MASTER'S THESIS
New Future Ring Layout with Parallel Boreholes
Results from Small - Scale Blasting Tests
Youssef Hamoudeh 2013
Master of Science (120 credits) Civil Engineering
Luleå University of Technology
Department of Civil, Environmental and Natural Resources Engineering
New future ring layout with parallel boreholes Results from small - scale blasting tests
Master thesis
By
Youssef Hamoudeh
Division of Mining and Geotechnical Engineering
Department of Civil, Environmental and Natural Resources Engineering Luleå University of Technology
Preface
The work in this master thesis has been carried out at the division of Mining and Geotechnical Engineering at Luleå University of Technology, Sweden in conformity with the requirements for the degree of Master of Science in Civil Engineering with specialization in Mining and Geotechnical Engineering.
I would like to express my sincere appreciation to my supervisor Dr. Daniel Johansson for his encouragement and guidance throughout the course of my thesis.
My supervisor at LKAB Anders Nordqvist is deeply appreciated for his unlimited support and kind caring. I am grateful to PhD student Nikolaos Petropoulos for his interest in this thesis and for his valuable support.
I am thankful to Jonny Olofsson and Leif Keskitalo Technical staff at LKAB for providing technical support and guidance during the blasting tests at LKAB Kiruna.
Sincere thanks goes to the LKAB mining company for the use of sieving lab in the sorting plant and for the magnetic samples and also to LKAB Kimit AB for allowing me to use the testing site and the staff at University’s test lab for their assistance.
Thanks also to all my friends since I started in LTU.
Finally, I would like to dedicate all my life achievements, including this master’s study, to my family and for most thanks to Allah for everything.
Youssef Hamoudeh Luleå/ Sweden April-2013
i
Abstract
The blast holes in a sublevel caving ring are today drilled in a fan pattern. This result in an uneven distribution of explosives in the ring and consequently a variation in fragmentation could be expected. The main problem with the current ring design is that the hole spacing varies too much from the region near to the drift to the region at the top.
A possibility to avoid this problem is to distribute the explosives more evenly through the whole ring and this can be done by using a parallel hole design. This can be achieved by using a special drilling technique which changes the drill design from a fan pattern to a parallel pattern (Hustrulid & Kvapil, 2008). Model scale tests have been done to evaluate the fragmentation both for a standard and a curved ring layout. The blasting tests have been carried out at LKAB Kimit AB blasting testing site. The models were made of magnetic mortar and the rings been scaled down by using a scale factor of 1:54 giving a model height of 1 m. In order to simulate confined conditions as in real sublevel caving crushed granite 0-16 mm was used.
Two standard rings have been blasted under free face condition, three standard rings have been blasted under confined condition and one ring has been blasted for the future ring layout under confined condition. The specific charge was varied from 1.63 kg/m3 to 4.6 kg/m3 by using decoupled PETN with different cord strengths. The fragmentation has been analyzed by dry sieving of all blasted material and the results have been compared with previous research work and discussed within this thesis work.
Keywords: sublevel caving, fragmentation, blasting, small scale tests
ii
Abbreviations
A Rock mass factor
B Burden; mm
b Undulation parameter cp P – wave velocity; m/s
da Thickness of the confined material; mm E Explosive energy; MJ/kg
LKAB Luossavaara Kirunavaara Aktie Bolag
P Porosity; %
PETN Pentaerythritol tetranitrate q Specific charge; kg/m3 SLC Sub-level caving
SweBrec Swedish Blasting Research Centre UCS Strength of the confined material; MPa VOD Velocity of detonation; m/s
x50 Average fragment size; mm xmax Maximum fragment size; mm Ρ Density of the magnetite; kg/m3 ρa Degree of packing; %
σΒΤ Magnetite Brazilian tensile strength; MPa P80 80% passing sieve size; %
iii
Important terminologies in mining
Burden (B): The distance from a blasthole to the nearest free face.
Delay time: Time interval between the initiation and detonation of a detonator.
Detonator: A device containing a small detonating charge that is used for detonating an explosive, including, but not limited to, blasting caps, exploders, electric detonators, and delay electric blasting caps.
Free face: This is an exposed rock surface towards which the explosive charge can break out.
Explosive: Any rapidly combustive or expanding substance. The energy released during this rapid combustion or expansion can be used to break rock.
Ore dilution: The contamination of ore with waste rock.
Specific charge: This is the ratio between the mass of explosives required to break a given quantity of rock and is normally expressed in kg/m3 or kg/ton.
Spacing (S): This is the distance between adjacent blast holes and measured perpendicular to the burden.
Waste: The barren rock in a mine which is too low grade to be of economic value.
Velocity of detonation, (VOD), (m/s): the velocity at which the detonation waves travels through a column or mass of explosive. The detonation velocity of an explosive depends on the type of explosive, particle size, density, diameter, packing, confinement and initiation.
1 Table of Contents
Preface
Abstract ... i
Abbreviations ... ii
Important terminologies in mining ... iii
1. Literature review ... 3
1.1 Sublevel caving mining method ... 3
1.2 History and development of SLC ... 3
1.3 Description of SLC sequence at Kiruna mine ... 5
1.4 Advantages & Disadvantages of Sublevel Caving Mining ... 6
1.5 Confined condition ... 7
1.6 Fragmentation evaluation ... 9
1.6.1 Kuz – Ram model ... 9
1.6.2 The KCO model ... 9
1.6.3 The JKMRC model ... 11
1.7 Small-scale experimental work by previous researchers ... 13
2. Objectives ... 15
3. Methodology ... 17
3.1 Selection of the ring for the tests ... 17
3.2 Experimental section ... 17
3.2.1 Preparation of the magnetic moulds ... 18
3.2.2 Confined condition ... 22
3.2.3 The blasting tests procedure ... 23
3.2.4 Firing procedure ... 25
3.2.5 Evaluation of fragmentation ... 28
4. Results ... 31
4.1 Size distribution curves for the standard ring (T5) versus curved layout TC (Magnetic Material) ... 33
4.2 Problems faced in the tests set-up ... 35
4.3 Comparing the curved ring layout TC with block tests ... 36
4.4 Comparing the standard tests and TC for debris ... 40
5. Discussion and conclusions... 43
6. References ... 45
7. Appendix-Sieving data for the tests ... 49
2
3
1. Literature review
1.1 Sublevel caving mining method
Sublevel caving mining method (SLC) is underground mass mining method which it is often used to extract large steeply dipping ore bodies (Hustrulid, 2000) and is shown in Figure 1. SLC involves a confined blasting environment (Cullum, 1974). The use of the SLC is increasing worldwide due to its high productivity, simplicity and profitability.
Though, the disadvantages are poor recovery and high dilution (Hustrulid, 2000). LKAB is a mining company using SLC for mining iron ore in two underground mines located in Kiruna (single large orebody) and Malmberget (many small orebodies) in the northern part of Sweden.
Figure 1. Sublevel caving (after Hamrin et. al, 2001).
1.2 History and development of SLC
The scale of the SLC has been increased distinctly through the history of this method and it is prominent that LKAB being the leader in the development this method. The scale was very small as comparing with the scale used today. Initially the sublevel spacing was 9 m with drift size of 5 x 3.5 m, and a sublevel drift spacing of 10 m center-to-center.
4
Currently the sublevel spacing is about 29 m with drift size 7 x 5 m and the sublevel drift spacing of 25 m center-to-center (Hustrulid & Kvapil, 2008). Figure 2 shows the history of the SLC geometry in Kiruna mine and Table 1 shows the most important design parameters.
Figure 2. Geometry at the Kiruna mine at 1963, 1983 and 2003(Hustrulid & Kvapil, 2008).
Table 1. Summary of some important design parameters (after Hustrulid & Kvapil, 2008).
Parameter/year 1963 1983 2003
Drift width(m) 5 5 7
Drift height (m) 3.5 4 5
Sublevel height (m) 9 12 27
Sublevel drift spacing (m) 10 11 25
Blasthole diameter (mm) 45 57-76 115
Burden (m) 1.6 1.8 3
Holes/ring 9 9 10
Tons/ring (t) 660 1080 9300
Tons/meter of drift (t/m) 400 600 3100
5
1.3 Description of SLC sequence at Kiruna mine
Sublevel caving mining method used in the Kiruna mine involves the following stages throughout the extraction process of the ore (Hamrin et al., 2001; Hustrulid, 2000), see Figure 3.
1- Drilling
When the cross-cuts are ready, production holes (ϕ = 115 mm) are drilled in a fan hole blast pattern. The borehole length varies from about 25 m to 55 m for full size rings.
Every ring contain 10 holes with an inclination of 80°, the side angle of the side holes is about 73°. The average length of the cross-cuts is about 80 m and with a burden of 3 m.
This cross cut will take around 25 fan cut. The drilling technology used is a water powered hammer, which can drill long holes with fairly small deviations. This technology was developed LKAB Wassara AB.
2- Charging and blasting
The main production starts after completing the production drilling. The explosive used in the Kiruna mine is bulk emulsion developed by LKAB’s subsidiary LKAB Kimit AB.
The explosive is charged into the boreholes and between 20 to 30 tons of explosives detonates in production blasts every day in the Kiruna mine.
3- Loading
When the ring is blasted and toxic fumes ventilated, the blasted ore is loaded by using load haul dump machines (LHD:s). The load haul dump machines is both operator controlled and remotely controlled. The bucket capacity is 17 – 25 tons and after filling the bucket the ore is transported to the ore pass for further transportation at the main level.
6 4- Transporting and crushing
The crude ore is transported from the ore passes to the crushers by driverless trains. The trains are emptied into rock bins and later the ore is feeded into crushers. The material is crushed down to 10 cm and further transported to the skips which lift the ore to the surface of the mine.
Figure 3. Mining activates in Kiruna’s mine. (Source, www.lkab.com)
Only blasting activates will be discussed in details in this thesis.
1.4 Advantages & Disadvantages of Sublevel Caving Mining
The advantages and disadvantages of sublevel caving mining method can be summarized as follows:
7 Advantages
The SLC mining method has a number of advantages over other mining methods (Bull &
Page, 2000)
Low Cost: Back fil1 is not required and manual labor requirements are low.
High safety: Miners are not exposed to open stopes areas and most of the work is carried out in well supported areas.
Efficiency: The mining process is applied and adaptable to mechanization.
Improvements in mining machinery and automation can often be integrated to this method resulting in improved overall efficiency of the mine.
High production: High production is achieved due to the large number of draw points. The ore is often extracted from different sublevels simultaneously.
Disadvantages
High Dilution: The dilution is relatively high and can reach up to 40%. The main source of the dilution is the flow behavior of the waste and ore (Janelid & Kvapil, 1966)
Low Recoveries: Recovery can range from 75 to 90%. Low recovery makes the method uneconomical for high-grade deposits with high in situ value.
Mine Instability: Progression of mining to deeper levels may create instability in the pillars between production drifts. Consideration must be given to subsidence and ground water problems.
Gravity flow: The mechanics of gravity flow of blasted and caved material is not well understood (Brady & Brown, 2004, Wimmer, 2012).
1.5 Confined condition
The mining method is characterized by confined blasting conditions. This condition leads to considerably coarser fragmentation compared with unconfined condition since that the loosening materials for the previous blasting lies in front of the new face for new blasting (Cullum, 1974; Johansson,2011).
8
Different research found conflicting results regarding improving the fragmentation under confined blasting conditions and free face (unconfined) blasting conditions. The free face blasting is when the loosening materials from the previous blasting are removed from the new face.
Many researchers have done several experiments for the confined blasting condition for both small and half scale tests.
Cullum (1974) examined the impact of increased confinement for each blast rows retreated from a starting slot by doing small scale blasting tests. He found that as each row was blasted, the confinement will increase for the next ring and the resulting fragmentation became coarser (Cullum, 1974). Volchenko (1977) found that the fragmentation became finer under confined conditions when he did tests in small scale.
Olsson (1987) prepared half-scale blast tests under confined conditions with a bench design. The tests were made at LKAB Malmberget mine and the purpose of the tests was to obtain a varied void ratio between 10 and 100 % per row. Olsson’s results indicated that the fragmentation was improved under confined condition. Johansson (2007,2008 &
2011) prepared small- scale blast tests and the aim of the tests was to determine the fragmentation under confined conditions. The tests showed that the fragmentation became coarser under confined conditions (Johansson, 2007, 2008 & 2011).
Johansson (2011) made a series of small-scale tests on mortar blocks, both free face and confined. The main purpose of this set-up was to investigate the influence of delay times on fragmentation. The block size was 650/660x205x300mm (LxWxH). Burden and spacing was 110 mm and 70 mm respectively. The average fragment size x50 was plotted against delay times for different confining face conditions. The results showed that the confinement made the fragmentation considerably coarser compared with a free face shot. The properties of confining material have also significant influence on the fragmentation of the blasted material.
9 1.6 Fragmentation evaluation
The purpose of the rock fragmentation by blasting is that to create rock fragments that can be loaded, hauled, crushed in desired fractions with desired properties in quarrying and mines. Many ways have been established to evaluate the fragmentation of the blasted rock (Ouchterlony, 2005). Researchers have introduced many indicators that can lead to evaluate the fragmentation of the blasted rock. These indicators are x30, x80, maximum fragment size (xmax) and average (i.e median) fragment size (x50). These can be measured through sieving process and can be done for both wet and dry materials. Most of the sieving data come from small -scale tests and some come from full scale tests due to the extensive amount of the sieving materials. To investigate the fragmentation in full-scale is very difficult and extensive work has to be made due to the mining method itself (Johansson, 2011).
Numerous models have been presented to predict the fragmentation of the blasted rock.
These models can be divided in two approaches:
Empirical models which concludes finer fragmentation and mechanistic models which concludes the physics of the explosives and the transfer of the energy in the rock for definite blast layout. The empirical approach is more used and applicable for daily blast design. The mechanistic approach which drive the all range of blasting results. It tracks the detonation physics and the transfers process the energy in well-defined rock for specific layouts blast. It is hard to apply for daily blast design, since it difficult to calculate the sufficient data about the explosion and the rock. It is also limited in scale and it is less accurate since, it requires less or more degree of experimental work.
(Cunningham, 2005)
Here follows brief descriptions of the most commonly used fragmentation models 1.6.1 Kuz – Ram model
Kuz – Ram model is considered to be the most used fragmentation model. It is based on the Rosin–Rammler distributions (Cunningham, 2005). The model consists of four
10
equations; from which, one describes the fragmentation curve (Rosin–Ramler distribution), one gives the value for the average fragment size (x50) as a function of the blasting parameters, the third gives a value for the rock mass factor (A) and the last gives a value for the uniformity index (n). The equations which describe the Kuz – Ram model are as follow:
Fragmentation curve (Rossin–Ramler distribution)
( ) ( ⁄ ) ( ⁄ ) Eq (1) Where,
P(x)= Percentage of passing materials x50= Average ( Median) fragment size (m) x= Size of the materials (m)
n= Uniformity index
This model does not concern the maximum fragment size xmax (m).
Average fragment size (x50)
( ⁄ ) ⁄ Eq (2) Where,
x50 = Average (median) fragment size (mm) A = Rock mass factor
q = Specific charge (kg/m3)
Q = Total amount of explosive per blast hole (kg)
SANFO= weight strength of the explosive in % compared to ANFO
The uniformity index is defined as:
n = (2.2-0.014·B/Ø)· (1-SD/B)·√[(1+S/B)/2]·[(Lb-Lc)/Ltot+0.1]0.1·(Ltot /H) Eq (3) Where,
B = burden (m) S = spacing (m)
Ø = drill-hole diameter (m)
11 Lb = length of bottom charge (m)
Lc = length of column charge (m) Ltot = total charge length (m) H =bench height (m)
SD = std deviation of drilling accuracy (m)
The rock mass factor (A)
( ) Eq (4) Where,
RMD= Rock mass descriptions JF = Joint factor = JPS + JPA
JPS = Joint Plane Spacing = 10, if Sj (av.joint spacing) < 0.1 m 20, if Sj < 0.1 x0, oversized fragment 50, if Sj > 0.1 x0, oversized fragment JPA = Joint Plane Angle = 20, if joints dip out of face
30, if joints is perpendicular to face 40, if joints dip into face
RDI(Rock density influence) = (0.025∙ρ) – 50 (kg/m3) HF = Hardness factor = E/3 if E < 50 MPa
σ
c/3 if E > 50 MPaThe Kuz-Ram model is a tool to examine how different parameters influence the blast fragmentation. This model does not take into account the maximum fragment size (Cunningham, 2005).
1.6.2 The KCO model
Kuznetsov–Cunningham–Ouchterlony (KCO) model is an improved version of Kuz–Ram model (Ouchterlony, 2005a). These improvements were done by replacing the Rosin–
Rammler function used to describe the fragmentation curve in Kuz – Ram model with the Swebrec function®. The Swebrec function® involves three main parameters, the
12
maximum fragment size xmax, the median fragment size x50 and the undulation parameter b (Ouchterlony, 2005a).
The equations which describe the Swebrec function® are as follow:
Swebrec function ( )
( ( ( )
) )
Eq (5)
Average fragment size x50
( ) Eq (6) Where,
x50= Average fragment size(mm) A= Rock mass factor
q= Specific charge(kg/m3)
Q=Total amount of explosive per blast hole (kg) SANFO=Weight strength of explosives used The undulation parameter b
( ) Eq (7) The maximum fragment size xmax; can be estimated from geological data or from the b- equation.
1.6.3 The JKMRC model
The JKMRC model is based on two fragmentation models, the two component model (TCM) and the crushed zone model (CZM). These two models are based on assumption that fragmentation is caused separately by two different mechanisms i.e. a fine part and a coarser part. According to the model the finer material is generated mainly in the crushed zone while the coarser material is generated by tensile fracturing and pre-existing fractures in the rock mass as shown in the Figure 4.
13
Figure 4. Schematic diagram showing crushing zone, fracture zone and fragment formation zone.
1.7 Small-scale experimental work by previous researchers
Many researchers as Rustan (1970) and Cullum (1974) used small scale experiments to measure the fragmentation when firing into a compressible material.
Rustan (1970) made a series of small scale tests to investigate the impact of experimental and blast designs on blast fragmentation and swell. Rapid hardening cement was used as a binding medium for the models with cement contents of 3 percent, 5 percent, or 10 percent. Additionally, artificial joints were introduced into some of the models with the use of crushed microscopic glass. The models were scaled to 1:75 of the standard SLC rings used at that time in LKAB. Twelve holes in a fan hole shape were placed in the concrete models and detonating cords were used as explosive source, resulting in specific charge of 5.31 kg/m3. The results indicated that the model material, inter-hole initiation timing, and the pressure exerted by the compressible material had a significant impact upon blast fragmentation and swell. The finest fragmentation P80 was achieved for inter- hole delays of between 0.1 ms and 1.0 ms, while the maximum average horizontal swell factor (1.4) was achieved for a delay time of 0.1 ms. An inverted parabolic relationship was observed between the pressure exerted by the compressible material (cave material pressure) and blast fragmentation P80 with the coarsest P80 obtained for a pressure of 0.8 MPa. An increase of the pressure exerted by the compressible material also resulted in less horizontal swell of the blasted material.
14
Cullum (1974) performed small-scale tests to investigate the effect of increased confinement on fragmentation and swell of the blasted material. The concrete models were confined by an upper and lower buffer steel plate slice. The number of holes was three or four in each row and they were parallel. The diameter of the holes and the depth were 11 mm and 60 mm respectively. The spacing and the burden for each row was 60 mm and 55 mm respectively. Each hole was filled with 0.24 g of PETN, which gave a specific charge of 0.91 kg/m3. The results showed a coarser fragmentation when the level of confinement increased. The bulk density of the blasted material was dependent upon the initiation scheme, the model size and rigidity. The conclusion of these tests was that the blasted rows moved away from the starting slot, the resulting fragmentation distribution became coarser.
Johansson (2011) made a series of small-scale tests on mortar blocks, bothe free face and confined. The main purpose of this setup was to investigate influence of delay times on fragmentation. The block size was 650/660x205x300mm (LxWxH). Burden and spacing was 110 mm and 70 mm respectively. Conclusion drawn from these tests was that the confinement makes the fragmentation coarser and x50 increases nearly by 100 %.
15
2. Objectives
The main objective of the thesis was to compare two different ring layouts in terms of fragmentation as shown in Figure 5.
Following activities were carried out to reach the objective:
Literature study
Preparation of test samples (blocks).
Carry out small-scale blasting tests on standard rings and on the new future ring layout.
Sieving analysis and size distribution curves.
Analysis of the results and comparing the two different layouts.
Figure 5. Standard ring layout to the left and a new future ring layout to the right (Details of the curved ring not 100% correct) (Hustrulid & Kvapil, 2008).
16
17
3. Methodology
3.1 Selection of the ring for the tests
The selection of the model geometry was based on a situation where rings in one of the two neighboring drifts have been blasted but not both (see Figure 6). The scale factor 1:54 has been selected according to the longest borehole, so that it would be < 1 m in height. All the parameters have been scaled down according to the selected scale factor for both the standard and the new future ring layout.
Figure 6. Front view of sublevel caving rings in Kiruna mine at the level 907 m to 964 m (LKAB Company).
3.2 Experimental section
The purpose of the tests was to measure and compare the fragmentation between the standard ring with fan holes and the future ring with parallel holes. In Figure 7 the schematic procedure in all different phases of the experimental work can be seen
18
Figure 7. Different phases of experimental work.
3.2.1 Preparation of the magnetic moulds
The magnetic mortar for conducting these tests was moulded according to the composition of materials shown in Table 2. Glenium 51 (plasticizer) is an innovative admixture based on modified polycarboxylic ether (PCE) polymers and it has been used to increase the durability of the magnetic mould during the manufacturing. Tributylphosphate is an effective solvent and is used in blending of material which is difficult to dissolve. It is used as antifoaming agent for concrete.
19
Table 2. Recipe of magnetic mortar.
Ingredient %
Portland cement 25.6
Water 12.6
Glenium 51 (plasticizer) 0.25 Tributylphosfate (deformer) 0.13
Magnetite powder 29.6
Quartz Sand 31.7
Firstly all the drawings were designed by AutoCAD (Figure 8 and Figure 9). The cross- sectional area of the mould was 0.242 m2 and the burden 0.055 m. The moulds were manufactured by using plywood and fixed properly with the help of hinges and screws.
Plastic sticks of ϕ10 mm were inserted inside the mould which acted as the boreholes, see Figure 10. To get the curved boreholes as seen in Figure 9, the plastic sticks were heated and bended to the desired shape. They were later placed and fixed into the specific moulds. The plastic sticks were placed carefully in the moulds. Each mould had four rows as seen in Figure 8.
Figure 8. Cross-section for the testing mould.
20
Figure 9. Curved ring layout with parallel boreholes and fan ring layout
Figure 10. Manufacturing process of moulds before casting.
21
The magnetic mortar was manufactured in the concrete lab at LTU. After fixing the model mould the magnetic mortar has been poured into the mould. The plastic sticks were carefully pulled out after 8 hours of curing time, see Figure 11. The moulds were left in the lab for curing under constant temperature (20 degree Celsius) for 28 days.
Figure 11. Filled mould with magnetic mortar and the final shape of the model.
The mechanical properties for the magnetic mortar had been measured earlier by using two methods, uniaxial compression test and the Brazilian test see Table 3 (Johansson, 2008).
Table 3. Mechanical properties of the magnetic mortar (after Johansson, 2008).
Mechanical properties Value
Uniaxial compressive strength 50.7 ± 4.8 MPa
Young’s Modulus 21.9 ± GPa
Tensile strength 5.2 ± 0.34 MPa
Density 2 459.5 ±25 kg/m3
P-wave 3808 ±73 m/s
22 3.2.2 Confined condition
Crushed granite (0-16 mm) was used as confining material and placed in front of each burden. The mean size of the crushed granite (x50) was 8 mm and the properties of it has been determined by earlier research (Johansson, 2008), see Table 4.
Table 4. Crushed granite properties (after Johansson, 2008).
Confined Material
Porosity (%)
Cp (m/s)
Average density (kg/m3)
Swebrec distribution parameters.
Crushed
granite 36 1168 1696
x50=8mm, xmax= 16 mm and b=2.2
Side concrete triangles had been prepared; the aim of it was to fulfil the requirements of the confined condition of SLC in a real case and to avoid the breakage of the magnetic mould after each blasting test. Two side concrete triangles were placed carefully on both sides of the magnetic mould as shown in Figure 12. The composition of the materials which has been used for manufacturing the triangles is shown in Table 5.
Figure 12. Curved ring layout showing the position of concrete triangles.
23
Table 5. Compositions for side concrete triangles.
Ingredient %
Portland cement 18.82
Water 7.53
Glenium 51 (plasticizer) ~0.1
Macadam (8 – 16)mm 29.7
Filler 2.92
Quartz Sand 41
3.2.3 The blasting tests procedure
A steel box was designed with the dimension (1m x 1m x 1m) for the blasting tests. The steel box was equipped with a movable steel plate to maintain a fairly constant thickness of the confined material in front of the ring. The ring shaped specimen was placed in the steel box and for making the blasting process fully confined concrete side triangles have been placed carefully at both sides. Sand bags were placed and packed at the back side of the magnetic mortar and also on top of the triangles in order to obtain confined conditions. It was also assumed that the sand bags could act as a wave trap during the blasting. This to avoid the back breakage after each shot, (see Figure 13). The side walls of the model were covered with crushed granite (0-16mm) (see Figure 14). It was placed in a steel chamber with a rubber liner inside and conveyor belts covering the openings to avoid the losses of the materials after each blasting (see Figure 15). After each blast the material was collected in buckets and marked for further analyzes. The model after blasting can be seen in Figure 16.
24
Figure 13. Fixing the side magnetic model
Figure 14. Filling the confined materials and preparing the test for blasting.
Figure 15. Closure of chamber and ready for blast.
25
Figure 16. Magnetic model after blasting.
3.2.4 Firing procedure
The VoD (Velocity of Detonation) was measured using the Datatrap II instrument (produced by MREL Company). The detonating cord was 2 m in length and the sampling frequency was 10 MHz. An electric detonator was used to initiate the PETN cord and explosive used was Riocord (produced by Maxam Company). Table 6 and Figures 18-20 shows the velocity of detonation for different cord strengths. The initiation sequence of the blast holes started from the center boreholes and moved towards the outer boreholes;
the detonating was in pairs of boreholes i.e. the same situation as in full-scale as shown in Figure 17. The delay time was 30 µsec between the holes for the tests with confined conditions in order the results to be comparable with the full scale and based on the experience of the previous researchers to be adjusted between the blast holes (Johansson, 2008). Also two rings were blasted simultaneously under free condition to check the difference in fragmentation with the delayed confined condition. Table 7 shows the charging length for both standard ring and future ring layout with 73 degree side hole.
26
Table 6. Velocity of detonation with different cords.
PETN cord strength, g/m 3.6 5 8.6
VOD, m/s 7155 7010 7016
Table 7. The charging length of boreholes.
Bh
#
Delay time (full scale)
ms
Delay time (model)
μs
Length (full scale)
m
Length (model)
m
Un-charged length (model)
m
Charging length (model)
m
1 275 90 25.8 0.47 0.04 0.43
2 250 60 35.1 0.65 0.15 0.5
3 225 30 44.0 0.84 0.06 0.76
4 200 0 54.4 1.00 0.26 0.74
5 200 0 54.6 1.00 0.06 0.94
6 225 30 43.7 0.81 0.19 0.61
7 250 60 34.8 0.64 0.07 0.56
8 275 90 25.4 0.47 0.04 0.43
Figure 17. Standard initiation at Kiruna mine. (Source:LKAB)
27
Figure 18. Maxam Riocord 3.6 g/m VOD test.
Figure 19. Maxam Riocord 5 g/m VOD test.
28
Figure 20. Maxam Riocord VoD test, 3.6 and 5 g/m cord twisted together resulting in a strength of 8.6 g/m.
3.2.5 Evaluation of fragmentation
The blasted material was collected from the chamber and placed into buckets and transported to a sieving lab at LKAB’s Kiruna mine. The collected blasted material consisting of magnetic mortar and also the debris was sieved by using a sieving machine with different mesh sizes. Materials smaller than 16mm was collected again and transported to a mineral processing lab at Luleå university for further processing. The material (samples) was separated in magnetic mortar and debris by using a magnetic separator (see in Figure 21). Finally, the magnetic mortar and debris were sieved for getting the fine part of the size distribution curve. Figure 22 show different stages of the sieving process.
29
Figure 21. Magnetic separator at LTU lab.
Figure 22. Stages of the sieving process (after Johansson, 2011).
Blasted material collected and further transported to the sieving lab at the
Kiruna mine.
Dry sieving (80-16) mm
Material less than 16mm collected and further
transported to LTU Magnetic seperation in the
mineral processing lab at LTU
Test materials (magnetic)
Dry sieving (16-0.063)mm Confining materials
Dry sieving (16-0.063)mm
30
The properties for five different tests for the standard ring layouts and one from the curved ring design (TC) with parallel bore holes are shown in Table 8.
Table 8.Properties of the blasting tests.
Test No T1 T2 T3 T4 T5 TC
Confinement Free Confined Free Confined Confined Confined Specific charge [ kg/m3] 1.9 4.5 1.2 1.6 2.9 2.8
Delay time [µsec] 0 30 0 30 30 30
Cord strength [g/m] 5 12 3.6 5 9 7.5
Burden [mm] 55 55 64 64 64 55
31
4. Results
Three tests have been done with the standard ring shape under confined conditions and two tests have been blasted under free conditions. All the properties of the tests have been presented in Table 8. The size distribution curves have been plotted for all the standard tests as shown in Figure 23. The specific charge and the average fragment size x50 have been presented in Table 9 with the fitted Swebrec function to the sieving data.
Table 9. The Swebrec function parameters for the magnetic specimens.
Test q
[ kg/m3]
x50 [mm]
xmax [mm]
b r2
T1(Free) 1.9 24.95 89.84 2.32 0.994
T2(Confined) 4.5 11.01 181.99 2.71 0.995
T3(Free) 1.2 30.95 103.89 2.21 0.996
T4(Confined) 1.6 29.83 80.01 1.58 0.996
T5(Confined) 2.9 29.32 99.71 2.01 0.997
TC(Confined) 2.8 20.39 80.00 1.67 0.998
The fragmentation results show that when the specific charge increases the fragmentation becomes finer. For instance, both of the tests T1 and T3 were free face shots and had cord strengths 5 g/m and 3.5 g/m respectively. For these tests, x50 is 24.9 mm for T1 (q= 1.9 kg/m3) and 30.9 mm for T3 (q= 1.2 kg/m3). Also by comparing the tests under confined conditions similar results are obtained. These tests were T2, T4 and T5 and having the cord strengths 12 g/m, 5 g/m and 9 g/m respectively.
32
Figure 23. Size distribution curves for the standard tests.
Here x50 for T2, T4 and T5 is 11.01mm, 29.6mm and 29.2mm respectively. These values show that the fragmentation is finer for T2 compared with T4 and T5. T2 has the highest specific charge (4.5 kg/m3). T5 shows unexpected jump (see Figure 23) and some explanations could be draw backs in the mould during blast test or some losses while collection of the blasted materials- the total amount of the blasted materials was 32.5 kg and about 50% of the materials have been sieved that also could be a reason. In Figure 24, x50 is plotted versus the specific charge. The effect of specific charge and confinement can be seen, though a more detailed analysis of each test is presented in Section 4.1.
1 7 70
0 1 5 50
Mass passing(%)
Sieve size (mm)
T1(Free) T2(Confined) T3(Free) T4(Confined) T5(Confined)
33 Average fragment size x50 (mm)
Figure 24. Average fragment size x50 versus with the specific charge q.
4.1 Size distribution curves for the standard ring (T5) versus curved layout TC (Magnetic Material)
TC and T5 have been compared, since both of the tests have been blasted with same properties and the specific charge was nearly the same (2.8 kg/m3 for TC and 2.9 kg/m3 for T5). The size distribution curves plotted (see Figure 25) and shows that the fragmentation is finer for TC. The specific charge and x50 for TC and T5 are shown in Table 10.
Table 10. Specific charge versus x50
Test No TC T5
State Confined Confined
Specific charge [ kg/m3] 2.8 2.9
x50 [mm] 20.4 29.2
1 10 100
1 x 50( mm)
Specific charge(kg/m3)
T1(Free) T2(Confined) T3(Free) T4(Confined) T5(Confined)
34
Figure 25. Size distribution curves for T5 and TC.
By analyzing the slope for free shots in the diagram for specific charge vs. average fragment size (see Figure 26). The results seems that is not subjected to equation (8), since the power of q factor (0.449) is not close to the 0.8 in the Kuz-Ram equation (Cunningham, 1987) and also not in the range 0.96 – 1.53 which has been found by Ouchterlony and Moser (2006).
For the confined shots (see Figure 26) the exponent is 0.979 i.e. a closer value to 0.8 in the Kuz-Ram equation (Cunningham, 1987) and lies in the range 0.96 – 1.53(Ouchterlony and Moser, 2006).
x50 α 1/q0.8 Eq(8) Where,
x50 = average fragment size (mm) q = specific charge (kg/m3)
0,1 1 10 100
0,01 0,1 1 10 100
Mass passing (%)
Sieve size(mm)
T5(Confined) TC(Confined)
35
Figure 26. Average fragment size x50 versus with the specific charge q.
4.2 Problems faced in the tests set-up
The curved ring layout TC has been compared with other small scale blasting tests which have been performed by researchers at LTU University since it was hard to continue the blasting tests which were planned due to some draw backs of the ring design (see Figure 27). The main problems which have been faced the blasting tests can be presented as follow:
The scale factor which has been used to scale down the ring for sublevel caving 1:54 considered small and by this scale factor it is difficult to scale down all the parameters like blast hole diameter and the delay time. The concrete side triangles for the magnetic moulds did not show effective way to simulate the confined condition for the blasting process and it was hard to fix it properly in the site, since it need time for curing and the blasting experiments was daily performed. Moreover, spacing between the concrete side triangles and the magnetic moulds was appearing after each blasting test and these consequences need also to fix it again in the proper position with the magnetic mould.
y = 33.129x-0,449 R² = 1
y = 48,403x-0,979 R² = 0.9994
1 10 100
1 10
Free Confined
Average fragment size X50 (mm)
Specific charge (kg/m3)
36
The sand bags which have been used on the side of the moulds was always damaging under the effect of the explosive energy and throwing of the fragmented blasted materials.
Figure 27. The main problems of the tests.
4.3 Comparing the curved ring layout TC with block tests
Small scale blasting tests had been carried out by Petropoulos (2011) and he used rectangular mortar blocks (see Figure 28) for his blast tests . He workedwith two different blast hole patterns The first design had 5 holes (ϕ =11 mm) and a specific charge of 2.56 kg/m3 and the second design had 7 holes (ϕ =11 mm) with a specific charge 4.16 kg/m3. The dimensions of the blocks were 660x215x270 mm and tthe scale factor was 1:51 for simulating the real case of SLC. The burden used was 58.3 mm. the block was confined by a U-shaped yoke made of reinforced concrete.
37
Figure 28. Set-up for confined shots (Petropoulos, 2011)
Petropoulos tested both under free face and under confinement. The tests under confined conditions with the specific charge 2.56 kg/m3 have been selected and compared with the curved ring design TC.
N1 and N2 were the most comparable tests in terms of specific charge with TC as it shown in Table 11.
Table 11. Specific charge versus x50 for curved SLC-design and blocks.
Test No TC N1 N2
State Confined Confined Confined
Specific charge [ kg/m3] 2.8 2.5 2.5
x50 [mm] 20.4 35.9 23.5
By comparing the average fragment sizes from N1 and N2 with the curved SLC-test (TC) the fragmentation, it shows finer fragmentation for TC. The conclusion of the comparison between these three tests is that the fragmentation of the ring test with curved holes TC is finer than the tests with the block test design when both of the tests blasted under the same boundary confined condition and has nearly the same specific charge. The number of tests is lacking, so more tests are needed to draw distinct conclusions. Another
38
constraint for this conclusion is the layout design for both of the tests and the confinement on the sides.
A series of small-scale blast tests was carried out by Johansson (2011) and he used magnetic mortar blocks as well in his set-up (see Figure 29). The planned block design was 650/660×205×300 mm (L× W× H) and had two rows with five Ø 10 mm blast holes in each row. The spacing and burden was 110 mm and 70 mm respectively. To minimize reflecting waves and to emulate full-scale geometry, the block was confined by a U- shaped yoke (Johansson, 2011, Petropoulos, 2011).
Figure 29. Setup for confined shots (after Johanson, 2011).
He carried out both free face and confined shots. The explosive used was decoupled PETN-cord with the strength of 20 g/m, giving a specific charge of 2.6 kg/m3 (Johansson, 2011). The results showed that the confinement made the fragmentation coarser and x50
increased nearly by 100 % when confining material was in front of the face to be blasted.
(Johansson 2011). The confined shots have been compared with the curved ring design TC.
39
Four tests have been selected from the trials. Two different inter-hole delays of 146 µsec and 73 µsec were used in these trials (as shown in Table 12) and each delay was repeated to check the repeatability of the tests (Johansson, 2011).
Table 12.x50, specific charge and nominal delay time for Johansson tests and TC(after Johansson, 2011)
Test no.
Row x50
[mm]
Specific charge [kg/m3]
Nominal delay [µsec]
10 1 66.8 2.6 146
10 2 54.6 2.6 146
11 1 64.9 2.6 73
11 2 45.8 2.6 73
12 1 52.2 2.6 146
12 2 46.2 2.6 146
13 1 56.6 2.6 73
13 2 44.9 2.6 73
TC 1 20.4 2.8 30
The fragmentation is finer for TC compared with Johansson tests (DT), see Figure 30.
The inter hole delay time is smaller for TC (30 µsec) compared with Johansson tests (73 and 146 µsec). The boundary conditions for both of the curved ring layout and Johansson test was confined. The rectangular blocks were confined by a U-shaped yoke and the curved ring layout was confined by using sand bags and steel box. The number of curved ring layout tests is only one so more research has to be carried out in order to get more reliable results.
40
Figure 30. Inter hole delay time versus x50.
4.4 Comparing the standard tests and TC for debris
The confined material for the tests has been prepared using a mixture of sand (0-8mm) and crushed granite (8-16 mm) which was mixed by the proportion of 50 % of each material giving a x50 of 8 mm. The parameters for the confined tests are presented in Table 13.
The size distribution curves in Figure 31 show that x50 is close to 7 mm for all the tests.
The confined material has become slightly finer around 12.5% when performing the blasts.
Table 13. Parameters for the confined tests.
Test no. Burden (mm)
Cord Strength (g/m)
Specific charge
(kg/m3) Condition
T2 55 12 4.5 Confined
T4 64 5 1.6 Confined
T5 64 9 2.9 Confined
TC 55 7.5 2.8 Confined
0 10 20 30 40 50 60 70 80
0 50 100 150 200
DT(Confined) TC(Confined)
Inter hole delay time(µsec) Average fragment size x50(mm)
41
Figure 31. Average fragment size x50 versus with the specific charge q for debris.
0,1 1 10 100
0,01 0,1 1 10 100
Mass Passing(%)
Sieve Size(mm)
T2 T4 T5 TC
42
43
5. Discussion and conclusions
The small-scaled blast tests reported in this thesis is a first step towards understanding of the actual behavior in a ring layout in terms of fragmentation. Tests have been carried out for both standard and the curved ring layout using the same material properties for both designs. The models were made of magnetic mortar in order to simulate the real situation as in sublevel caving and the models were confined with crushed granite 0-16 mm. The models were constructed by using the scale factor 1:54 which was selected according to the longest borehole in a real sublevel caving ring and the steel box had a height of 1 m. A total of six blast tests have been performed during this thesis work. The specific charge was varied for all the tests by changing the burden and the amount of explosive. Both free and confined shots were carried out and the results have been compared. The test set-up for the first five tests was a standard ring fan shaped design and the last test was the curved ring layout with parallel boreholes in the upper region of the SLC-ring. After some drawbacks and failure in the design of the set-up, only one test for the curved ring layout has been carried out. The curved ring layout has also been compared with earlier small scale blasting tests in blocks carried out by Johansson (2011) and Petropoulos 2011. They blasted confined shots using about the nearly the same specific charge, so the main difference between the tests was the geometry (blocks versus a ring) and the confinement on the sides. The fragmented materials were collected after every blasting test and laboratory sieving was performed. The size distribution curves were plotted for both the blasted burden and the confined material.
The fragmentation becomes finer for the standard rings when increasing the specific charge. Similar results can be expected for the curved ring layout, but only one test shot has been carried out in this thesis work. The fragmentation is finer for the curved ring layout compared with the standard ring layout, but this conclusion is based on only one shot. However more tests should be performed with the curved ring layout in order to get more reliable results.
44 Suggested future work for small-scale blast tests
Improve the design and the set-up of the tests by using bigger scale factor and use reinforced concrete instead of sand bags for the confined condition. That might help to prevent the back breakage.
The study the dynamic behavior of the blasted burden during the blasts tests was not carried out in this thesis work as planned due to the draw backs in the set-up.
Future study by using small scale tests could help to investigate this phenomenon.
Evaluate the fragmentation and the burden movement for different future ring designs (different angles for the side holes for example 60 degrees)
45
6. References
Brady, B. H. G., & Brown, E. T. (2004). Rock mechanics: For underground mining Springer.G. Bull and C. H. Page: ‘Sublevel caving – today’s dependable lowcost
‘ore factory’, Proc. Conf. MassMin 2000, Brisbane, Qld, Australia, October–
November 2000, Australasian Institute of Mining and Metallurgy, pp.537–556.
Cullum, A. (1974a). The Effects of Confined Blasting on Rock Fragmentation and Flow Characteristics in Sub Level Caving, MSc thesis, 161 pp, Brisbane: Univ. of Queensland.
Cunningham, C. (1983). The kuz-ram model for prediction of fragmentation from blasting. Proceedings of the First International Symposium on Rock Fragmentation by Blasting, Lulea, Sweden, Vol. 2, pp 439-453.
Cunningham, C. (1987). Fragmentation estimations and the kuz-ram model-four years on.
2nd International Symposium on Rock Fragmentation by Blasting, Keystone, Colorado, pp 475-487.
Data trap II, (http://www.mrel.com/blasting_instrumentation/flash/datatrap.html)
Hustrulid, W (2000) ‘Method selection for large-scale underground mining’. In MassMin 2000, Proc. 3rd Int. Conf. & Exhib. on Mass Mining, pp 29–57, Chitombo, G.
(ed.).AusIMM, Brisbane, Australia.
Hamrin, H., Hustrulid, W., & Bullock, R. (2001). Underground mining methods and applications. Underground Mining Methods: Engineering Fundamentals and International Case Studies, SME, Littleton, CO, , pp. 3-14.
Hustrulid, W., & Kvapil, R. (2008). Sublevel caving–past and future. Proceedings 5th International Conference and Exhibition on Mass Mining, MassMin,pp 107-132.
Janelid, I., & Kvapil, R. (1966). Sublevel caving. International Journal of Rock Mechanics and Mining Sciences & Geomechanics Abstracts, , 3(2) 129-132.
Jarlenfors, J and Holmberg, R, 1980. Blasting against a compressible rock mass – model scale experiments in PMMA. Swedish Detonic Research Foundation, Report DS 1980:5, (in Swedish), Stockholm, Sweden, p 35.
Johansson, D., Ouchterlony, F., & Nyberg, U. (2007). Blasting against aggregate confinement, fragmentation and swelling in model scale. In Proceedings of the 4th EFEE World Conf on Explosives and Blasting, EFEE, England, pp. 13-27.
46
Johansson, D. (2008). Fragmentation and waste rock compaction in small-scale confined blasting. Licentiate thesis 2008:30. Lulea: Lulea Univ. Techn.
Johansson, D., 2011. Effects of confinement and initiation delay on fragmentation and waste rock compaction.Results from small scale blasting tests.:Doctoral Thesis.
Luleå University of Technology Luleå, Sweden.
Johansson, D. & Ouchterlony, F., 2011. Fragmentation in small-scale confined blasting..
s.l.:International Journel.Mining and Mineral Engineering,Vol 3,No 1.
Katsabanis, P., Tawadrous, A., Kennedy, C., & Brown, C. (2006). Timing effects on fragmentation. PROCEEDINGS OF THE ANNUAL CONFERENCE ON EXPLOSIVES AND BLASTING TECHNIQUE, , 32(2) 243.
Olsson, M (1987). Blasting against a compressible bed of fragmented ore. Swedish Detonic Foundation, Report 1987:5 (in Swedish), 56pp, Stockholm, Sweden.
Ouchterlony, F. (2005a). The Swebrec© function: Linking fragmentation by blasting and crushing. Mining Technology: Transactions of the Institute of Mining and Metallurgy, Section A, 114(1), 29-44. PP A29 – A44.
Ouchterlony, F. (2005b). What does the fragment size distribution of blasted rock look like? Brighton Conference Proceedings: Third European Federation of Explosives Engineers (EFEE) World Conference on Explosives and Blasting, pp189-199
Ouchterlony, F., Olsson, M., Nyberg, U., Andersson, P., & Gustavsson, L. (2006).
Constructing the fragment size distribution of a bench blasting round, using the new swebrec function. Proc. 8th Int. Symp. On ‘Rock Fragmentation in Blasting’, Santiago, Chile, Fragblast 8, no. 333.
Ouchterlony,F. and Peter Moser (2006). "Likenesses and differences in the fragmentation of full-scale and model-scale blasts." Proceedings of the 8th International Symposium on Rock Fragmentation by Blasting, pp.207–220.
Petropoulos, N. 2011. Influence of confinement on fragmentation and investigation of the burden movement. Master's thesis. Luleå,Sweden: Luleå University of Technology Luleå,Sweden.
Riocord, (http://www.maxam.net/buscador_ceprods_action/9887/9717)
Volchenko, N. (1977). Influence of charge arrangement geometry and short-delay blasting on the crushing indices in compression blasting. Journal of Mining Science, 13(5), pp. 488-493.
47
Wimmer, M, Nordqvist, A,Ouchterlony, F. Selldén, H & Lenz, G (2012) . "3D mapping of sublevel caving (SLC) blast rings and ore flow disturbances in the LKAB Kiruna mine." In 6th Int Conf and Exhib on Mass Mining.
Wimmer, Matthias (2012). Towards Understanding Breakage and Flow in Sublevel Caving (SLC) - Development of new measurement techniques and results from full- scale tests.: Doctoral thesis. Luleå University of Technology Luleå, Sweden.
48
7. Appendix-Sieving data for the tests
I. Sieving data for all the tests for magnetic mortar.
Test no Confinement Sieving method Charge[g/m] Delay time[µsec]
1 Free Dry 5 0
Sieve size(mm) Mass passing (%)
80 100
65 93.73
50 83.25
45 82.63
32 68.32
25 50.09
20 37.18
16 30.35
11.2 23.04
5.6 15.14
2 10.21
1 8.521
0.5 4.801
0.25 3.214
0.125 2.047
0.063 1.287
Test no Confinement Sieving method Charge[g/m] Delay time[µsec]
2 Confined Dry 12 30
Sieve size(mm) Mass passing (%)
80 100
65 100
50 85.51
45 85.51
32 77.33
25 71.09
20 66.06
16 61.37
11.2 50.99
5.6 34.60
2 22.30
1 17.31
0.5 12.28
0.25 8.23
0.125 5.28
0.063 3.82
Test no Confinement Sieving method Charge[g/m] Delay time[µsec]
3 Free Dry 3.6 0
Sieve size(mm) Mass passing (%)
80 100
65 84.68
50 77.93
45 67.89
32 54.39
25 40.78
20 31.39
16 27.51
11.2 20.64
5.6 11.93
2 7.14
1 5.65
0.5 4.29
0.25 3.12
0.125 2.03
0.063 1.41
Test no. Confinement Sieving method Charge[g/m] Delay time[µsec]
4 Confined Dry 5 30
Sieve size(mm) Mass passing (%)
80 100
65 93.92
50 78.29
45 68.83
32 52.64
25 41.51
20 36.09
16 30.60
11,2 25.07
5,6 18.18
2 13.89
1 11.23
0.5 8.05
0.25 5.54
0.125 3.35
0.063 1.62
Test no. Confinement Sieving method Charge[g/m] Delay time[µsec]
5 Confined Dry 9 30
Sieve size(mm) Mass passing (%)
80 100
65 88.03
50 76.78
45 67.09
32 55.13
25 44.87
20 37.07
16 32.11
11.2 21.28
5.6 15.53
2 10.97
1 6.87
0.5 3.28
0.25 2.08
0.125 1.51
0.063 0.97
Test no. Confinement Sieving method Charge[g/m] Delay time[µsec]
TC Confined Dry 7.5 30
Sieve size(mm) Mass passing (%)
80 100
65 96.12
50 83.46
45 82.38
32 68.84
25 56.05
20 47.91
16 42.59
11.2 34.93
5.6 24.23
2 17.27
1 14.19
0,5 10.96
0.25 8.16
0.125 5.65
0.063 3.99
II. Sieving data for the debris for the tests with confinements
Test no Confinement Sieving method Charge[g/m] Delay time[µsec]
2 Confined Dry 12 30
Test no. Confinement Sieving method Charge[g/m] Delay time[µsec]
4 Confined Dry 5 30
Sieve size(mm) Mass passing (%)
20 100
11.2 56.59
5.6 46.41
2 35.47
1 27.97
0.5 15.21
0.25 6.59
0.125 2.11
0.063 0.96
Sieve size(mm) Mass passing (%)
20 100
11.2 56.68
5.6 49.03
2 38.24
1 30.68
0.5 16.21
0.25 5.81
0.125 2.48
0.063 1.24
Test no. Confinement Sieving method Charge[g/m] Delay time[µsec]
5 Confined Dry 9 30
Sieve size(mm) Mass passing (%)
20 100
11.2 58.93
5.6 52.71
2 42.68
1 34.52
0.5 18.60
0.25 7.17
0.125 3.73
0.063 2.48
Test no. Confinement Sieving method Charge[g/m] Delay time[µsec]
TC Confined Dry 7.5 30
Sieve size(mm) Mass passing (%)
20 100
11.2 58.80
5.6 45.97
2 36.07
1 30.49
0.5 20.77
0.25 9.03
0.125 4.87
0.063 2.78
III. Fragment size distributions curves for magnetic mortar
Figure 32. Fragment size distributions curve for T1.
Figure 33. Fragment size distributions curve for T2.
Test 1 (Magnetic mortar) Eqn 8001 (a,b,c)
r^2=0.99487827 DF Adj r^2=0.99359784 FitStdErr=2.7684307 Fstat=1262.6031 a=2.3269818 b=3.6004189
c=89.840807
0.01 0.1 1 10 100
Mesh size [mm]
1 10
Mass Passing [%]
1 10
Mass Passing [%]
-3 -1 1 3 5 7
Residuals [5]
-3 -1 1 3 5 7
Residuals [5]
2nd Eqn 8001 (a,b,c)
r^2=0.99557528 DF Adj r^2=0.99436854 FitStdErr=2.4151079 Fstat=1350.019 a=2.7121535 b=16.531468
c=181.99155
0.01 0.1 1 10 100
Mesh size [mm]
1 10
Mass Passing [%]
1 10
Mass Passing [%]
-3 -1 1 3 5 7
Residuals [5]
-3 -1 1 3 5 7
Residuals [5]
