFFMAN, E.B.: Potyamide and Aromatic Fibers o he 1990s; (76) Cou.lN, H.: Mechanical Conveyors or Bui Solids; Bsevier, 1 985.
solids handling, Vol. 1 0, ( 1 990) o.3. pp. 283-288. [77] Roeems. A.W.: Investigation of Drag Chain Feeder Performance for Reclaim­
[44) BAHKE, T: Dimensioning and Application o Sett Conveyors ing Bu Solids from Bins and Stockpiles; Proc. Powders and Bulk Solids
Belt Drive (TT System); bu solids hand! ng. Vol. 2 { 982) Con erence, Chicago, 1 993.

281