1 COMBINED BLAST AND FRAGMENT LOADING EFFECTS ON REINFORCED CONCRETE STRUCTURES Yau Jia Ming Spencer 1 , Kang Kok Wei 2 1 Anglo Chinese Junior College, 25 Dover Close East Singapore 139745 2 Defence Science and Technology Agency, 1 Depot Road, Singapore 109679 ABSTRACT Concrete has been used since the Roman Empire and has been made stronger in modern times with the addition of steel reinforcing bars (rebars), which form reinforced concrete (RC). The cost effectiveness, reasonable strength and high malleability has contributed to its popularity in the construction industry. This research study aims to understand the response of RC slabs against combined blast and fragment loadings in the design of protective structures. Since small countries such as Singapore suffer from land space constraint, practical experiments are limited thus the need for computation software such as LS-DYNA, which is used in this study. Through LS-DYNA, parameters such as arrangement of rebars and boundary conditions, have been varied to study the response of RC against blast as well as combined blast and fragment loadings. Findings include the decreasing relationship of the damage extent of the RC slab to increments of rebars as well as the stark difference in the response of the RC against combined blast and fragment loading compared to its response against blast loading solely. INTRODUCTION Resilience of reinforced concrete (RC) structures to dynamic loadings has been well researched in the protection of human or equipment within buildings. The loading effects from conventional weapons include both blast and fragments. The latter is generated from the breakup of metal casing around the explosives within. While blast and fragment loadings are well documented individually, there are limited data on the combined blast and fragment loading effects. The objective of this research is to analyse the physical response of an RC slab to combined blast and fragments loadings. Since small countries such as Singapore suffer from land constraints, practical experiments to study the combined blast and fragment effects are limited thus the need for computation software such as LS-DYNA. Apart from saving resources, numerical analysis using these computational software develop trends or patterns without the need for numerous live testing. However, such software and simulations run on equations that are built on data only found through experiments. Hence, some live testing is still required. This paper is divided into 2 sections: Blast Loading on RC Slabs and Combined Blast and Fragments Loading on RC Slabs. Each section will discuss about their respective loading, present the methods and discuss the results. RESEARCH APPROACH In the analysis of the RC slab subjected to various dynamic loadings, the software, LS- DYNA, which originated from the Lawrence Livermore National Laboratory in 1976, is used. This software is primarily used to numerically analyse structures which are subjected to a variety of impact loading. For this research, we will be analysing the stress visually and z- displacement graphically of the RC slab when it is loaded. An example of stress analysis can
16
Embed
COMBINED BLAST AND FRAGMENT LOADING EFFECTS ON REINFORCED CONCRETE STRUCTURES · · 2016-03-111 COMBINED BLAST AND FRAGMENT LOADING EFFECTS ON REINFORCED CONCRETE STRUCTURES Yau
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
COMBINED BLAST AND FRAGMENT LOADING EFFECTS ON REINFORCED
CONCRETE STRUCTURES
Yau Jia Ming Spencer
1, Kang Kok Wei
2
1Anglo Chinese Junior College, 25 Dover Close East Singapore 139745
2Defence Science and Technology Agency, 1 Depot Road, Singapore 109679
ABSTRACT
Concrete has been used since the Roman Empire and has been made stronger in modern
times with the addition of steel reinforcing bars (rebars), which form reinforced concrete
(RC). The cost effectiveness, reasonable strength and high malleability has contributed to its
popularity in the construction industry. This research study aims to understand the response
of RC slabs against combined blast and fragment loadings in the design of protective
structures. Since small countries such as Singapore suffer from land space constraint,
practical experiments are limited thus the need for computation software such as LS-DYNA,
which is used in this study. Through LS-DYNA, parameters such as arrangement of rebars
and boundary conditions, have been varied to study the response of RC against blast as well
as combined blast and fragment loadings. Findings include the decreasing relationship of the
damage extent of the RC slab to increments of rebars as well as the stark difference in the
response of the RC against combined blast and fragment loading compared to its response
against blast loading solely.
INTRODUCTION
Resilience of reinforced concrete (RC) structures to dynamic loadings has been well
researched in the protection of human or equipment within buildings. The loading effects
from conventional weapons include both blast and fragments. The latter is generated from the
breakup of metal casing around the explosives within. While blast and fragment loadings are
well documented individually, there are limited data on the combined blast and fragment
loading effects. The objective of this research is to analyse the physical response of an RC
slab to combined blast and fragments loadings.
Since small countries such as Singapore suffer from land constraints, practical experiments to
study the combined blast and fragment effects are limited thus the need for computation
software such as LS-DYNA. Apart from saving resources, numerical analysis using these
computational software develop trends or patterns without the need for numerous live testing.
However, such software and simulations run on equations that are built on data only found
through experiments. Hence, some live testing is still required.
This paper is divided into 2 sections: Blast Loading on RC Slabs and Combined Blast and
Fragments Loading on RC Slabs. Each section will discuss about their respective loading,
present the methods and discuss the results.
RESEARCH APPROACH
In the analysis of the RC slab subjected to various dynamic loadings, the software, LS-
DYNA, which originated from the Lawrence Livermore National Laboratory in 1976, is
used. This software is primarily used to numerically analyse structures which are subjected to
a variety of impact loading. For this research, we will be analysing the stress visually and z-
displacement graphically of the RC slab when it is loaded. An example of stress analysis can
2
be seen in Figure 1 [1]. Usually, the red regions will indicate high stress values and vice versa
for blue regions.
Figure 1: An example of visual stress analysis [1]
Packaged together with the LS-DYNA Solver is a software called LS-PrePost. This is a both
a PreProcessor as well as a PostProcessor. The former allows users to setup models prior to
the analysis while the latter enables users to analyse and visualise the output from LS-DYNA.
For this research study, the PreProcessor is used to create a slab model, which measures
3x1x0.2m and the following parameters are varied:
Boundary conditions,
Element Size
Arrangement and number of loading segments
Arrangement and number of rebars
Figure 2 Measurements of the RC slab model
Details of the parameters above will be described in subsequent sections. The rest of the
parameters such as the ones below can be created using LS-PrePost but, in this study, a
separate file is written using the computer program WordPad to specify these values into the
model:
charge mass,
stand-off distance, and
the type of loading (blast and combined blast and fragments for this research)
An example of the file is included in the Annex A.
After analysing the model using LS-DYNA, the PostProcessor allows the user to analysis the
damage and response of the RC slab loading visually and graphically. For the graphs, only
the first peak deflection is considered as the subsequent oscillations of the slab is deemed
unphysical. Given more time, this problem can be numerically resolved.
0.2m
1m
3
BLAST LOADING
This section describes the analysis of a RC slab subjected to blast loadings, which assumes
the use of bare Trinitrotoluene (TNT). The shape of the charge is assumed to be spherical and
it is detonated in the air. This is known as airblast. Variations of the quantity and arrangement
of rebars, boundary conditions, element size and scaled distances were studied in this
research. Fortunately, LS-DYNA can calculate the blast load data. Hence there is little need
for manual calculations. Grade 30 concrete and Grade 460 steel for the rebars in the model.
Quantity & Arrangement of rebars
In RC slabs, rebars are important as they contribute greatly to the tensile capacity of RC.
Concrete is highly resistant to compressive stresses but responds poorly to tensile stress.
Therefore, the quantity and arrangement of rebars can alter the capacity of RC slabs to blast
loading. For this study, 20x20x20mm solid elements are used to model concrete while the
rebars are modelled using 20mm long beam elements. An explosive charge weighing 10kg,
which is placed 2m above the top surface of the slab, is detonated in this case. Three models
are created in this study. The models in Figures 3(a) and 3(b) have two layers of rebars but
the number of rebars per layer is varied: 5 rebars per layer for Figure 3(a) and 10 rebars per
layer for Figure 3(b). The model in Figure 3(c) is similar to the one in Figure 3(a) but without
a layer of rebars near the top surface.
(a) (b) (c)
Figure 3 Models used to study the effects of varying the rebar quantity and
arrangement
By comparing images in Figures 4(a) and 4(b) and the curves in Figure 5 (Figure 4(a) having
10 rebars and Figure 4(b) having 20 rebars), there is a decrease in the overall midspan
vertical- or z-displacement of the slab and damage when the quantity of rebars increase. The
same can be said when the number of rebar layers increase in the case of Figures 4(a) and
4(c) (Figure 4(c) having 5 rebars). When the slabs are bent as shown in Figure 6, the top
surface of the slab is subjected to compression while the bottom of the slab is subjected to
tension. Since rebars have a higher tensile strength than concrete, the overall tensile strength
of the slab increases with increments in rebar content, which reduces the overall z-
displacement of the slab as well as the damage.
(a) (b) (c)
Figure 4 Damage of the soffit of the slabs with various rebar arrangements after loading
4
Figure 5 Midspan displacement histories of the slabs with various rebar arrangements
Boundary Conditions
Figures 6 Bending response of the RC slabs under blast loading
There are 2 main types of supports for RC structures: pinned and fixed. Pinned support
provides translational restraints but not rotational restraint at the ends whereas fixed support
provides both translational and rotational restraints. Two models were created to study the
effects of these boundary conditions. Figures 6(a) & 6(b) represent fixed and pinned supports
respectively. The entire surfaces of the model in Figure 6(a), which are darkened, are
restrained whereas, for the model in Figure 6(b), only an edge on the soffit of the slab has
been restrained. They are also illustrations of both boundary conditions and their appearances
in the models [2]. The other parameters remain unchanged and the models are similar to the
one in Figure 3(a).
(a) (b)
Figure 6 Models used to study the effects of varying the boundary conditions
The RC slab with fixed supports in Figure 7(a) suffered less damage than the slab with
pinned supports in Figure 7(b). The z-displacement is significantly greater for the slab with
5
pinned supports than the slab with fixed supports as seen in Figure 8. Pinned supports are
commonly seen in pre-cast concrete buildings such as the many HDB flats in Singapore. An
example of a structure with fixed supports is a bomb shelter as it is a single piece of structure
with no weak linkages. Since fixed supports offer more resistance to rotational movement
(moment) than Pinned supports, the slab with fixed supports will be more resilient to loading
forces and pressure exerted, resulting in less damage; as observed in Figures 7(a) and 7(b).
(a) (b)
Figure 7 Damage of the soffit damage of the slabs with (a) fixed and (b) pinned supports
after loading
Figure 8 Midspan displacement histories of the slabs with fixed and pinned supports
Element Size
In LS-PrePost, 2 meshes of the RC slab described in Figure 3(a) are formed but the number
of elements are varied as it was believed that this parameter has an effect on the RC slab
response to impact loading. The slab in Figure 9(a) consists of 1,000 solid elements of
300x100x20mm while the slab in Figure 9(b) comprises of 75,000 solid elements of
20x20x20 mm.
(a) (b)
Figures 9 Models used to study the effects of varying the element size
The slab made of small solid elements in Figure 10(a) has suffered significantly less damage
from the impact loading than the slab made of large elements in Figure 10(b). At stand-off
distances of 1m & 2m and charge mass of 10kg, there is almost no difference in the z-
displacement of the slab as seen in Figure 11. However, for the stand-off distance at 2.15m
and charge mass of 100kg, the slab in Figure 10(a) has a smaller z-displacement than the slab
in Figure 10(b) as seen in Figure 11; showing an increasing relationship of the z-displacement
and the stress as the quantity of elements increases. It could be due to the average impulse,
6
from the blast loading, per element being higher for the slab in Figure 10(a) than the slab in
Figure 10(b); tallying with the results.
(a) (b)
Figure 10 Damage of the soffit of the slabs with (a) small and (b) large elements after
loading
Figure 11 Midspan displacement histories of the slabs with various element sizes at
various stand-off distances and charge mass
Scaled Distance
Scaled distance is defined as the ratio R/C1/3
where R is the stand-off distance of the
explosive charge from the target and C the explosive charge mass. Most blast parameters can
be derived from scaled distance as shown in Figure 12. Two scaled distances are studied in
this section: 0.46m/kg1/3
and 0.92m/kg1/3
. The former is based on a 10kg explosive mass at
stand-off distance of 1m while the latter is based on the same explosive quantity at a stand-off
distance of 2m. For the former, another scenario of a 100kg explosive mass with a stand-off
distance of 2.15m is studied.
Figure 12 Derivation of various blast parameters based on scaled distance [3]
Figures 13(a), 13(b) and 13(c) shows the damage of the three scenario studied. As expected, a
higher scaled distance will lead to less damage by comparing Figures 13(a) and 13(b), in
which the charge weights are the same but the stand-off distances are 1m and 2m
7
respectively. It is also noted that although the scaled distances are the same for the slabs in
Figure 13(a) and 13(c), the use of a higher charge mass on the slab in Figure 13(c) resulted in
a larger amount of explosive materials reacted, producing more kinetic energy and thus a
very powerful shockwave. This will lead to more damage. The z-displacement histories of the
three scenarios are plotted in Figure 14 and the results relate well with the damage
comparisons in Figure 13. Thus, in addition to scaled distance, the individual components of
this expression i.e. the explosive charge mass and stand-off distances are also important.
(a) (b) (c)
Figure 13 Damage of the soffit of the slabs with various scaled distances of (a)