2046 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
was studied, tested, and simplified in this work. The model
testing and verification were done with the industrial-scale
screen tests results made originally in 1998. The compari-
son between the results from the model and 1998s tests was
done by comparing the main parameters: the screening effi-
ciency, the throughput rate, the material bed depth along
the screen deck and the material cumulative size distribu-
tion for the oversize of the screen.
In the verification stage, the passing probability model
throughput was compared to 1998s test results throughput.
Four out of ten tests were within the selected error limit
(4%). The results of the model for the main parameters
indicate that it can predict the screening case precisely if
the screen is not overloaded.
The passing probability model can simulate screen-
ing situations when there are different sized apertures in
the screen deck, as the model divides the material bed on
the screen deck into consecutive calculation elements. For
each of the calculation elements, the size of the aperture
can be determined. It calculates the thickness of the mate-
rial bed that has a probability of passage. From this layer,
every discrete fraction is examined separately to calculate
the amount of undersized material that has the probability
of passage.
The passing probability was built so that it can predict
the screening efficiency, the throughput rate and the mate-
rial bed depth along the screen deck and the size distribu-
tions of the oversize and undersize material. In each of the
calculation elements, the screening efficiency is calculated
based on the amount of undersize material in the feed and
the amount of passed material from the calculation ele-
ment. The amount of passed material is the throughput
rate in a certain calculation element, which is calculated as
the amount of material in the actual passage layer multi-
plied by the probability of passage. The material bed depth
is calculated when the amount of feed material to the cal-
culation element, the bulk density, the width of the screen
and the transport speed are known. Because in every calcu-
lation element, these parameters are calculated separately,
the development of each of these main parameters can be
predicted along the screen deck. Also, the size distributions
of the undersize and oversize materials are calculated in all
the calculation elements. At the discharge end of the screen,
the size distributions can be predicted for a certain screen’s
overflow and underflow.
To integrate the model for software purposes so that
different sized apertures could be selected for the screen-
ing media, more work and time are needed. The developed
model works under regular conditions of charged material.
However, when the screen is overloaded, the model fails to
get enough precision. Also, precision in the model should
be tested for several decks on the screen (that was excluded
from this work). After that, the model could be integrated
into software.
When the model development stage is evaluated, it is a
fact that there were only ten tests that were suitable to study.
However, the implemented test results from 1998 are valu-
able because the throughput rate from each 1-meter-long
sector of the test screen deck can be calculated from the
amount of undersized material collected. The importance
of these 1998s test results is realized as the throughput rate
along the screen deck can be compared between the passing
probability model and the test results.
There were a small number of repeated tests, but for fur-
ther model development, it is required to implement more
industrial-scale tests, as were executed in the tests in 1998,
with many repetitions and sample series. Also, the screen
machine parameters and the screened material character-
istics, such as the screen rotation speed, the stroke length,
the transport speed of the material and so on, should be
documented precisely. The source of possible errors should
also be recorded. The downside of this implementation is
that it is expensive and time-consuming and there are many
possible sources of error.
There is a need to develop a model applicable to labora-
tory scale tests and equipment useful to scale up later. This
means that the sources of error can be managed better.
There is a possibility that the further development of
the passing probability model would lead to a change in
some design principles of the multi-slope screens because a
larger number of variables can be considered compared to
the traditional VSMA (1998) method.
REFERENCES
Metsälä, H. (2008) Probability Model for Screen Performance
Calculations. Tampere University of Technology.
Pohjasenaho, E. (2023) Passing Probability Model for Non-
Uniform Screen Deck.
Viilo, K. (1998) Test Report: Trible Slope 1800×6000 mm
Screen Tests. Tampere.
Viilo, K. (2020) ‘Screen Sizing Theories’. Tampere: Metso
Outotec. Viilo, K. (2022) Screening Model. Tampere.
VSMA (1998) Vibrating Screen Handbook. Milwaukee:
Vibrating Screen Manufacturers Association.
Whiten, W. (1972) Simulation and Model Building
for Mineral Processing. University of Queensland,
Queensland. 239 p.
was studied, tested, and simplified in this work. The model
testing and verification were done with the industrial-scale
screen tests results made originally in 1998. The compari-
son between the results from the model and 1998s tests was
done by comparing the main parameters: the screening effi-
ciency, the throughput rate, the material bed depth along
the screen deck and the material cumulative size distribu-
tion for the oversize of the screen.
In the verification stage, the passing probability model
throughput was compared to 1998s test results throughput.
Four out of ten tests were within the selected error limit
(4%). The results of the model for the main parameters
indicate that it can predict the screening case precisely if
the screen is not overloaded.
The passing probability model can simulate screen-
ing situations when there are different sized apertures in
the screen deck, as the model divides the material bed on
the screen deck into consecutive calculation elements. For
each of the calculation elements, the size of the aperture
can be determined. It calculates the thickness of the mate-
rial bed that has a probability of passage. From this layer,
every discrete fraction is examined separately to calculate
the amount of undersized material that has the probability
of passage.
The passing probability was built so that it can predict
the screening efficiency, the throughput rate and the mate-
rial bed depth along the screen deck and the size distribu-
tions of the oversize and undersize material. In each of the
calculation elements, the screening efficiency is calculated
based on the amount of undersize material in the feed and
the amount of passed material from the calculation ele-
ment. The amount of passed material is the throughput
rate in a certain calculation element, which is calculated as
the amount of material in the actual passage layer multi-
plied by the probability of passage. The material bed depth
is calculated when the amount of feed material to the cal-
culation element, the bulk density, the width of the screen
and the transport speed are known. Because in every calcu-
lation element, these parameters are calculated separately,
the development of each of these main parameters can be
predicted along the screen deck. Also, the size distributions
of the undersize and oversize materials are calculated in all
the calculation elements. At the discharge end of the screen,
the size distributions can be predicted for a certain screen’s
overflow and underflow.
To integrate the model for software purposes so that
different sized apertures could be selected for the screen-
ing media, more work and time are needed. The developed
model works under regular conditions of charged material.
However, when the screen is overloaded, the model fails to
get enough precision. Also, precision in the model should
be tested for several decks on the screen (that was excluded
from this work). After that, the model could be integrated
into software.
When the model development stage is evaluated, it is a
fact that there were only ten tests that were suitable to study.
However, the implemented test results from 1998 are valu-
able because the throughput rate from each 1-meter-long
sector of the test screen deck can be calculated from the
amount of undersized material collected. The importance
of these 1998s test results is realized as the throughput rate
along the screen deck can be compared between the passing
probability model and the test results.
There were a small number of repeated tests, but for fur-
ther model development, it is required to implement more
industrial-scale tests, as were executed in the tests in 1998,
with many repetitions and sample series. Also, the screen
machine parameters and the screened material character-
istics, such as the screen rotation speed, the stroke length,
the transport speed of the material and so on, should be
documented precisely. The source of possible errors should
also be recorded. The downside of this implementation is
that it is expensive and time-consuming and there are many
possible sources of error.
There is a need to develop a model applicable to labora-
tory scale tests and equipment useful to scale up later. This
means that the sources of error can be managed better.
There is a possibility that the further development of
the passing probability model would lead to a change in
some design principles of the multi-slope screens because a
larger number of variables can be considered compared to
the traditional VSMA (1998) method.
REFERENCES
Metsälä, H. (2008) Probability Model for Screen Performance
Calculations. Tampere University of Technology.
Pohjasenaho, E. (2023) Passing Probability Model for Non-
Uniform Screen Deck.
Viilo, K. (1998) Test Report: Trible Slope 1800×6000 mm
Screen Tests. Tampere.
Viilo, K. (2020) ‘Screen Sizing Theories’. Tampere: Metso
Outotec. Viilo, K. (2022) Screening Model. Tampere.
VSMA (1998) Vibrating Screen Handbook. Milwaukee:
Vibrating Screen Manufacturers Association.
Whiten, W. (1972) Simulation and Model Building
for Mineral Processing. University of Queensland,
Queensland. 239 p.