Download as pdf or txt
Download as pdf or txt
You are on page 1of 4

M aking an impact at Kohat Cement

M arc Piccinin, Process Engineer, M agot t eaux, describes t he opt imisat ion
of a raw mill circuit using a vert ical shaf t impact or in a closed circuit in f ront of
the ball mill, and a Sturtevant ® SD high efficiency separator instead of a static one.

Introduction ball charge gradation would hard-


The Kohat Cement Company’s plant ly be an advantage.
is situated in the north of Pakistan • It would have been necessary to
near the Afghan border, 100 km from use a 1000 kW fan in case of a
Peshawar. In 1994, KCCL contacted pneumatical transport in the new
Magotteaux for a quote for the optimi- operating conditions.
sation of their raw and cement mills.
The plant target was to increase pro- From a specific consumption point of
duction of the existing Romanian line view, it certainly was not the best
from 1000 tpd to 1800 tpd of clinker. choice.
This article examines the factors The decision was also made to
The M ag’impact ® . keep the existing cyclones and fan in
related to the raw mill section.
the new configuration.
Before modification higher humidity content: gas was The other major modification pro-
The raw mill is an airswept mono- posed was the addition of a pre-
at 140 ˚C instead of 300-350 ˚C
chamber without a drying compart- crushing system in a closed circuit
normally used.
ment, whereby material leaving the before and on line with the mill. This
mill is drawn through a static separa- • Granulometry of the fresh materi- proposal was justified by the need for
tor. al; too coarse with 12% residues an important production increase
The tails are returned to the mill on 50 mm instead of 5% R30 mm. requested by the plant.
entrance, and the fines are passed • Important accumulation of All monochamber raw mills expe-
through four cyclones where the sep- uncrushed stones inside the mill rience the problems caused by
aration between gas and finished itself (in the first metre and in the excessively coarse fresh feed and
product is realised. The gas passes last part). inadequate internals.
to the main mill fan where part is recir- Scientific research shows that the
culated to the mill exit (in order to Proposed solution fresh feed size must be 0-25 mm or 0-
help the pneumatical transport of the Magotteaux held various meetings 30 mm, with a maximum residue of 5%
material), and the rest goes to the kiln with the plant management in order to between 25 mm and 40 mm.However,
electroprecipitator. establish objectives and preferences. these theoretical values are not prac-
The main technical data of the cir- A drastic modification of the existing tical given the well known problems
cuit before modification is given in installation by replacing the static facing the raw mill section, such as
Table 1. separator with a Sturtevant® SD third worn crusher hammers or no grid at
generation dynamic separator fed by the crusher outlet.
Problems before a bucket elevator was chosen. As a consequence there is often
modification There were two main reasons for an accumulation of uncrushed parti-
The whole mill circuit was checked this choice. cles inside the mill and a decrease of
before any proposals were made. • The retention time of the material the total circuit efficiency.
The following typical problems were in an airswept mill is always too The guarantee given by
found: short. Therefore, a finer fresh feed Magotteaux included an output of 140
• Inadequate drying gas tempera- coming from a precrushing sys- tph dry or 144 tph humid if there was
ture, especially in the case of tem, as well as a finer classified 3% moisture, and so, a 50% increase.

Table 1. Technical data


Mill Circuit Production data

• diameter: 3.8 m • static separator: 5.5 m • average fresh feed: 95 tph


• length: 7.68 m (L/D = 2) • 4 cyclones • composition:
• speed of rotation: 17.4 t/min (80.2%Vcrt) • mill fan: 220 000 m3/h, mix stockpile: 95-100%
• installed power: 1600 kW 80 ˚C, clay-limestone: 0-3%
• ball charge: 91 t of balls or 24.8%VL 650 mm WG, laterite: 0-2%
• initial ball charge gradation: ø90 mm 6.5 t 630 kW, • average moisture:
ø80 mm 11 t 1000 rpm 2.5% (up to 5-6% during the wet season)
ø70 mm 7.5 t • average fineness of product: 21-24% R90µ
ø60 mm 18 t • mill absorbed power: 1400 kW
ø50 mm 17 t • specific power (mill only): 14.74 kW/t
ø40 mm 19 t • mill inlet gas temperature:
ø30 mm 12 t +/-140˚ (gas coming from the cooling tower)
Total 91 t

1
Fineness on the 90µ sieve of 21% The raw material is precrushed and Table 2. Proposed internals of ball mill
residue was also assured. In order to the outlet product transported by Lining:
reach this guarantee, the flow-sheet another belt conveyor to the fresh on 1.50 m (or 20%): Monostep lifting
shown in Figure 1 was chosen. material conveyor. on 6.18 m (or 80%): Classodrift classifying
All components of the precrush-
New equipment ing system are thus linked by belt
Installation of an OPTIMEX outlet diaphragm
Ball charge:
The precrushing system conveyors. This system is more
102 t of ø90-80-70-60-50-40-30-25 mm or 28%
The fresh material is stocked in three extensive in a horizontal area than in volume load
small silos after the stock pile. Three a vertical one because the plant was
Detail of the charge at 80% and before
weigh feeders feed a conveyor belt to unwilling to use a bucket elevator in
optimisation (t):
the vibrating screen. A metal detector this circuit. 80 mm dia 15
is installed on this conveyor, but the Finally, the passing product from 70 mm dia 13
plant has now also installed a metal the vibrating screen feeds the ball 60 mm dia 1
extractor in order to avoid untimely mill. Both circuits are directly on line 50 mm dia 10
stoppages. without an intermediate silo. 40 mm dia 7
The vibrating screen, made by 30 mm dia 23
The ball mill
Haver & Boecker, has a capacity of Total: 79
The mill was equipped with a poor lift-
300 tph, an angle of 15˚, a first deck
ing lining without classification effect.
of 25 mm and last one of 10 mm. The mill gas also passes through the
The new internals proposed are
The material is screened in this separator and the same quantity leaves
described in Table 2.
section in order to get a passage of the circuit after the fan and enters the
0-10 mm. The tails (or material higher The separator circuit kiln’s electroprecipitator (Figure 1).
than 10 mm) fall from a conveyor belt The separator is a Sturtevant® SD120
onto the MAG’impact® precrusher. with air re-circulation. The existing Optimisation and results
The main characteristics of the cyclones and mill fan were used in First step at 80% of the ball
MAG’impact® model 2400 are: the new configuration. charge
The separator has the following
• Table diameter: 960 mm.
features:
The mill was producing an average of
• Impellers number: 5 (with H+ 110-115 tph with an average fineness
inserts). • Cage diameter: 2.94 m. of 18-20% residue on 90µ. In order to
• Installed power: 200 kW (1 motor). • Installed power: 110 kW. add the last 20% of balls, Magotteaux
had to carry out an axial test inside
• Maximum speed of rotation: 1500 • Cage speeds: 90-230 rpm.
the mill (Figure 2).
t/min. • Maximum air quantity: 200 000 m3/h.
These curves clearly show that
• Maximum feed size: 130 mm. • Fines output: maximum 180 tph. there are no uncrushed particles as
• Maximum output: 350 tph. • Feed quantity: 160-500 tph. from sample 3.

Figure 1. Proposed f low sheet f or t he raw mill at Kohat Cement , Pakist an.

2
Table 3. Final ball charge Table 4. First sample in precrushing system (80% of ball charge)
mm dia t % % of passing 1 mm 2.8 mm 4.75 mm 6.7 mm 9.5 mm 16 mm 20 mm 25 mm 31.5 mm
80 15 14.6 Fresh feed 25 39.4 49.6 57.5 65.9 79.3 84 87.4 91.5
70 14 13.6 Screen feed 25.8 39.3 50.3 59.1 69.1 82.8 87.1 91.7 95.4
60 12 11.6 VSI inlet 0.5 0.5 0.6 0.8 12.3 56 68.6 78.8 89.6
50 11 10.7 VSI product 20.1 42.1 58.9 72.6 82.9 93.7 95.5 97.1 98.3
40 12 11. Screen passing 34.9 59.6 77.2 90.6 100 100 100 100 100
30 28 27.2
25 11 10.7 Circul. load 1.36 1.52 1.54 1.54 1.54 1.64 1.70 1.64 1.79
Total: 103 Efficiency 19.7% 41.8% 58.7% 72.4% 80.5% 85.7% 85.7% 86.3% 83.7%

The last 23 t Table 5. Second sample in precrushing system (100% of ball charge)
could be a majority % of passing 1 mm 2.8 mm 4.8 mm 6.7 mm 9.5 mm 16 mm 20 mm 25 mm 32 mm
of small balls. The
Fresh feed 25.1 46.6 49.3 57.8 67.8 81.4 85 88.7 91.7
final ball charge is
shown in Table 3. Screen feed 19.8 38.4 49.7 57.9 67.3 80.5 87.5 91.1 94.1

A complete sam- VSI inlet 0.8 0.8 0.8 1.9 11.7 52.6 65 77 87.8
ple was also taken in VSI product 26.6 42.2 57.5 75 83.3 89.7 90.6 95.3 97.5
the circuit of the pre- Screen passing 34.6 62.4 77.9 90.2 99 100 100 100 100
crusher in order to
know the material
Circul. load 1.78 1.64 1.58 1.58 1.57 1.70 1.56 1.63 1.94
flows, the efficiency
of the MAG’impact ® Efficiency 26.0% 41.7% 57.2% 74.5% 81.1% 78.3% 73.1% 79.6% 79.5%
and the granulome-
try of the material
going to the mill (Table 4). • Mill power: 1270 kW. Final step at 100% of the
Prior to the mill crash stop, the ball charge
production data was as follows:
• Mill specific power: 10.95 kW/t.
• MAG’impact® speed of rotation: A second sampling in the precrushing
• Output: 116 tph. 1500 t/min. system was taken at maximum pro-
• Circulating load to the separator: • MAG’impact® power: 104 kW. duction. The results in Table 5 show
± 1.5 (A/F). that the system is stable and reliable.
• MAG’impact® specific The circulating load, efficiency
power: 0.9 kw/t of fin-
and particle size distribution of the
ished product.
material going to the mill are very sim-
100
• Total specific power: ilar to the results found at 116 tph.
11.85 kw/t. Figure 4 shows the particle size
80
• MAG’impact® specific distribution of the different samples at
Residue (%)

60 power on the precrushed 100% of the ball charge.


product: 1.65 kw/t. Before the sampling was imple-
40
The circulating load is rela- mented, the production data was as
20
tively low because the fresh follows:
feed is fine. A rate of 116 • Output: 151 tph.
tph means that only 63 tph
0
N.1 N.2 N.3 N.4 N.5 N.6
are precrushed in the
• Circulating load to the separator:
Samples number ± 1.65 (A/F).
MAG’impact® . The efficien-
0.090 mm 0.150 mm 0.212 mm 1 mm
cy on the sieve of 9.5 mm is • Mill power: 1560 kW.

2.80 mm 4.75 mm 6.70 mm

an impressive 80%. Figure Mill specific power: 10.33 kw/t.


Figure 2. The result s of t he axial t est inside t he 3 shows the particle size • MAG’impact® speed of rotation:
mill. distribution of the samples. 1500 t/min.

0.1 99.9
0.1 99.9
1 99 1 99
5 95 95
5
10 90 10 90
20 80 20 80
30 70 30 70
36.8 63.2 36.8 63.2
50 50 50 50
60 40 60 40
70 30 70 30
80 80
85 20 85 20
15 15
90 90
10
10
95
95
5
5 4
96
96 4 97 3
97 3
98 2
98 2
99 1
99 1
99.5 0.5
99.5 0.5 0.1 1 10 100 100
0.1 1 10 100 1000
dimension mm dimension mm

Figure 3. Part icle size dist ribut ion of samples at 80% of t he Figure 4. Part icle size dist ribut ion of samples at 100 % of t he
charge. charge.

3
• MAG’impact® power: 136 kW. • Total: 150.6 tph.
• MAG’impact® specific power: 0.9 • Fineness: 20.2% R90µ.
kW/t of finished product. • Moisture fresh feed: 1.9%.
• Mill+ MAGimpact® specific power: • Tails of separator: 137 tph (i.e. a
11.23 kw/t. circulating factor off 1.9).
• MAG’impact specific power of the • Separator speed: 33-34%.
precrushed product: 1.65 kw/t. • Mill completely in manual.
It was not possible to continue the tri-
Final result als because of problems on some old
The commissioning which began on existing auxiliaries.
the 13th March 1997 was achieved
with the following results. During 16 Conclusion
hours from 18.00 on 21/3/97 to 13.00 Improvements arising from the circuit
on 22/3/97, the average production modifications are summarised in
data was: Table 6.
• Mix stockpile: 147.75 tph. In conclusion, the modification of
• Laterite: 2.84 tph. a raw mill circuit by replacing the sta-
The M ag’impact ® in Kohat . tic separator with a high-efficiency
one and adding a precrushing sys-
Table 6. Improvements arising from circuit modifications tem can lead to an important produc-
Before modification After modification Improvement (% )
tion increase and a representative
efficiency improvement.
Output (tph) 95 150 58.5
Millpower (kW) 1400 1560 11.4
Acknowledgement
Precrusher power (kW) - 136 -
Magotteaux would like to thank the
Mill specific power (kW/t) 14.74 10.36 29.7 management and the personnel of
Precrusher specific power (kW/t) - 0.9 - Kohat Cement for their complete
Total specific power (kW/t) 14.74 11.26 23.6 co-operation in this successful pro-
Fineness ± 10 ject.

You might also like