04 Milling

Download as pdf or txt
Download as pdf or txt
You are on page 1of 21

CEMENT PROCESS ENGINEERING

VADE-MECUM

4. MILLING

Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

Table of Contents
1.

Mill ....................................................................................................... 4.1


1.1 Ball Mill General .......................................................................... 4.1
1.2 Ball Charge and Internals.............................................................. 4.3
1.3 Ball Charge Design (Finish Mill)................................................... 4.6
1.4 Grinding Laws.............................................................................. 4.9

2.

Separator............................................................................................ 4.12
2.1 Circulating Load......................................................................... 4.12
2.2 Tromp Curve.............................................................................. 4.12
2.3 Indicators for Cement Milling and Typical Values ....................... 4.14
2.4 Sturtevant/O'Sepa (Bath #B mill) ................................................ 4.15
2.5 Recommended Steps for Sizing a HES ........................................ 4.16

3.

Heat and Water Balance .................................................................... 4.17

4.

Grinding Aid ...................................................................................... 4.18

5.

Sieve ................................................................................................... 4.19

Index - i
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

1. Mill
1.1 Ball Mill General
a) Mill design
General L/D ratio
Raw mills: 1.5 < L/D < 3.2
Finish / cement mills: 2.8 < L/D < 3.2
Length of first Compartments relative to total mill length
Raw mills: First compartment length equals 35 45% of total mill effective length.
Cement mill: First compartment length equals 30 35% of total mill effective length.
When L/D>1.5, classifying liners might be used.
The lower the L/D, the higher the circulating load needs to be (see below).
b) Percent loading of mill
2
r 2 r sin (h r )
% volume load = 360

r 2

where:
- r is the radius
- h is the free height
hr
- = arccos
r
- in degrees = 3.14

0.9
h/d
0.8
0.7
0.6
0.5 0

10

20

30

40

50%

% volume load

Rules of thumb
% vol. Load = 111.87 123.98 (h/d), 25 50%: error max 0.6%.
It is estimated that material increases the actual ball filling ratio by about 2%.
Another method (quick but not as accurate) consists in counting the number of visible shell liner plates (n) and
to divide by the total number of shell liner plates per circumference (N): Angle = n x 360 / N.
Values of angle h/d ratio in relation to the ball load (% filling degree)
Ball load (%)
20
21
22
23
24
25
26
27
28
29
30

h/d
.7459
.737
.7281
.7193
.7106
.702
.6926
.685
.6765
.6682
.6598

n/N
.667
.653
.639
.625
.611
.601

Ball load (%)


31
32
33
34
35
36
37
38
39
40
41
42

h/d
.6516
.6434
.6352
.627
.6189
.6109
.6028
.5948
.5868
.5789
.5709
.563

n/N
.590
.580
.569
.558
.549
.539

4.1
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

c) Mill Critical Speed

C
m
P

000000000000000000000000000000000000000000000000000000000000000000000000
r

60 2 g

Mill critical speed: nc =

4 2 r

C = m 2 r =

G 2 r
g

where:
- G = Weight of grinding ball in kg
- = angular velocity of mill tube (rad/sec)
- D = inside mill diameter (m)
- n = rev per minute
- C = centrifugal power kg
- D1 = inside mill diameter (ft)
P = G * sin
(P is the resulting force of gravity)
To maintain the ball in this position on the mill wall, it is
necessary that C P.

76.6
42.3
(=
), with D in meters ( D1 in feet)
D1
D

% Critical speed:
Practically, mill speed between 68and 80% of critical speed.
% critical speed is the mill actual speed in RPM divided by nc.
Example:
3.98 meter mill with rotational speed of 15.6 rpm then nc = 21.2, % critical speed = 73.6 %.
d) Retention Time

Rules of thumb:
Retention time:

Open circuits: 12 min


Closed circuits: 5 min
The feed is pushing the material through the mill so that, If mill throughput increases: retention time decreases:
C
8 <
< 12
where: C is the ball charge weight, M is the material weight
M

Fluoresceine test:
2g/t of mill production. Prepare the fluoresceine with 800-ml alcohol and impregnate 2 kg of mill feed
material (in a plastic bag).
Put the material at mill inlet, start the time and sample every 30 s during 30 min. (others use salt).
e) Mill Throughput
Using elevator power and after calibrating we have:

A=

(kW kW0 ). 3600 .


9,81. H

where:
A
kW
kW0

H
-

=
=
=
=
=

Material flow (mtph)


Actual elevator power ( in kW)
Elevator power empty
Elevator efficiency
Inter axis elevator height

4.2
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

f) Required air velocities for mill ventilation

Rules of thumb
Recommended 1.5 m/s above the ball charge:
inside the trunnion: 22-25 m/s.
partitions: 8-14 m/s (<20 m/s).
hood: <5 m/s to prevent dust from being sucked up (dust pick-up is proportional to speed^2).
dropout box: <2 m/s.
0.3-0.5 Nm3/kg ck
0.6-0.8 Nm3/kg raw mix
Wet bulb temperature should be 30oC below the dry bulb temperature.
g) Optimum filling ratio:
U= (volume of powder in the mill)/ (volume of voids in the charge): between 60% and 110%, optimum around
90%.
In practical terms, material level should equal ball level.

1.2 Ball Charge and Internals


a) Biggest Ball

Bond Formula

d KMAX

where:
- d KMAX is the biggest ball diameter (mm)
- d 20 is the sieve dimension () with 20% retained

Quick evaluation
For clinker:

K is a constant (350 for a dry mill open or close circuit,


300 for wet)
is the specific mass of material (g/cm3)
Wi is the Bond work index (kWh/t)
Du is the mill inside diameter (m)
is the ratio between the actual / critical speed (%)

Grinding Ball vs Clinker Size

B = 24 d 80
(Other formula exist that result in value differences
of 5%)
B = ball dimension (mm)
d 80 is the sieve with 80% passing
-

Optimum Ball Diameter (mm)

d
Wi
= 20.17 20 .3
K . Du

100

10
.1

10

100

Clinker Size d80

Rowland Formula

. Wi
d 80
.
K 100 . . 3.281 Du

B = 25.4

d 80 is the sieve with 80% passing

4.3
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

Material bulk density and Bond index


kWh/st
(g/cm3)
Wi
Clinker:
Limestone
Shale
Slag
Sand stone
Silica sand
Coal
Clay
Gypsum
Kiln feed

13.49
10.18
16.40
15.76
11.53
16.46
11.37
7.10
8.16
10.57

Bulk density
Sand
Sand
Iron
Bauxite
Brick
Gypsum
Fluid coke
Limestone (crushed)
Silica fume
Bottom Ash
Cement T I-II
T 10
T III
Clinker
Clinker
(underburnt)
Raw mix

3.09
2.68
2.58
2.93
2.68
2.65
1.63
2.23
2.69
2.67

g/l or kg/m3
1387
1679
2629
1980
1502
1677
926
1803
1024
1241
1234
1207
1054
1575
1400

lb/ft3
86.6
104.9
164.2
123.6
93.8
104.7
57.8
112.6
63.9
77.5
77.1
75.4
65.8
98.4
87.4

1041

65.0

b) Grinding Balls Data


Grinding Ball dimensions
Weight
Surface
Diameter
(g)
(cm2)
mm
inch
4,001.153
314.159
100.00
4"
2,916.841
254.469
90.00
3"
2,048.590
201.062
80.00
1,826.658
186.265
77.00
3"
1,372.396
153.938
70.00
1,048.878
128.680
64.00
2"
864.249
113.097
60.00
500.144
78.540
50.00
2"
256.074
50.265
40.00
219.551
45.365
38.00
1"
171.549
38.485
35.00
128.061
31.669
31.75
1"
108.031
28.274
30.00
62.518
19.635
25.00
1"
48.682
16.619
23.00
43.895
15.511
22.22
=7/8"
32.009
12.566
20.00
3/4"
19.658
9.079
17.00
5.8"
(Unit weight and specific surface of MAGOTTEAUX grinding media)

Quick calculation:
Ball diameter (mm) =

250 P

Specific surface of balls of diameter =

Number of balls per


metric tons

Weight of 1 m3 of
balls (kg)

Specific surface
(m2 / mt)

250
343
488
548
729
954
1,157
2,000
3,905
4,555
5,830
7,809
9,257
15,996
20,542
22,782
31,242
50,870

4560
4590
4620

7.854
8.728
9.812
10.207
11.222
12.276
13.085
15,708
19.628
20.664
22.437
24.730
26.173
31.408
34.139
35.337
39.259
46.185

4640
4660
4708
4760

4850
4894

4948
4989

(P = weight in g)

785 2
m / mt
d

(d = diameter in mm)

4.4
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

Wear rates:
Ball diameter mm
100
90
80
70
60
50
40
30
25
20
17

SECTION 4 MILLING

Wear GT ball
g/h.T

Wear / Ball
g/100h

Wear diam
mm/100h

12.9
14.4
16.7
19.3
23.1
28.9
38
46.5
58.5
68.2

38.6
30.4
23.6
17.2
11.9
7.6
4.22
2.98
1.92
1.38

0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4

Bulk density for ball load


(coarse to medium ball size distribution):
In first compartment; 4.3 4.5 metric tonnes per tonne of balls (3.0 to 2.0 inch balls would be fine).
In second compartment; 4.5 4.65 metric tonnes per tonne of balls (2.0 to 0.75 inch balls would be considered
medium to fine).
In single compartment: 4.5 4.55 metric tonnes per tonne of balls.
c) Others internals
Partitions
Total slot area: 10 to 20 cm2/tph production:
Slot Size
Central Part
Discharge Part
FM
9 mm 1 mm
7 mm 1 mm
Max: min ball size
RM
12 mm 1 mm
10 mm 1 mm
Liners
Liners must be changed when 60% of their effective lifting height has worn away:
-8 to 10 % production
reference points to measure lifting height are the lowest point on the liner to the highest release point
(contact points between grinding ball and liner plate)
American Lorrain pattern: diameter (ft)*2=# bolt holes/row, 18.8 center to center.
DIN pattern: diameter (m)*10== # bolt holes/row, 31.4 cm center to center.
Classifying liners if L/D>1.5 and volume load<35%.
Without classifying liners, keep a maximum of 3-4 ball sizes.

4.5
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

d) Mill Internal Inspection Sheet


Ball Charge Remarks
Ball Coating Remarks
Ball Classification Remarks
Discharge Grate Slot Size-Average
Discharge Grate Slot Size-Maxim.
Discharge Grate Metal Thickness
Discharge Grate Percent Blinded
Discharge Center Screen Percent Blinded
Material Position in Mill

Points to audit
Shell Liner Thickness
Shell Liner Lifter Thickness
Shell Liner Remarks
Inlet Head Liner Thickness
Inlet Head Liner Remarks
Inlet Opening Remarks
Height Liner, to Balls - Average
Width Across Balls - Average
Calculated Percent Fill

1.3 Ball Charge Design (Finish Mill)


a) Recommended volume loading
(see BP Ball Charge Management)

Recommended Volume Loading


1 Compartment
2nd Compartment
3rd Compartment
Minimum kWh/t1
26 28%
28 30%
28 30%
Maximum Production
32 34 %
34 36%
34 36%
(Ball level in the trunion should not be higher than 2 to 3 inches.)
st

b) Polysius Design
As a rule of thumb, it suits raw mills and especially monochambers very well, especially if no classifying
liners are used.
D
ln
9.6
D = 9.6 e013.x x =
0.13
where:
D = ball (cm)
x = effective mill length (m)
Process step-by-step, calculating each effective length starting from the input and with the largest ball:
1. Calculate effective lengths and the ball sizes you plan to use.
2. Double the first effective length which is both the first interval width and the first cumulative length.
3. Calculate each succeeding interval width by taking the effective length and substract the preceding
cumulative length and doubling it. Add this value to the previous cumulative length to get the new one.
4. If an interval overlaps with the partition divide the interval at the point of overlap. The excess is carried
over to the next compartment. At the end of the mill, the interval is truncated at the point of overlap.
5. Once the intervals have been adjusted for compartment lengths as described in step (4), divide the adjusted
interval by compartment length and multiply by 100. This is the percent weight for each size to be used in
the compartment.

c) Slegten Model
1
The recommended volume loading for minimum kWh/t is based on an acceptable compromise with production. For minimum kWh/t
the volume loading can be as low as 22%.

4.6
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

Compatible mostly with classifying linings in the second compartment.

4.7
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

First Compartment Crushing


Same number (n) of balls in each size range.80, 70 and 60 mm and then add some 90 mm to deal with
oversize clinker. This equilibrium charge will not change as you add 90 mm make-up balls to maintain
volume load.
Ball (mm)
% of Weight (x)
% of Weight
90
100-5x
20.0
80
2-4x
38.4
70
1.6x
25.6
60
x
16.0
- x = is taken to be the number of balls in the last size.

Number/ 10 t of Charge
670
1820
1820
1820

In recent years, Slegten has favored a 3-ball size distribution in first compartments over a 4- ball size as shown
in table above.

Transition Zone
This is at the beginning of the second compartment and basically its job is to clean up anything which
penetrated the partition that is oversize for the second compartment charge to fracture.
The design for this area is to use "n" balls of 50 and 40 mm.
Ball (mm)
50
40

Number/ 10 t of Charge
1820
1820

The transition zone is made of the largest ball size used in this transition zone is sometimes identical to the
smallest ball size used in the first compartment.

Second Compartment Fine Grinding


The envelope curve for the balls smaller than 40 mm follows the following formula:
where:
D = 3.3e 010.x
D = ball (cm)
x = distance from transition zone finish (m)
The 30 mm balls start at the completion of the transition zone and the exponential curve follows.
Rule of thumb:
The smallest ball size should, as a minimum, be at least twice the width of the slots in the grates (ex. 16 mm
balls if slots are 8 mm wide). For this reason, it is generally recommended to use (19 mm) balls as the
smallest size in Finish mills. 5/8 balls are fine when the grates are new but often become problematic as the
grate slots enlarge.
d) Example: Comparison Slegten/ Polysius
1st compartment useful length = 3.81 m, 2nd compartment useful length = 7.66 m
Using an average ball weight of 1.65 kg per ball and 3 ball sizes in the first compartment for the Slegten model.
1st compartment
2nd compartment

Ball size and % compartment load


3
3
2
2
1
1
1

5/8

Polysius design
31.0%
31.2%
37.8%
2.31%
23.73%
34.05%
2.57%
37.34%
-

Slegten design
32.1%
43.1%
24.8 %
7.67%
2.94%
10.08%
48.18%
31.13%
(Some)1

Transition
zone

4.8
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

A limited amount of 5/8 balls should theoretically be added but the designer decided to use as the smallest ball size.

e) Fineness in Finish Mills:

In the first compartment before intermediate diaphragm


95% passing of 2.365 mm (2360 or 8 mesh) for the material leaving the first compartment 33% of energy.
Particle size distribution recommended on other sieves:
- 86 92 % passing 1.0 mm (1000 m or 18 mesh)
- 80 90 % passing 0.6 mm (595 m or 30 mesh)
- 75 85 % passing 0.5 mm (500 m or 35 mesh)
In the second compartment before discharge diaphragm
95% passing 0.5 mm (500 m or 35 mesh)
70- 80 % passing 0.2 mm (212 m or 70 mesh)

1.4 Grinding Laws


a) Absorbed Power of a Mill
Only 5-10 % of the energy is used for grinding, 90% is wasted into heat, wear
With similar ball charge gradation and similar liners' lifting effect, the absorbed power is related to:
Tonnage of balls
Mill rpm
% volume load
Mill diameter

Slegten formula

P =W*

rpm
V cr

1.27

W=

and

* K j * K Fr

* Fr 2 * L * J * d

where:
P : the motor absorbed power (kW)
J : the ratio between the apparent ball
W : the weight of the load (T)
volume and the internal volume
- rpm: is mill speed (rpm)
Fr : internal diameter (inside liners) (m)
d is the apparent density of load (t/m3)
#1 comp :
d = 4.5
#2 comp :
d = 4.65, if fine ball size distribution (say with average ball weight < 40 g)
d = 4.6, if coarser ball size distribution (average ball weight > 40 g)
Average :
d = 4.6
Vcr is the critical speed inside liners=

42.3
Fr

, L : the useful length of mill (m)

K j = 1.36 1.2 J , K Fr = C .Fr

K Fr is the influence of the location of the center of gravity for the moving ball charge vs. the mill center
(C is a constant depending on the material and the liners).
C=
11.262 for Clinker mill closed circuit with Slegten equipment
10.7 for clinker + slag, 12.16 for raw mix, 10.1 for slurry

0.379

4.9
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

P = L*

rpm
Vcr

1.27

* J*K j*

SECTION 4 MILLING

* Fr2.379 * d * C

Simplified formula

P =T *

Fr
RPM 100
*Kj *
* 9.5
*
Vcr
75
1.366

Kj Function of Volume Load


Volume load
Kj
40%
0.9
30%
1
20%
1.1
Rules of Thumb
One metric ton of balls increases the mill power draw by 10kW.
Usually, 8 to 12 kWh/t is absorbed in the first compartment for clinker grinding.
b) Grinding Laws

General Law: Charles

dW = cx n dx
- If W = Comminution work,
particles (initial, final)

Value of n
Energy Law
Rittinger
Kick
Bond

Value of n:
2
1
1.5

x = Size of

Applies well over range of:


10 1000 m

Normalized Blaine fineness equation


Fineness equation, generally accepted within Lafarge Corp:

Blaine 1
W2 = W1
Blaine 2

n = 1.4 for high efficiency separator (HES) circuit, n = 1.6 for Sturtevant separators, bearing in mind
that 16 and 18 Sturtevant separators are more efficient than the larger 20 and 22 Sturtevant.
W: communition work, W is proportional to production rates.

Proposed by Polysius: C2 = C1 * e0.43( Blaine 1 Blaine 2 ) / 1000


where C2 and C1 are production capacities
Renes Study: +1% passing at 10m: +10.8 SSB

Rules of thumb
Raw material: 10-16 kWh/t (target fineness: passing 200m>99%, passing 100m>90%)

4.10
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

Clinker: 45 15 kWh/t at 3500 SSB

4.11
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

2. Separator
2.1 Circulating Load
a) Junction with Three Streams
A, R, F are the feed, reject and fine of the separator
A
- ai , ri , f i are the cumulated % passing at a defined sieve(i).
- da, dr, df are the % retained corresponding to the sieve interval dx.

A=
A da

With:

R+F
=
Rdr + Fdf
da = ai + 1 ai ,

b) Drawing
Plot ( f i a ) vs ( f i ri )
If the mill circuit is steady, the graph has to be a
straight line:
( f a) = + ( f r )
- should be close to 0
R
- is the most probable value of
A

R
- The circulating load is defined as:
=
F 1

R df da
=
,
A df dr

F dr da
=
A dr df

c) CL calculation
Using the least square line calculations,
with = 0
d) Quick CL calculation
With one set of results of sieving:
R f a
=
F ar

2.2 Tromp Curve


a) Tromp Curve
On the Gausso-logarithmic paper, let's plot the probability for a given particle of a certain size entering the
dr( x )* R
with:
separator to go to the rejects =
da( x )* A
n

( f i ai )( f i ri )

R i =0
=
n
A

i =0

( f i ri ) 2

The Tromp curve can be divided into two straight lines:


The right one (higher sieves) has a slope which is representative of the separator efficiency (a perfect one
would be vertical).
The left one (smaller sieves) is flat.

4.12
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM
b) Imperfection
I = d75 d 25
2 * d 50

SECTION 4 MILLING

where:
- d25 is the size of the particle which has 25 % chance of going to rejects
- d50 is the size of the particle which has 50 % chance of going to rejects
- d75 is the size of the particle which has 75 % chance of going to rejects
I=

0.4 -0.5 for a high efficiency separator


0.6 - 0.7 for a Sturtevant
0.45 - 0.6 for a Raymond separator

Imperfection vs Circulation Load


0.44

Imperfection

0.42
0.40
0.38
0.36
0

100

200
300
Circ. load (%)

400

c) Acuity Limit
AL is the abscissa of the intersection of the two Tromp curve lines.
Its the size at which selection is initiated

Rule of thumb
Cement mill = Acuity limit: 20-30 m, Raw mill = Acuity limit: 30-60 m
d) Bypass

Definition:
By-pass is the ordinate of the intersection of the two Tromp curve lines.
The bypass is the lowest percentage of feed that will go to the separator rejects.

60
40
20
0

50

100
150
200
250
Feedrate to Separator (t/h)

300

80
70
60
Bypass (%)

Bypass (%)

80

00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000

100

Bypass vs. feed rate OSepa/Sturtevant

50
40
30
20
10
0

0000000000000000000000000000000000000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000

Bypass vs. feed rate Sturtevant


The following graph shows the Bypass of an
18Sturtevant versus its feed rate.

Sturtevant

O-Sepa

1.0
1.5
2.0
2.5
3.0
Qf/Qa (kg feed/m3 separator sweep)

3.5

4.13
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

Bypass (%)

20
10

Bypass = 1 + e
- f1: coefficient for the separator

2
3
Qf/Qa (kg/m3)

RRnumber vs. Feed to Air Ratio


1.20
Rosin-Ramler Number (n)

e) Rosin Rammler Number


The steeper the size distribution (RR# high) the
more efficient the grinding and separating
process.
Cement Mill RR#
Open Circuit 0.75 0.85
Raymond Sturtevant 0.85 1.00
2nd general HW 1.00 1.20
High Efficiency Separator 1.10 1.40

1.15
1.10
1.05
1.00

0 0 0 0 0
0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
0 0 0 0 0

Qf
Qa

f1

30

0 00 00
00 0000 0000
00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000000000000000000000000000000000000000000
00 0000 0000

40

QF/Qa vs. bypass


If Qf is the separator feed rate (kg/h) and Qa the
separator ventilation (m3/h),
Qf/Qa is an important ratio for the separator
efficiency.

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Qf/Qa (kg/m3)

Raw mix RR# are usually lower

Separation Performance
Rate of recuperation in the fines of particles smaller than a given dimension.
F f
r= *
A a

2.3 Indicators for Cement Milling and Typical Values


IMLt
IMLm:
100/c:
100/c C1:
100/c C2:
NRR fines:
% recovery, 45 m:
Acuity:
Imperfection:

Bypass:
Circulating load:
HES Qf/Qa:
% Passing 45 m:

> 37 (laser / sieve)


in the 17 19 range
63 91 % range
60 85 % range
80 110 % range
1.1 1.4 for HES
0.85 1.0 for 1st generation separators (Sturtevant, Raymond)
1.2 for second generation separators
55% for HES
20 30 m
0.4 0.5 for HES
0.45 0.6 for Raymond separators
0.6 0.7 for Sturtevant separators
5 10% range for HES
150 200 % with HES
1.5 2.0 range
93% minimum (45 m = 325 mesh)

4.14
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

2.4 Sturtevant/O'Sepa (Bath #B mill)


Data before (22' Sturtevant)-after (O'Sepa)
T10
Before
Prod(T/H)
81
kWh/t
total
42.6
Mill+elev+sep
42.1
Mill
38.3
Blaine
380
45
89
Strength
3d (MPa)
24.3
7d (MPa)
28.4
28d (MPa)
34.5
Water dem
(% for NC)
24.8
Flow@
48%W/C
113
Gypsum
dehydrant(%)
65%
Setting time
VIS(min)
120
VFS(min)
218
Blaine vs
325
40.6
Circuit characteristics
Mill
Diameter:
#1 comp length:
#2 comp length:
Power Connect:
Lining #1 comp:
Lining #2 comp:
Partition slots:
Discharge slots:

After
93.3
40.3
36
34.9
358
91.6
24.8
28.9
34.4
25.8
107
25
120
225
39.0

T1
Before
71
48.6
48.0
43.7
380
90
25.2
29.7
35.2
25.3
121
65
117
218

After
79.4
47.1
42.4
41.0
361
94.7
24.7
29.2
35.2
26.4
113
25
124
231

Separator
O'Sepa N-2000
Airflow design:
Pressure drop:
Rotor speed:
Feed capacity:
Power:

3.96 m
5.04 m eff
9.72 m eff
3581 kW
Lifting
Classifying
8mm-1
8mm+1

2000 m3/mi
8"
100-230 rpm
420 tph
200 HP AC var. freq

Ball charge (B mill)


comp

Volume load
comp

Volume load
Power kw

3"1/2
3"
2"1/2
2"1/2
2"
1"1/2
1"1/4
1"
3/4"

Before
20%
60%
20%
33%
7%
4%
8%
9%
12%
60%
34%
3100

After
20%
60%
20%
33%
3%
6%
9%
9%
22%
51%
36%
3260

Mill sweep
Separator
Primary
Aux
Secondary
Tertiary
TOTAL
Outlet fan

Temp
C
78.3

Press
"WG
-11

Flow
m3/h
34101

Flow
Nm/h
25781

15

66487

15
15

0
0

24079
13814

61

143381

63024
1301
22825
13095
100245
117156

4.15
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM
Separator sp =
Mill feed =
Mill #1 sound =
Mill #2 sound =
Tromp curve (A mill)
R/A) mean
CL
Acuity limit

174 RPM
79.2 t/h
78.2
62.5

Mill out temp =


Elevator =
Mill =
Type 10

= 0.656
= 191%
= 12 m

SECTION 4 MILLING

84 C
34 kW
3101 kW

Bypass
Imperfection
Rosin-Rammler(fine)

=8%
= 0.39
= 1.15

2.5 Recommended Steps for Sizing a HES


a)

Estimate new production rate


Assume a specific power consumption for an optimized ball mill with HES (37 kWh/t TI @3700 Blaine)
Back calculate the production rate based on the available motor power
Use the standard Blaine adjustment formula (Prod1/Prod2 = [Blaine2/Blaine1]^1.4).

b) Fix the Qf/Qa target with the circulating load


Decreasing Qf/Qa (kg feed per Am3 of air) increases the separator efficiency but also increases the capital and
the operating costs.
The CTS process target is Qf/Qa = 1.6 @ CL (R/F) = 150%; Qa @ 65C; 3800 Blaine.
The CTS upper limit is Qf/Qa = 2.0. Lower than this only marginally improve the separator Imperfection and
Bypass of 2-5% and therefore has no effect on production or product quality
Other suppliers rules of thumb:
Fuller typically sizes at a Qf/Qa = 2.0-2.5 @ CL = 180%.
Christian Pfeiffer
- Finish mill:
Qf/Qa = 1.8 @ CL = 200%; Qa @ 90C; 3500 Blaine; Qp/Qa (kg fines per m3 of air)
3000 Blaine
0.75-0.80 kg/Am3
4000
0.55-0.60
5000
< 0.50
- Raw mix:
Qf/Qa = 2.2 @ CL = 200%; Qa @ 90C; 12-14%R 90 m., Qp/Qa: 0.55 kg/m3.
- Slag: Christian Pfeiffer sizes slightly larger for slag circuits due to lower density and higher CL, as does
Polysius.
c)

Fan and Bag House Sizing


Use the new production rate (T/h), Qf/Qa (kg/ Am3) and circulating load (%) to specify the air flow.
Most separators can operate at +/- 20% of nominal.
Only a margin of 5 - 8% above the separator airflow is recommended for the BH.
Margin of 5-10% is recommended on top of the BH for the fan.
Correctly specifying the static pressure:
- Pressure drop can be estimated by the dP of the separator (8-12"), D/C (4-8"), ducting. (3-6") and if
present, silencer (1-2").
- The recommendation is 26 in WG with a minimum of 24 in WG.
Include in the circuit design, the possibility to recirculate from 0-80% of the separator airflow.

4.16
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

3. Heat and Water Balance


a) Water
Gypsum dehydration
Water generated = 0.156977*M*K*D
where:
- M = kg/h of SO3 source excluding clinker,
- K = % pure CaSO4(2H2O)
- D = % dehydration (from DSC)
Spray cooling with water
Water flow(kg/h) = Q*Cp*(Tf-Ti)/((100-Tw)+538.9*f)
where:
- Tf and Ti are the temperatures of material or gas before and after cooling (in C)
- Cp (kcal/kg) is the specific heat of material or gas and Q its flow rate (kg/h)
- Tw is the water temperature (C) and f is the % water evaporated.
Rule of thumb
Usually, water flow ranges from 0 to 3 % of the mill production
b) Heat balance
Mill heat generation
Kcal/h = kWh (power measured)*factor*860
Factor = 0.75 for vertical mill and 0.9 for ball mill
Furnace wall losses
About 5kcal/kg fuel
c) Mill heat balance sheet
Mill Product Rate (STPH)
Feed Temperature ___________ Percent H2O
%
Mill Diameter_ Ft. Mill Length____ Ft. Mill Motor HP
Mill Liner Thickness ________in Ball Charge
%
Separator Ventilator
Volume
ACFM
Temperature
F
Pressure
in.H2O
Mill Ventilator
Volume
ACFM
Temperature
F
Pressure
in.H2O
Baghouse Ventilator
Volume
ACFM
Temperature
F
Pressure
in.H2O
Auxiliary Ventilator
Volume
ACFM
Temperature
F
Pressure
in.H2O

Percent H2O in

Separator Rejects
Product
Mill Discharge
Percent Recirculating Load
%
Percent Leakage Into Separator (CFM)
Maximum Separator Inlet Temperature
Ambient Temperature
F
Percent Relative Humidity
%
Plant Altitude
FT
Fuel Type
Heat
Value
Separator Rejects Temperature F
Product Temperature
F
Baghouse Cloth Area
FT2
Number of Compartments
Number of Compartments Cleaning Together

4.17
Rev. 2002

%
%
%
%
F

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

4. Grinding Aid
Type of Products
Surface active agents tend to saturate the free valence and inhibit the pack-set. Typical surface-active agents
are:
- ligno-sulphonates
- polyoils
- amines
- organic acids
Polar compounds (water, ammonia) are known to have some action on such bonds through their polar moment.
However, their practical use as surface agents is limited by their other impacts on the cement properties.
Other agents, particularly coal dust, have been used in the past.
Commercial products available as grinding aids are essentially (60-800 g/t ck):
- Triethanolamine
- Polypropyleneglycols and polyethylene
HEA2, DDA& and other products cause a definite reduction of pack-set but do not prevent agglomeration or
lump-formation problems that are caused by:
- Alkalis ( K 2 SO4 )
- Moisture

The effect of grinding aid on milling process:


- Enhances the flowability and prevents agglomeration
- Prevents coating on liners and grinding media- Decreases the "Blaine: Passing 325" ratio
- Is lLowers effect on coarser product (below 320 m2/kg)
- Reduces contraction
Example Bath
HEA2 on the feeding belt, Range: 0.1-0.2 kg/t of kk
Specific gravity: 1.195 kg/l, % active agent: 70%
0.15
* 1.31 = 0.234 $/t kk
Price: 1.31 $/l, Cost:
0.7 * 1.195
At kWh/t cte, 200g/t of glycol: +80 SSB
Production increase and pack set decrease. Mixed with water (3/1) and injected in #1 comp
HEA2 (Grace)
0.025% weight per weight of kk
2.02$/kg
6% production increase
reduce cracks

HEA2/rm (Grace)
0.078% weight per weight of kk
1.48$/kg
for a higher setting time (20 min)

4.18
Rev. 2002

CEMENT PROCESS ENGINEERING


VADE-MECUM

SECTION 4 MILLING

5. Sieve
Sieve Screen
#400
#325
#270
#230
#200
#170
#140
#120
#100
#80
#70
#60
#50
#45
#40
#35
#30
#25
#20
#18
#16

Micron
37
44
53
63
74
88
105
125
149
177
210
250
297
354
420
500
595
707
1000

Iso alter
38
45
53
63
75
90
106
125
150
180
212
250
300
355
425
500
600
710
850
1000
1180

Screen
#14
#12
#10
#8
#7
#6
#5
#4
#3.5
1/4"
5/16"
3/8"
7/16"
1/2"
5/8"
3/4"
7/8"
1"
1"1/4
1"1/2
2"

Micron

2000

6350
8000
9510
11200
12700
16000
19000
22600
25400
32000
38100
50800

Iso alter
1400
1700
2000
2360
2800
3350
4000
4750
5600
6300
8000
9500
11200
12500
16000
19000
22400
25000
31500
38100
50000

4.19
Rev. 2002

You might also like