Minerals Engineering 23 (2010) 1213–1219

Contents lists available at ScienceDirect

Minerals Engineering
journal homepage: www.elsevier.com/locate/mineng

Spreadsheet-based modeling of liner wear impact on charge motion

in tumbling mills
M. Yahyaei, S. Banisi ⇑
Mining Engineering Group, Shahid Bahonar University of Kerman, P.O. Box 76175-133, Kerman, Iran

a r t i c l e i n f o a b s t r a c t

Article history: Grinding remains the major constituent of the total cost of processing minerals in most applications.
Received 6 July 2010 Charge motion is one of the key parameters affecting grinding efficiency and mill power draw. Although
Accepted 16 August 2010 there have been numerous investigations on the effect of liner design on charge motion, the effect of
charge profile due to liner wear during mill operation along the mill length has not received much atten-
tion. In this research, Powell’s analytical approach to calculate the charge trajectory with respect to the
Keywords: liner profile was used to develop a software based on Microsoft ExcelÓ . As a case study, the liner wear
Comminution
profile of the Sarcheshmeh copper complex SAG mill was used to model the liner wear rate and calculate
SAG milling
Modeling
the changes of lifter face angle and lifter height during mill operation. Results were then used to deter-
mine charge motion in the SAG mill at any given operation time. The results indicated that after 4000 h of
operation the lifters face angle increased from 14° to 47.1° and the height of lifters decreased from 15.2 to
5.8 cm. Modeling charge motion in the mill after 3000 h of operation showed 34° difference between the
maximum and minimum of angles of impact along the mill length due to the nonuniform wear profile. It
was also found that the variation in the pattern of the charge motion depends on the mill working hours
and the distance of the desired point from the feed end. It was determined that the ratio of spacing to the
height of lifters (S/H) plays an important role in the grinding efficiency and throughput. After 4000 h of
operation, S/H ratio of the Sarcheshmeh SAG mill increased from 1.7 to 4.6.
Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction Radziszewski, 1997; Van Nierop et al., 2001; Kiangi and Moys,
2006; Kalala et al., 2008). Mill liners designs affect the charge
Comminution is the most energy intensive operation which profile resulting in a specific grinding mechanism. Smooth liners
constitutes the major portion of operating and capital costs of increase the abrasion mechanism which results in a finer product.
the mineral processing plants. In the last decade, a vast number Lifters improve the lifting action of liners and prevails shatter and
of research works aimed at studying various aspects of grinding impact mechanisms in mill environment which results in a coarser
with the aim of performance improvement (Davis, 1919; Mishra product (Wills and Napier-Munn, 2007). A significant change in the
and Rajamani, 1990; Kalala et al., 2008; McElroy et al., 2009). balls impact points relative to charge toe due to liner profile and
Due to inherent complexity of grinding a comprehensive model resulting direct impact of balls to liners could lead to broken liners
to describe the process has not yet been proposed. One major rea- and subsequent costly mill downtimes (Banisi and Hadizadeh,
son has been the simplification of dynamic conditions of the charge 2007). Although a large number of studies have been carried out
inside the mill (Morrell, 1993). Laboratory studies have shown that on the effect of liner profile on the charge motion and power draw,
the assumption of a fixed shape for the charge is far from the real- the effect of liner profile change during the operation on the charge
ity (Liddell, 1986). There have been many attempts to quantify motion along the mill length has not received considerable
analytically the effect of various parameters on the dynamics of attention.
mill charge (McIvor, 1983; Vermeulen and Howat, 1986; Powell,
1991; Augustine and Moys, 2006; Makokha et al., 2007; Cleary, 1.1. Liner design and charge motion: a background
2009). Mill liners transfer the input power to the charge and play
a significant role on the motion of the charge (McIvor, 1983; It appears that White (1905) was the first person who tried to
Mishra and Rajamani, 1990; Moys, 1993; Cilliers et al., 1994; study the charge motion analytically. Later Davis (1919) related
the charge motion to power draw. Their works on determining
the charge trajectory were similar; they assumed the material
⇑ Corresponding author. Tel./fax: +98 341 2112764. inside the mill was a locked charge which moved in a circular path
E-mail address: [email protected] (S. Banisi). until it reached the point of equilibrium. Grinding elements start

0892-6875/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.mineng.2010.08.013
1214 M. Yahyaei, S. Banisi / Minerals Engineering 23 (2010) 1213–1219

their free flight after the point of equilibrium in a parabolic track soft ExcelÓ spreadsheet using the method proposed by Powell
until they hit the mill shell. In this type of study parameters such (1991). In this approach if the grinding media is positioned at the
as liner profile, friction forces and interaction of grinding elements tip of lifter it will start its free flight after the point of equilibrium.
are ignored. Otherwise, it will start to role or to slide on the lifter surface until it
Powell (1991) has analyzed the charge motion in detail. He ap- reaches the edge of the lifter. Then it will fall into free flight (Pow-
plied dynamics principles to determine the effect of liner profile on ell, 1991). When the grinding media reaches the edge of liner, the
charge motion. Nevertheless he ignored the interactions of grind- reference frame should change to a Cartesian coordination which
ing elements. McIvor (1983) also investigated the effect of liner de- its origin is positioned at the mill center to simplify the calcula-
sign on charge trajectory but he assumed that the outer layer of tions. The speed of the particle composed of its linear velocity par-
mill load will immediately projected into free flight after it reaches allel with the lifter face and angular speed resulted from mill
the point of equilibrium (McIvor, 1983). Taking the McIvor ap- rotation. These two types of speeds should resolve into compo-
proach Powell extended the study of the charge motion from the nents parallel with Cartesian axes to calculate the charge trajec-
point of equilibrium until it reaches the edge of the lifter. Mishra tory. The free flight of charge will end when it encounters the
and Rajamani (1990) have used discreet element modeling mill shell.
(DEM) method to predict not only the trajectory but also the inter- GMT in comparison with other software packages has the
actions of grinding elements. In recent years, DEM modeling has advantages of using all MS ExcelÓ readily available functions and
been the common approach to investigate the effect of operation capabilities. The main feature of GMT is the ability to show the kid-
parameters on charge motion and power draw in tumbling mills. ney-type shape of the charge along with the trajectory which has
Morrell (1993) studied the effect of liner profile, mill speed and not been incorporated in similar software packages. The ratio of
mill total filling on charge shape. He used a laboratory mill with spacing to the height of lifters (S/H) along the mill length is also
one transparent end and photographed the load under various predicated. Morrell’s (1993) approach was applied to determine
operating conditions. The variation of the toe and shoulder posi- the charge shape in which he calculated positions of the toe and
tions for three different liner profiles at various mill speeds and fill- shoulder and inner charge radius based on mill speed and filling.
ings were studied in his work. Then he proposed various empirical In other software packages, the shape of the charge is indicated
equations to relate the positions of toe and shoulder and inner load with a straight line with a constant angle which has been shown
radius to mill speed and filling (Morrell, 1993). In his work the load to be not very realistic (Morrell, 1993). Because of the differences
trajectory was not studied. in the calculation of charge trajectory for rods and balls an option
Radziszewski (1997) combined various relationships proposed has been provided which enables the user to choose the media type.
by other researchers to model charge trajectory for various layers A powerful error detection system has also been incorporated in
within the mill taking into account the slippage between layers. GMT which checks input data entry and prevents inclusion of inva-
He then used it to estimate mill power draw. He also modeled lid values in the calculation procedure. Fig. 1 shows the data entry
the liner wear by analyzing forces acting on liners using DEM. section of the software. There are three sections to be completed by
There have been other works regarding the factors influencing the user: mill, grinding media and liner data sections. A graphical
the charge trajectory (Moys, 1993; Cilliers et al., 1994; Valderrama legend has also been provided which defines the mill effective
et al., 1995; Morrell and Kojovic, 1996; Van Nierop et al., 2001; diameter, lifter width, lifter face angle and lifter height. Since in
Makokha et al., 2007; Kalala et al., 2008; Cleary, 2009). the literature various ways of defining these parameters have been
adapted, the legend prevents any possible confusion. In the mill
section, the values indicating the effective diameter, speed and fill-
1.2. Effect of liner wear on charge trajectory
ing are entered. In the grinding media section, the media type (i.e.,
ball or rod), diameter and density are recorded. In the liner section,
During grinding liner profile (i.e., lifter height and face angle)
lifter face angle, liner height, lifter width, number of lifters, static
changes gradually. The lifter face angle has been reported as the
friction and dynamic friction should be entered. Negative values
most effective parameter influencing the charge trajectory (McIvor,
or values out of a normal range will be prevented from entering
1983; Powell, 1991; Makokha et al., 2007; Rezaeizadeh et al.,
by the error detection system. For the media trajectory calculation
2010). It has also been found that the lifter height plays a role in
the values shown in Table 1 were used in the GMT software.
changing the trajectory of the charge outer layer (McIvor, 1983).
It is therefore necessary to study the relationship between the liner
2.2. The Sarcheshmeh copper complex SAG mill
wear profile and charge behavior in order to understand the varia-
tion of grinding performance during operation.
This study was carried out at the Sarcheshmeh copper complex
Banisi and Hadizadeh (2007) measured the 3-D liner wear pro-
located in south-eastern Iran. At the new concentration plant of the
file at the Sarcheshmeh copper concentration plant SAG mill using
Sarcheshmeh copper complex a 9.75  4.87 m SAG mill is used to
an especially-designed measuring device. They measured the liner
grind the product of a gyratory crusher where particles are smaller
wear profile regularly and entered the data into a mechanical soft-
than 17.5 cm. The SAG mill uses two 5500 hp synchronous motors
ware called SolidworksÓ to make the 3-D shape of liners at any gi-
and works with constant 10.5 rpm rotational speed in two direc-
ven operating time. Their results could be used to model liner
tions. The mill shell is lined with two series of 60-row liners. The
profile variations during operation. The objective of the present
mill liners are chrome-molybdenum steel casting type with a Brin-
study was to predict charge trajectory change due to the liner wear
nell hardness between 325 and 375. The lifter face angle, lifter
along the mill length for an industrial SAG mill using Powell (1991)
height and plate thickness are 14°, 152 mm and 78 mm, respec-
methodology.
tively. The liner length is 2084 mm which weighs 1130 kg. The
manufacturer recommends that the liners should be changed when
2. Materials and methods the lifter height reaches one third of its initial height (i.e., 51 cm).
The shape of liners was obtained by taking six measurements
2.1. Charge trajectory prediction software (profiles) with equal intervals along the length of the liner (Yahyaei
et al., 2009). Since each profile was characterized by 25 points
In order to determine the charge trajectory a software called corresponding to 25 needles of the measurement device, it was
GMT (Grinding Media Trajectory) was developed based on Micro- possible to find a model to describe the height decrease of each
 M. Yahyaei, S. Banisi / Minerals Engineering 23 (2010) 1213–1219 1215

Fig. 1. Data entry section of the GMT software.

each profile. Therefore, in total 150 (6  25) equations each show-

Table 1 ing the lifter height vs. time made the modeling of the liner profile
SAG mill characteristics used in GMT software to simulate charge motion.
shape change possible. The general form of the equations was:
Mill Mill speed (% Mill Ball Static Dynamic
diameter of critical total diameter friction friction hi ðtÞ ¼ hi ð0Þ þ k1i  t 2 þ k2i  t ð1Þ
(m) speed) filling (mm) coefficienta coefficienta
(%) where hi(0), hi(t) and kji are the initial height of the ith point, the
9.6 77 35 127 0.2 0.15 height after t hours of operation and constants at the corresponding
a
point, respectively (Yahyaei et al., 2009). In order to use the GMT
These values were taken from Powell’s work (Powell, 1991).
software to predict media trajectory, the lifter height and lifter face
angle were needed. These values were calculated upon knowing the
profile for any given operation time using Eq. (1).
point with operation time. At any given operation time, for any
profile, 25 point measurements were taken.
3.3. Lifter face angle and height change

3. Results and discussion In order to evaluate the accuracy of the proposed model to esti-
mate the lifter height for any given operation time, lifter heights
3.1. Media trajectory predication by GMT software for different profiles over the liner life time were compared with
the values obtained from the model. One criterion which is com-
Once the user has finished the completion of the data entry sec- monly used to evaluate the models accuracy is their corresponding
tion in the GMT software, the trajectory along with the charge correlation coefficients. The correlation coefficients for the models
shape could be calculated by pressing the ‘‘Simulate” button. A typ- of six profiles varied from 0.91 to 0.98 which indicated an accept-
ical result page of the software is shown in Fig. 2. In addition to the able accuracy (Yahyaei et al., 2009). This assured that the use of
graphical representation of the charge shape and trajectory some model to study the variations of lifter face angles and heights dur-
useful information regarding the impact point (Xi, Yi, ui), impact ing operation was legitimate. The lifter face angle was measured as
speed (Vxi, Vyi, VT), impact energy (Ek), positions of toe (t) and the angle between the lifter face and the direction which is normal
shoulder (s) and distance to height ratio of the lifters (S/H) are to the liner plate (Fig. 1).
also provided. The largest change in the lifter face angle during operation oc-
curred at the distance between 0.8 and 1.7 m from the feed end
3.2. Liner wear profile measurement: the Sarcheshmeh copper complex of the mill (Fig. 5). Due to the nonuniform liner wear profile along
SAG mill the mill length the lifter angle face for any given operation hours
varied significantly. For example, the largest change in the lifter
A typical 3-D model of a worn liner after 5384 h of operation is angle face for the operation hours of 3000 was from 14° to 34.6°
shown in Fig. 3. It is evident that during the operation lifter height which occurred between 0.8 and 1.7 m from the feed end. As ex-
and lifter face angle changes rather significantly. During the liner pected the change in the discharge end because of finer particle
life time (5384 h), four series of measurements were performed size and uniform distribution compared to the feed end was not
at 2120, 3259, 3759 and 5384 h of operation, respectively (for clar- significant. The comparison of the curves for the distances 1.7
ity only two of the four operation hours curves are shown in Fig. 4). and 4.2 m from the feed end in Fig. 5 clearly verifies the
In this manner, the heights of liners at any specific locations (i.e., observation.
profiles 1–6) and the corresponding operation times became avail- The change in the lifter height during the operation is shown in
able. A mathematical relationship was found to describe the height Fig. 6. The largest reduction in the lifter height occurred in the
decrease versus increase in operation time for any of 25 points on 0.8 to 1.7 m from the feed end while the discharge end did not
1216 M. Yahyaei, S. Banisi / Minerals Engineering 23 (2010) 1213–1219

Fig. 2. A typical result page of the GMT software.

Fig. 3. Typical 3-D model of new and worn liners after 5384 h of operation.

Fig. 5. Lifter angle change along the mill length during operation.

Fig. 4. The change of lifter profile (no. 3) over the liner life time.

experience any significant changes. The line showing the change

in the lifter height for the profile 4.2 m away from the feed end
is almost horizontal indicating a minor change during a period of Fig. 6. Lifter height change along the mill length during operation.
5384 h of operation. Given the significant changes in the lifter
face angle and height during operation it is expected that charge
trajectory also experiences considerable variations. by the GMT software. The values regarding the characteristics of
the mill were the ones shown in Table 1 and the lifter face angle
3.4. Charge trajectory change and height for any position along the mill length were calculated.
One such trajectory predication for a point 2.5 m away from the
The results obtained from modeling of the wear profile of the feed end for a period of 4000 h of operation is illustrated in
Sarcheshmeh SAG mill were used to predict the charge trajectory Fig. 7. The effect of liner profile change during the operation was
 M. Yahyaei, S. Banisi / Minerals Engineering 23 (2010) 1213–1219 1217

Fig. 9. Variations of charge impact point along the mill length for various operation
hours.

into consideration in determining the liner removal time rather

than the amount of scrap metal of liners which is common in most
plants. As it was shown, the wear profile along the mill length was
not uniform which makes the suggestion regarding the liner re-
moval time reasonable.
In order to study variations of the charge trajectory along the
mill length the trajectories were obtained after a certain operation
Fig. 7. Charge trajectories for a point 2.5 m away from the feed end for a period of time at distances of 0.8, 1.7, 2.5, 3.4., 4.2 m from the feed end,
4000 h of operation.
respectively. Fig. 8 shows the charge trajectories for an operation
time of 3000 h. The significant variations of the trajectories indi-
cate nonuniform liner wear profile along the mill length. The im-
pact point was measured in degrees starting from the horizontal
line passing the mill center (i.e., 3 o’clock position) and moving
counterclockwise. The largest difference in the angles of impact
points along the mill length was found to be 34°. It is interesting
to note that the shape of mill load remained constant. This is be-
cause the relationships used to calculate the mill load shape do
not take into consideration the change in the liner profile (Morrell,
1993). Further study is warranted to investigate this issue.
The variations of the impact angle along the mill length for
operation hours of 1000, 3000 and 4000, respectively, are shown
in Fig. 9. Since in the first 1000 h of operation the liner wear rate
is not significant, the charge impact point is approximately 230°
and does not change. In this region, cataracting is the prominent
mode of charge motion. As the operation hours increase and the
liner wear profiles become nonuniform impact angles vary. At
3000 h of operation the impact angle curve is not uniform anymore
and between 1.2 and 2 m from the feed end the contact angle is in
the range of 265–269°. This range corresponds to the area with
high rate of wear where the lifer height is low (7.8 cm) and face an-
gle is high (35°). The higher impact angles result from lower lifter
heights and higher face angles both indicative of worn liners. For
4000 h of operation curve, in the region between 0.8 and 2 m from
the feed end the maximum indicates where the face angle is max-
imum (47.1°) and lifter height is minimum (5.8 cm) both indicative
Fig. 8. Charge trajectories for various distances from the feed end an operation time of severe wearing. It was found that after 5000 h of operation the
of 3000 h. average impact angle reached 302° (excluding the first and last
profiles) which clearly showed rolling down the charge to toe. Im-
pact angles above 300° indicate conditions where the balls do not
found to be significant in particular after 3000 h of operation. If this have any vertical free flight and after the equilibrium point roll
change in the trajectory is not compensated by other factors such down. This cascading led to finer grinding and increased liner wear.
as increasing the mill speed it could have an adverse effect on As a result, a significant reduction in the mill throughput was expe-
the efficiency of the grinding. This problem is of importance in rienced and the plant had to remove liners after 5384 h of
the case of the Sarcheshmeh SAG mill where a constant speed drive operation.
is used. The charge shape which is also shown in Fig. 7 (dotted One criterion which is commonly used to study the efficiency of
lines) suggests the possibility of direct impact of balls and liners. grinding is the lifter spacing to height ratio (S/H). When the liners
The broken liners detected at the early weeks of the operation were are installed the S/H ratio is low (i.e., 1.8–2) because the lifter
indicative of the problem. The charge trajectory should be taken height is at the maximum. As the operation continues the liner
1218 M. Yahyaei, S. Banisi / Minerals Engineering 23 (2010) 1213–1219

4. Conclusions

1. A spreadsheet-based software was developed to model the

charge trajectory in tumbling mills using the methodology pro-
posed by Powell (1991).
2. The 3-D liner wear profiles at various operation hours were
used to predict the charge trajectory along the mill length in a
9.75  4.88 m SAG mill at the Sarcheshmeh copper complex.
3. By 3-D modeling of liner wear profiles it was found that after
3000 h of operation, at the areas with high wear rates, the lifter
face angle increased from 14° to 35.2° and lifter height
decreased from 15.2 to 8 cm.
4. The largest difference in the angles of charge impact points after
3000 h of operation along the mill length due to nonuniform
liner wear profile was found to be 34°.
Fig. 10. Variations of the S/H ratios along the mill length for three different
operation hours. 5. Modeling of the charge trajectories indicated that the amount of
variation not only was related to the operation hours but the
location of the measured profile along the mill length also
played an important role.
6. There is a ratio of lifter spacing to height ratio (S/H) in which
the mill throughput is maximum; for the Sarcheshmeh SAG mill
the ratio was found to be between 3 and 4.
7. By using detailed liner monitoring techniques linked to mill tra-
jectory software the operation of large SAG mills can be
improved through adjusting the mill speed considering existing
liner profile, on time replacement of liners and designing non-
uniform liners based on the liners wear pattern.

Acknowledgements

The help received from Dr. Morrell, S. Mohammadi, and M.

Hadizadeh in understanding the charge trajectory and shape equa-
tions and also liner wear profile measurements are greatly
Fig. 11. Relationship between the mill throughput and the S/H ratio for the acknowledged. The authors would like to thank National Iranian
Sarcheshmeh copper complex SAG mill. Copper Industries Company (N.I.C.I.Co.) for supporting this re-
search and permission to publish this article. Special appreciation
is also extended to the operating, maintenance, metallurgy and
wear becomes significant resulting in higher S/H ratios. Over the R&D personnel for their continued help.
years various S/H ratios have been proposed for new liners but
the recent trend has been between 3 and 4 for low–medium–high
lifter arrangements (Meekel et al., 2001). The main concern regard- References
ing the S/H ratio is the material packing between lifters as the S/H Augustine, B.M., Moys, M.H., 2006. Towards optimizing ball milling capacity: effect
ratio decreases. The variations of the S/H ratios along the mill of lifter design. Minerals Engineering 19, 1439–1445.
length for three different operation hours are illustrated in Banisi, S., Hadizadeh, M., 2007. 3-D liner wear profile measurement and analysis in
industrial SAG mills. Minerals Engineering 20, 132–139.
Fig. 10. As the operation hours increased from 1000 to 3000 the
Cilliers, J.J., Austin, L.G., Leger, P., Deneys, A., 1994. A method of investigating rod
S/H ratio increased especially in areas where the wear rate was motion in a laboratory rod mill. Minerals Engineering 7 (5/6), 533–549.
high. The area between 1.2 and 1.6 m from the feed end showed Cleary, P.W., 2009. Ball motion, axial segregation and power consumption in a full
the highest value of the S/H ratio; this area had the highest rate scale two chamber cement mill. Minerals Engineering 22, 809–820.
Davis, E.W., 1919. Fine crushing in ball mills. AIME Transactions 61, 250–296.
of wear. When operation hours reached 5000, the lifter height re- Kalala, T.J., Breetzke, M., Moys, M.H., 2008. Study of the influence of liner wear on
duced to 3.2 cm which was far from the original value of the load behavior of an industrial dry tumbling mill using the discrete element
15.2 cm. This resulted in very high values of S/H ratios (7–7.2) indi- method (DEM). International Journal of Mineral Processing 86, 33–39.
Kiangi, K.K., Moys, M.H., 2006. Measurement of the load behaviour in a dry pilot mill
cating cascading mode for the charge. This suggested that the lin- using an inductive proximity probe. Minerals Engineering 19, 1348–1356.
ers should have been changed before 5000 h of operation. Some Liddell, K.S., 1986. The effect of mill speed, filling and pulp rheology on the
researchers have reported a correlation between the S/H ratio dynamics behaviour of the load in a rotary grinding mill, MSc thesis. University
of Witwatersrand, Johannesburg, South Africa.
and the mill throughput (Meaders and MacPherson, 1964; Mako- Makokha, A.B., Moys, M.H., Bwalya, M.M., Kimera, K., 2007. A new approach to
kha et al., 2007; Cleary, 2009). The relationship between the mill optimizing the life and performance of worn liners in ball mills: experimental
throughput and the S/H ratio for the Sarcheshmeh copper complex study and DEM simulation. International Journal of Mineral Processing 84, 221–
227.
SAG mill is shown in Fig. 11. The tonnage reported is the average McElroy, L., Bao, J., Yang, R.Y., Yu, A.B., 2009. A soft-sensor approach to flow regime
tonnage up to the time of liner profiles measurement. For a given detection for milling processes. Powder Technology 188, 234–241.
throughput the maximum S/H ratio was selected. The optimum McIvor, R.E., 1983. Effects of speed and liner configuration on ball mill performance.
Mining Engineering 1, 617–622.
range for the S/H ratio was found to be between 3 and 4. For low
Meaders, R.C., MacPherson, A.R., 1964. Technical design of autogenous mills. Mining
S/H ratios there is a possibility of material packing between the Engineering 9, 81–83.
lifters which reduces the intensity of impact mechanism. For high Meekel, W., Adams, A., Hanna, K., 2001. Mill liner development at Highland Valley
values of S/H ratios charge cascading is predominant due to severe copper. In: Proceedings International Autogenous and Semiautogenous
Grinding Technology, vol. 3, pp. 224–239 (vancover).
liner wear and reduced lifter height. This also results in a lower Mishra, B.K., Rajamani, R.K., 1990. Numerical simulation of charge motion in ball
grinding efficiency. mill. In: Proceedings 7th European Symposium on Comminution, pp. 555–563.
 M. Yahyaei, S. Banisi / Minerals Engineering 23 (2010) 1213–1219 1219

Morrell, S., 1993. The prediction of power draw in wet tumbling mills. PhD thesis, Valderrama, W., Grandela, P., Magne, L., 1995. Liner shape and power consumption
University of Queensland, Brisbane. in rotary mills. In: Proceedings The XIX International Mineral Processing
Morrell, S., Kojovic, T., 1996. The influence of slurry transport on the power draw of Conference, pp. 177–182.
autogenous and semi-autogenous mills. In Proceedings International Van Nierop, M.A., Glover, G., Hinde, A.L., Moys, M.H., 2001. A discrete element
Autogenous and Semiautogenous Grinding Technology, vol. 1, pp. 373–389 method investigation of the charge motion and power draw of an experimental
(vancouver). two-dimensional mill. International Journal of Mineral Processing 61, 77–92.
Moys, M.H., 1993. A model of mill power as affected by mill speed, load volume, and Vermeulen, L.A., Howat, D.D., 1986. Fluctuations in the slip of the grinding charge in
liner design. The Journal of South African Institute of Mining and Metallurgy 93 rotary mills with smooth liners. International Journal of Mineral Processing 16,
(6), 135–141. 153–168.
Powell, M.S., 1991. The effect of liner design on the motion of the outer grinding White, H.A., 1905. Theory of tube-mill action. Journal of Chemical, Metallurgical and
elements in a rotary mill. International Journal of Mineral Processing 31, 163– Mining Society of South Africa V, 290.
193. Wills, B.A., Napier-Munn, T.J., 2007. Will’s mineral processing technology, seventh
Radziszewski, P., 1997. Ball charge dynamics and liner wear simulation. In: ed., Elsevier Science and Technology Books, pp. 147–149.
Proceedings 29th Annual Meeting of CMP, pp. 91–107 (Quebec). Yahyaei, M., Banisi, S., Hadizadeh, M., 2009. Modification of SAG mill liner shape
Rezaeizadeh, M., Fooladi, M., Powell, M.S., Weerasekara, N.S., 2010. An experimental based on 3-D liner wear profile measurements. International Journal of Mineral
investigation of the effects of operating parameters on the wear of lifters in Processing 91, 111–115.
tumbling mills. Minerals Engineering 23 (7), 58–562.