Analysis for predicting the Input Interactions of HBF Performance at -10 μm Particles Size for treating at Iron Ore Fines grade 24% to 29%.

Invention Journal of Research Technology in Engineering & Management (IJRTEM) ISSN: 2455-3689
www.ijrtem.com ǁ Volume 1 ǁ Issue 9 ǁ
| Volume 1 | Issue 9| www.ijrtem.com | 30 |
Analysis for predicting the Input Interactions of HBF Performance at
-10 µm Particles Size for treating at Iron Ore Fines grade 24% to 29%.
Roopa Navalli[1]
, Harshit B Kulkarni [2]
, Praveen kumar Hiremath [3]
, Sangamesh Desai[4]
1,2
(Faculty, Mechanical Engineering Department, KLS GIT Belgavi, Karnataka, India)
3
(Student M.Tech Production Management, KLS GIT Belgavi,)
4(DGM Maintenance OBP 2 JSW Toranagallu Ballari Karnataka, India)
ABSTRACT: Dewatering is an important process in any mineral industry. It is a process which removes the unwanted material from
the liquid solid suspension called slurry by using a filter element which separates the unwanted fluid material from the solids from the
feed. The paper attempts to establish the way towards analysis of Hyper Baric Filter (HBF) performance at -10μm particle size
treating iron ore fines (24% to 29%). Dewatering in HBF, requires reduction in moisture and material throughput rate in terms of per
hour so as to increase the performance of HBF. The present work carried out illustrates a method to predict the influence of process
input parameter such as vessel pressure, snap blow and filter disk rotation for reduction in moisture percentage level and material for
reduction moisture percentage level and material throughput rate for particle size in the range of 24% to 29%. Using Design of
Experiments (DOE) a linear regression model is developed to study the performance of HBF full factorial design method using
ANOVA to analyze the data. Validation of the results is performed by comparing the experimental values and predicted values for
Material through put rate in terms of cycles/hr and reduction in moisture percentage by weight and hot spots.
Keywords: Hyper Baric Filter, dewatering, design of experiments, size of particles, vessel pressure.
INTRODUCTION
The increasingly higher portion of fine particles causes several problems in the filtration of iron ore concentrates. One drawback is that
remarkable quantities of slurry are attached to the very fine particles because of their specific surface area. Furthermore, fine particles
also reduce the size of the capillaries in the filter cake. In fact, the Hyper baric filter is considered to be the efficient dewatering
equipment for the treating coal. The Hyperbaric Filter is a closed system (inside the pressure vessel) as shown in figure 1.1[1]
n
Snap Blow
Cake Formation Zone
Cake Dewatering
Zone
n
Snap Blow
Cake Formation Zone
Cake Dewatering
Zone
Fig. 1.1: Closed System Fig. 1. 2: Components of HBF Fig. 1.3: Filtration Zones of a Disc filter
(Courtesy: JSW STEELS, Toranagallu, Manual)
Fig.1.2 shows the main components and the function of a hyperbaric filter. The suspension is pumped into the trough (6) continuously
during filtration. The agitators (7) to maintain the suspension are either single devices between the filter discs or a paddle agitator is
installed at the bottom. After filtration, the cake discharged is conveyed to the double gate discharging system (8). The control disc
divides the filtration or dewatering process into three functions, as shown in Fig.1.3 The Zones are: Cake formation (The disc sectors
are submerged in the suspension), Dewatering (The air pressure applied removes the liquid from the filter cake) and the Snap-blow to
assisted cake discharge [1].
LITERATURE REVIEW
Many researchers have worked in this area and have investigated the most influencing factors as follows. Dong-Jin Sung et al. [2]
investigated the study of pressure filtration for dewatering of coal fines. The impact of five most influencing factors such as pressure
applied filtration period, lump thickness, concentration of solids in feed and slurry, pH on cake moisture, reduction of moisture
percentage on discharged lumps and air consumption were researched and studied. They found that the filtration duration, pressure
applied and lump thickness had major influence on consumption of air as well s reduction of moisture percentage in filtered pumps.
Manoj K Mohanty et al. [3] highlighted the dewatering of coal. The objective of the research was to obtain an improved understanding
the influence of feed solid content and volumetric flow that affected the clean coal recovery. M K Mohanty et al. also explored the
correctness or fitness of recently developed dewatering technique that is (SBF) steel belt filter, for filtration of coal fines. They used
Response Surface Methodology with factorial design method for optimization of filtration process. Kenneth J. Miller and Wu-Wey
Wen [4] employed a 6 inch continuous screen bowl centrifuge in pilot plant study designed to evaluate the effect of reagent addition,
coal particle size distribution, slurry feed rate, and slurry feed solids concentration on dewatering of finely ground Pittsburgh bed coal.

Invention Journal of Research Technology in Engineering & Management
| Volume 1| Issue 9| www.ijrtem.com | 31
|
T. Sivakumar, et al. [5] stressed on computing the reduction in moisture % of the discharged cake and to improve the performance of
the compressor. B K Parekh, et al. [6] stated that dewatering or filtration of coal feed is the major part in operation of coal washing.
The research concentrates on the froth flotation product obtained from a coal preparation plant processing Pittsburgh no seam coal,
through reduction in moisture percentage by weight.
EXPERIMENTAL INVESTIGATION
For the present work in conjunction with the previously published work by Praveen et.al [1], the experiments were carried on iron ore
fines with percentage variation of 24 % to 29 % by keeping particle size constant of -10 microns. The effect of process input
parameters such as Vessel pressure was analyzed. The effect of Snap blow and Filter disk rotation on iron ore fines for analyzing cake
moisture percentage reduction and material output rate during the filtration on Hyper baric filter is observed. The literature review
provides details of work carried out by various authors in the field of modeling, simulation and parametric optimization on cake
moisture reduction and material output rate of the product in Hyper baric filter using different process parameters.
3.1 Experimental procedure:
The Hyper baric filter (HBF) is checked for performing the filtration process for the set of standard runs as per the design matrix.
a)The desired process input parameters like vessel pressure, snap blow, and filter disk rotation are set through the controller of the
equipment.
b)The filtration process takes place and continues in the hyper baric filter. The output in the form of cake gets discharged from the
Vessel.
c)At a frequency of one hour the required responses cycles/hr can be recorded directly from the display unit of the controller.
i. Recording the material output rate in the form of cycles/hr.
ii. One Cycle = 5 tonnes
d) At a frequency of one hour a sample is collected from the discharged cake to measure the moisture percentage in the cake using
moisture analyzer.
e) At a frequency of one hour a sample is collected from the same discharged cake to measure the particles size in the cake using
particle analyzer.
SELECTION OF FACTORS, LEVELS AND RANGE
a. Input Parameters:
The selected major input process parameters for this work are Vessel pressure (VP), Snap blow (SB) and Filter Disk Rotation (FDR).
Each input process parameters has been assigned three levels that is low, medium and high as shown in below table (1). The table
shows three factors assigned with three levels. A total of 27 experiments were conducted using 33
full factorial design method. Three
major parameters were used to check the material throughput rate, so that the moisture percentage is maintained. The input parameters
are varied with low, medium and high levels on the hyperbaric filter. Hence the design matrix used is 33
full factorial design matrix.
Table 1: Design Matrix selected
Code Level
Process parameters
Vessel pressure
(Bar)
Snap
blow(Bar)
Filter disk rotation
(Rpm)
A B C
1 Low 2.5 0.5 0.8
2 Medium 2.8 0.6 1
3 High 3 0.7 1.2
b. Output Variables:
The reduction of Cake moisture percentage is the most important output parameter and it is an index of product quality. Material
throughput rate is a measure related to filtration method that determines production. Higher production rate implies higher productivity
hence the responses selected for this work are material output rate in terms of cycles per hour and reduction of moisture percentage by
weight.

|
EXPERIMENTAL READINGS
The experiment values obtained below in the Table 2 and Table 3 are noted by repeating the above procedure for each set
of standard runs.
Table 2 Experimental values of reduced moisture % by weight and material through put in term of cycles for -10 microns Particle size
(24 to 29%)
Experimental readings for -10 um size (24-29% )
SL NO VP SB FDR CYCLES MOISTURE
1 2.5 0.5 0.8 18 10.44
2 2.5 0.5 1 21 10.52
3 2.5 0.5 1.2 27 10.61
4 2.5 0.6 0.8 18 10.37
5 2.5 0.6 1 23 10.42
6 2.5 0.6 1.2 30 10.56
7 2.5 0.7 0.8 23 10.08
8 2.5 0.7 1 26 10.2
9 2.5 0.7 1.2 33 10.34
10 2.8 0.5 0.8 24 9.89
11 2.8 0.5 1 27 9.94
12 2.8 0.5 1.2 32 10.08
13 2.8 0.6 0.8 25 9.77
14 2.8 0.6 1 32 9.86
15 2.8 0.6 1.2 34 9.96
16 2.8 0.7 0.8 27 9.62
17 2.8 0.7 1 33 9.78
18 2.8 0.7 1.2 36 9.82
19 3 0.5 0.8 28 9.65
20 3 0.5 1 31 9.74
21 3 0.5 1.2 35 9.82
22 3 0.6 0.8 33 9.36
23 3 0.6 1 35 9.64
24 3 0.6 1.2 38 9.75
25 3 0.7 0.8 35 9.31
26 3 0.7 1 38 9.39
27 3 0.7 1.2 40 9.43

|
RESULTS AND DISCUSSION
The analysis of variance (ANOVA) table for 95% confidence level using Minitab-14 statistical analysis software and coefficients are
used to develop a linear regression forecast model for the responses, Cycles/hr and Moisture percentage. Effect of process input
parameters Vessel pressure, Snap blow, and Filter disk rotation on Cycles/hr and Moisture percentage are discussed.
Table 3 ANOVA table, graph and prediction model for CYCLES/hr
Source DF Seq SS Adj SS Adj MS F P
VP 1 489.98 18.35 18.35 14.40 0.001
SB 1 128.00 0.02 0.02 0.01 0.904
FDR 1 304.22 21.02 21.02 16.50 0.001
VP*SB 1 0.47 0.47 0.47 0.37 0.549
SB*FDR 1 0.00 0.00 0.00 0.00 1.00
VP*FDR 1 15.51 15.51 15.51 12.17 0.002
ERROR 20 2548 25.48 1.27 - -
TOTAL 26 963.63 - - - -
Figure 6.1 Main Effect Plots and Interaction plots for -10 microns particles size (24 to 29 %) for cycles/hr
The main effect plot reveals that during operation on hyper baric filter, the material output rate in terms of Cycles/hr are affected by all
the process input parameters that is Vessel pressure, Snap blow and Filter disk rotation. The Cycles/hr is increased by increasing any of
the process input parameters. From ANOVA table it is found that for Cycles/hr contribution of Vessel pressure is 51.77% and
contribution of Filter disk rotation is 31.57% and Snap blow pressure contributes by 13.28%.
Among the interactional effects the major contribution for achieving the material output rate in terms of cycles/hr are Vessel pressure
& Filter disk rotation. Linear Regression Model for Cycles/hr material through put rate for -10 microns particles size (24 to 29 %) can
written as follows by using coefficients obtained from ANOVA table.
General Linear Model: Moisture % versus Vessel pressure, Snap blow, Filter disk rotation for -10 microns particles size (24 to
29 %)
Cycles/hr = (-113.60) + (38.58*VP) + (4.82*SB) + (83.05*FDR) + (7.89*VP*SB) - (0*SB*FDR) - (22.588*VP*FDR)
Table 4 ANOVA table, graph and prediction model for moisture % by weight
Source DF Seq SS Adj SS Adj MS F P
VP 1 3.16333 0.03069 0.03069 8.63 0.008
SB 1 0.41102 0.00071 0.00071 0.20 0.660
FDR 1 0.19636 0.00008 0.00008 0.02 0.879
VP*SB 1 0.00062 0.00062 0.00062 0.17 0.681
SB*FDR 1 0.00021 0.00021 0.00021 0.06 0.811
VP*FDR 1 0.00021 0.00021 0.00021 0.06 0.810
ERROR 20 0.07112 0.07112 0.00356

|
TOTAL 26 3.84287
Figure 6.2 Main Effect Plots and Interaction plots for -10 microns particles size (30 to 35 %) for moisture %
The main effect plot reveals that during operation on Hyper Baric Filter, reduction in moisture % is affected by Vessel pressure. The
major contribution for achieving reduction in moisture percentage is Vessel pressure. From ANOVA table it is calculated that for
reduction in moisture percentage contribution of Vessel pressure is 82.31% and contribution of Snap blow is 10.69% and Filter disk
rotation contributes 5.10%.
Among the interactional effects the Major contribution for achieving reduction in moisture% are Vessel Pressure & Snap Blow.
Linear Regression Model for Moisture percentage for -10 microns particles size (24 to 29 %) can written as follows by using
coefficients obtained from ANOVA table.
Moisture% = 14.814 - (1.5781*VP) - (0.931*SB) + (0.167*FDR) - (0.2859*VP*SB) + (0.2083*SB*FDR) + (0.0833*VP*FDR)
Confirmation of Experiments
On the same HBF setup Validation was made and found to be within the required level of confidence. The verification of the
experiments for cycles/hr and moisture percentage by weight for -10 microns particle size (24 to 29%) was conducted from the in
between data values of the design matrix. The below table provides the percentage error calculated from experimental and predicted
results.
Table no 5 Verification experiments for cycles/hr and moisture % by weight for -10 microns particle size
24 % to 29 %
For -10 microns particle size 24 to 29 %
VP SB FDR moisture% by weight
Percentage error
Bar BAR Rpm Experimental Predicted
2.6 0.6 0.9 10.54 10.16 3.74 %
2.9 0.7 1.1 10.08 9.64 4.56 %
Influence of -10µm particle size with respect to material through put rate
Figure 6.3 Graph of -10um particle size Vs Avg Tonnes per hour
From the graph 6.3 it is observed that increase in percentage of -10µm size particles material through put or output tonnes per hour
TPH is decreasing. And it also reveals that decrease in percentage of -10µm size particles material through output rate increases that is
TPH increases. It means that cake thickness increases with decrease in -10µm particles.

|
Influence of -10µm particle size with respect to Moisture% by weight.
Figure 6.4 Graph of -10um particle size Vs moisture %
The graph 6.4 indicates that the higher the fines content, that is higher the percentage of -10µm size particles the lower the throughput
and the higher the residual moisture by weight. It is also observed from the above graph the higher the content of coarse particles and
the lower the fines, the higher the throughput and the reduction in moisture % of the cake discharged. Increase in -10um size particles
there will be increase in moisture. The fine particles (Near Size Particles) will clog the filter cloth Due to that moisture will increase
in the discharged cake.
Influence of Vessel Pressure with respect to Moisture%
Figure 6.5 Graph of vessel pressure Vs moisture %
The graph 6.5 reveals that with increase in Vessel pressure there will be reduction in moisture %. And when vessel pressure is
decreasing moisture % will increase of the discharged cake.
Influence of Filter disk rotation with respect to Cycles/Hr
Figure 6.6 Graph of Filter disk rotation VS cycles/hr
From the above graph 6.6 it is observed that increase in Rpm of Filter disk there will be increase in throughput (Cycles/hr) but
moisture % by weight will also increase. For Decrease in Rpm of Filter disk rotation o there will be decrease in throughput (Cycles/hr)
but moisture percentage decrease of the discharged cake.
CONCLUSION
The present work was carried out to predict the influence of process input parameters on the reduction of moisture percentage and
material throughput rate for different percentage of -10µm particle size that is for grade 24% to 29 %. The following are the
conclusions drawn from the study.
 The material through put rate (TPH) is increasing when the percentage of -10µm size particles are less than 30 % and also with
increase in vessel pressure and Filter disk rotation.
 The material through put rate (TPH) is decreasing if the percentage of -10µm size particles are more than 30 % and also with
increase in vessel pressure and Filter disk rotation.
 Higher the percentage of -10µm size particles i.e. more than 30% reduction of moisture percentage is less because of the near size
particles block the aperture of the filter cloth.
 Lower the percentage of -10µm size particles i.e. less than 30% reduction of moisture percent is high.
 Moisture percentage by weight is decreasing with increase in Vessel pressure.

|
 For the Response Throughput (cycles) TPH it has been observed that all three factors have positive contribution with Vessel
pressure having Maximum significance.
 With decrease in vessel pressure there will be decrease in material through put rate.
 Increase in vessel pressure there will be decrease in moisture percentage by weight.
 Decrease in vessel pressure there will be increase in moisture percentage by weight.
.
REFERENCES
[1] Praveen Kumar Hiremath , Roopa Navalli, Shivakumar.S, Sangamesh Desai Study of Process Input Interactions of HBF Performance at
Minus-10μm Particles Size Treating at Iron Ore Fines, Advanced Engineering and Applied Sciences: An International Journal, 2016.
[2] Sung, Dong-Jin, and Bhupendra K. Parekh. "Statistical evaluation of hyperbaric filtration for fine coal dewatering." Korean Journal of
Chemical Engineering 13, no. 3 (1996): 304-309.
[3] Zhang, Baojie, Paul Brodzik, and Manoj K. Mohanty. "Improving fine coal cleaning performance by high-efficiency particle size
classification." International Journal of Coal Preparation and Utilization 34, no. 3-4 (2014): 145-156.
[4] Miller, Kenneth J., and Wu-Wey Wen. Effect of operating parameters and reagent addition on fine coal dewatering in a screen bowl
centrifuge. No. DOE/PETC/TR-85/1. USDOE Pittsburgh Energy Technology Center, PA, 1984.
[5] Sivakumar, T., G. Vijayaraghavan, and A. Vimal Kumar. "ENHANCING THE PERFORMANCE OF ROTARY VACUUM DRUM
FILTER."
[6] Parekh, B. K., and A. E. Bland. "Fine Coal and Refuse Dewatering-Present State and Future consideration." In Flocculation and Dewatering,
Processing Engineering Foundation Conference, Scheiner and Moudgil (Eds.), pp. 383-398. 1989.

Analysis for predicting the Input Interactions of HBF Performance at -10 μm Particles Size for treating at Iron Ore Fines grade 24% to 29%.

More Related Content

What's hot (19)

Similar to Analysis for predicting the Input Interactions of HBF Performance at -10 μm Particles Size for treating at Iron Ore Fines grade 24% to 29%. (20)

More from journal ijrtem (20)

Recently uploaded (20)

Analysis for predicting the Input Interactions of HBF Performance at -10 μm Particles Size for treating at Iron Ore Fines grade 24% to 29%.