Civil Engineering and Construction Management
http://hdl.handle.net/11141/1068
See also: Construction Management [former]; Civil Engineering [former]Sun, 12 May 2019 01:51:08 GMT2019-05-12T01:51:08ZDesign and Analysis of Fiber Reinforced Concrete Structure for Transportation Infrastructures
http://hdl.handle.net/11141/2808
Design and Analysis of Fiber Reinforced Concrete Structure for Transportation Infrastructures
Wtaife, Salam Adil Mutlag
Concrete is one of the most popular materials in superstructures (buildings), substructures (foundation) and infrastructure facilities (bridges, pavements, and tunnels). However, concrete is a brittle material that cracks easily under tension.
Fiber may be used as reinforcement to impede the concrete from cracking as well as to increase the concrete flexural strength. Nevertheless, current design codes do not include this design enhancement. Furthermore, there are discrepancies in the design methodology concerning the inclusion of discrete fibers in concrete structures. Therefore; there is a need to develop new analysis methodology and design methodology for fiber reinforced concrete (FRC) and fiber reinforced cement composite (FRCC).
The aim of this study is to deepen the knowledge to analyze and design the fiber-reinforced concrete by developing a novel method to predict the stress compression and tension blocks at the design ultimate limit state. To accomplish this, an experimental program was developed to characterize FRC and FRCC using common test methods. Three group of fibers were investigated: steel, PVA, and synthetic fiber with volume fractions ranging from 0 to 2%. The experimental program consisted of compression tests, flexural tests, and direct tensile tests which were used to assess the fibers’ usefulness as reinforcement components and to compare to the assumptions of current design methods. Experimental results of this research, combined with an additional 1,120 data points obtained from other researchers, were statistically analyzed to develop a new model to predict the design stress block of FRC and FRCC so that their design. This dissertation is divided into four parts:1) development of stress block in compression, 2) development of stress block in tension, 3) development of analysis, and 4) development of design of the new FRC and FRCC components.
First, the development of stress block in compression requires the yield and ultimate strains to be know.
A new equation for yield strain was developed by modifying the American Concrete Institute (ACI) Committee 544, fiber reinforced concrete and Rilem equations that depend on the compressive strength. An ultimate strain of 0.0035 and 0.005 at the extreme concrete compression fiber for the volume fraction of less than 1% and for a volume fraction of more than or equal to 1%, respectively, are proposed. New parameters are also introduced to account for different fiber types. Steel, PVA, basalt and synthetic fibers, two compresive stress block shapes are proposed. The two shapes consist of a rectangular stress block commonly used in regular concrete and a tapezoid stress block.
For the rectangular compressive stress block, two constants, 𝐾1and 𝐾2, are proposed. These constants are affected by the volume fraction (Vf) of FRC and FRCC. For 𝑉𝑓≥1%, 𝐾1 and 𝐾2 are constant value of 0.75 and 0.375 for FRC and FRCC, respectively, while for 𝑉𝑓<1%, they depend on the concrete compressive strength of FRC and FRCC. To this end, there are two additional parameters, 𝛽 and ∝, which depend on the strain at the elastic, yield, and the ultimate stages. For the ultimate stage, β, which is also affected by the volume fraction of fiber, is proposed to be the same as the ACI 318 Code value for 𝑉𝑓<1%. While for 𝑉𝑓≥1%, 𝛽 depends on the FRC and FRCC concrete compressive strength. On the other hand, ∝ depends on the FRC and FRCC concrete compressive strength for 𝑉𝑓<1%. While for 𝑉𝑓≥1%, it may be observed that it becomes approximately a constant value of 1.0.
For the trapezoid stress block in compression, an idealized constitutive model which is a bilinear, elastic-perfectly plastic stress-strain response, has been a proposed. It is ssumed in compression that the linear portion of the response terminates at a yield point (∝fc',𝜀𝑐𝑦) and remains perfectly plastic at the compressive yield stress until the ultimate compressive strain 𝜀𝑈𝑙𝑡.. ∝ of fiber-reinforced concrete for ultimate design is taken to be 0.85, which is the same as the ∝ of ACI 318 for volume fraction less than 1%, and 1 for 𝑓𝑐′≥69 𝑀𝑝𝑎, ∝=1. The proposed models were compared to six design codes using 250 data points obtained from previous studies.
The evaluation for an area of the database which used the other codes more than twice underestimated the lower bond of the database. Therefore, the other codes do not give a valid evaluation for compressive strength due to neglecting the effect of fiber. For volume fraction of fiber more than 1, the rectangular and trapezoid area of stress blocks are almost matches. While for volume fraction less than 1 and compressive strength more than 40 MPa, there is a small difference between the rectangular and trapezoid stress blocks. For 𝐾2 value, the rectangular of stress block is in the lower bond of the database, while the trapezoid stress block model is near to the average of the database. As a result, the rectangular stress block is underestimated by the database and also is easy for a designer to use. Therefore, it will be used in the new design and analysis proposal model.
Second, determining tension stress block by knowing the first crack strength, first crack strain, elastic modulus, and ultimate strain in tension is necessary. More than 250 points of data are used to evaluate the first crack strength. The new empirical relations developed ACI 318 equation by multiple factor 𝜆. Factor 𝜆 depends on the length of the fiber. In this dissertation, factor 𝜆 is determined by the average and lower bond of the database.
Third, analysis of the new model for the ultimate stage is assumed to have a rectangular stress block with an average uniform stress. Also, this model adopts ∅ equal to 0.75 because of safe.
Comparing the proposal model with a moment of ACI 544 model shows that the proposed model is safer and more accurate than the ACI 544 model because the ACI 544 model has overestimated moment value for more than 3.5 Kn.m.
Finally, in the design of the new model for the ultimate stage, the majority of the design volume fraction for a proposal is more than the measurement volume fraction for the database. According to the database, this model works for volume fraction ≤2. It is necessary for the designer to know what kind of behavior is appropriate for each design. Through that, it is possible to precisely estimate the volume fraction of the requirement of each behavior as will be explained below. The volume fraction of fiber is critical for the response of strain-softening (hardening deflection). In the case of the strain of softening, the internal moment provided by FRC and FRCC is to resist the external moment more than the first crack moment. In addition to that, the volume fraction of fiber is critical for the response of strain-hardening. In the case of the strain hardening for FRC and FRCC, the strength of tension is defined for the stress block model as a constant value. FRC and FRCC strengths in tension are more than the first crack strength.
In conclusion of this dissertation, twelve equations were evaluated:
1. yield strain in compression,
2. ultimte strain in compression,
3. first crack strength in tension,
4. first crack strain in tension,
5. the elastic modulus in tension,
6. neutral axis for the elastic stage in tension,
7. ultimate strain in tension,
8. moment capacity for analysis and design model,
9. tensile strength for analysis and design model,
10. volume fraction equation of fiber for the design model,
11. volume fraction equation as a minimum requirement for deflection- hardening, and
12. volume fraction equation as a minimum requirement for strain-hardening.
In addition, there were the five parameters, in which 𝐾1𝑎𝑛𝑑 𝐾2 factor for stress block in compression and 𝛽 factor for rectangular stress block in compression and α factor for rectangular and trapezoid stress block in compression. In summary, this dissertation proposes new design and analysis for fiber-reinforced concrete and fiber-reinforced cement in order to increase safety and to provide an easier process method for the designer.
Thesis (Ph.D.) - Florida Institute of Technology, 2019
Wed, 01 May 2019 00:00:00 GMThttp://hdl.handle.net/11141/28082019-05-01T00:00:00ZEvaluation of Bond Properties between Fiber Reinforced Concrete Overlay and Substrate Concrete
http://hdl.handle.net/11141/2803
Evaluation of Bond Properties between Fiber Reinforced Concrete Overlay and Substrate Concrete
Alshammari, Emad Okrush
Bonded concrete overlay is the most economical and quick option in concrete
rehabilitation methods to provide strength for the existing structure. Bonded
concrete overlay strength rely on properties of both layers (substrate and new
layers). Poor bonding in interface zone leads to two main failures of concrete
overlays; debonding and cracks. There are several traditional factors affect bond
strength such as water to cement (w/c) ratio, moisture condition, and surface
roughness for substrate layer. However, the effect of using fiber consider a new
variable affecting bond strength.
Fiber reinforced concrete (FRC) is one type of overlay system which has been used
for applications such as pavements, slope stabilization, arches, and domes. Adding
discrete fibers in the mortar or concrete mixture significantly enhance the bond
strength between substrate layer and overlay by increasing the cohesion at the
interface, decreasing curling strain, restraining the development of cracks and
spread them into several finer cracks.
The objective of this study is to investigate how the bond strength between the
existing and overlay layer under two loading conditions (indirect tension and direct
shear) is influenced by adding Polyvinyl alcohol fiber (PVA) fiber and
Polypropylene fiber. This study was performed by using two bond tests: splitting prism tensile test and
direct double shear test. The splitting prism test form indirect tensile stress on the interface zone while the direct double shear test form direct shear stress on
interface zone. The substrate surfaces were roughened by wire brush and the
specimens were cured for 28 days of curing. Both layers use a 0.46 (w/c) ratio.
Two types of fiber were added to overlay layer with different volume fraction
dosages. Analysis of study results found that the bond strength (tension and shear)
significantly increases when using fiber reinforced concrete (FRC) as an overlay
layer. A relation was found between increasing fiber content and increasing bond
strength. However, there is no relation found between bond strength and fiber type.
In most FRC cases, the bond strength was greater than the control case, and the
improvement was more than 600% in some cases. Adding higher fiber volume fraction dosage leads to higher bond strength. However, volume fraction dosage
more than 1.5% causes less workability for the concrete mixture which leads to a
decrease in bond strength. Low volume fraction dosage (≤ 0.5%) did not show a
significant improvement in tensile and shear bond strength. Strong correlation was
found between shear bond strength and tensile bond strength.
Thesis (M.S.) - Florida Institute of Technology, 2018
Sat, 01 Dec 2018 00:00:00 GMThttp://hdl.handle.net/11141/28032018-12-01T00:00:00ZLateral and Torsional Seismic Vibration Control for Torsionally Irregular Buildings
http://hdl.handle.net/11141/2800
Lateral and Torsional Seismic Vibration Control for Torsionally Irregular Buildings
Akyϋrek, Osman
During strong earthquakes or wind gusts, it is likely that buildings with torsional
irregularity in the plan have an can be seriously damaged, partially collapsed or fully
collapsed. This is because Torsionally Irregular Buildings (TIBs) may have
significant aerodynamic torsion loads that increase the eccentricity between the
center of mass and the center of rigidity, especially in dominant torsion modes. For
this reason, torsion leads to excessive increase in lateral motions when dynamic loads
excite the buildings.
Torsional irregularity is one of the main failure causes during strong dynamic
excitations due to earthquakes or wind gusts. Ignoring torsional irregularity in
seismic design analysis can cause unexpected damages and losses. To enhance the
safety and performance of buildings, most of the current seismic provisions address
this irregularity in two main ways. The first is computing torsional moment at each
floor by using equations provided in various current seismic code provisions. After they are applied on each floor, the seismic analysis will be performed. The second is
shifting the center of mass (CM) or stiffness (CS) to eliminate the eccentricity by
putting additional masses or structural components such as braced frame systems on
buildings.
This research developed and validated a new torsionally effective control system for
the purpose of enhancing the performance/safety and mitigating structural failure in
Torsionally Irregular Buildings (TIBs) under bidirectional strong earthquake loads.
It introduces the new integrated control system (ICS) applied to a benchmark 9-story
steel building developed for the SAC project in California to suppress the undesirable
lateral and torsional coupling effects due to eccentricity. The dynamic responses of
the system were evaluated under N-S and W-E components of the real earthquake
excitations of the El Centro (1940), Loma Prieta (1989) and Kocaeli (1999)
earthquakes. First the traditional method (cross-braced frame systems) was
implemented in the benchmark building with different pre-determined placement
layouts. The most effective placement was determined and the benchmark building
was analyzed with that for comparison purpose. Secondly, tuned mass dampers
(TMDs) were designed and applied to start from the center of mass (CM) through
two translational directions under bi-directional seismic loads such as N-S and E-W
components of selected ground motions. Then the performance evaluation for TMDs
was determined. The effectiveness of the TMD system was evaluated in terms of
energy analyses and performance evaluation criteria including maximum floor displacement, maximum drift, and maximum floor acceleration. Based on these
comparisons, there is a substantial reduction of the amplitudes of the frequency
response validated the effectiveness of the ICS in controlling the seismic responses
for two-way eccentric elastic buildings. Unlike traditional TMDs placed in two
orthogonal directions, the ICS is more comprehended to control not only two
orthogonal (x- and y-) directions, but also effectively control rotational (θ-) direction.
By means of the proposed system configuration, the structures first-three dominants
modes can effectively be controlled by the ICS regardless of any external energy
sources. The ICS is also more robust in restricting the inter-story drift ratio as
compared with TMDs. It sufficiently mitigates the RMS and peak displacement on
the top floor of the Benchmark building. Thus, the ICS has a better performance than
the TMDs and the CFs placement in terms of response reductions. According to the
performance evaluation criteria, there are substantial reductions for both the tuning
case and the detuning case. For both cases, the performance indexes are overall less
than the bare Benchmark building and its respective application with the TMDs.
Thesis (Ph.D.) - Florida Institute of Technology, 2019
Wed, 01 May 2019 00:00:00 GMThttp://hdl.handle.net/11141/28002019-05-01T00:00:00ZDevelopment of a Small Diameter Rapid Pressuremeter Test for Unbound Pavement Layer Evaluations
http://hdl.handle.net/11141/2546
Development of a Small Diameter Rapid Pressuremeter Test for Unbound Pavement Layer Evaluations
Misilo, Thaddeus Joseph
Six inch and 12 inch long small diameter pressuremeter (SDPMT) probes were
developed and tested extensively to determine in situ stress-strain properties of
unbound pavement layers. They were deployed in the same hole as the one made
during nuclear density testing. The stress strain response of conventional or
incremental and continuous or rapid strain controlled tests in unbound pavement
layers were evaluated at four sites in Brevard County, Florida. One test site was
A-1-b base material, one was A-3 subgrade and two were compacted A-3 fill of
subgrade quality. The SDPMT results were compared with nuclear density dry
and wet soil unit weights, moduli from the Dynatest and Zorn light weight
deflectometers, impact values from Clegg impact tests, and index values from
dynamic cone penetrometer (DCP) tests.
The SDPMT tests were conducted by driving or drilling a hole 7 or 13 inches
deep, then inserting a 5/8 inch diameter 6 or 12 inch long expandable cylindrical
probe. Once the probe was in place, it was inflated with water as the soil
stress-strain response was monitored and digitally recorded. Five statical models
were used to evaluate the relationships between the pressuremeter parameters
and the results from the other testing. The models produced linear, logarithmic
and exponential correlations.
The testing phase of this research included 159 SDPMT tests. Additionally
107 nuclear density tests, 141 lightweight deflectometer tests, 96 Clegg impact
hammer tests, 36 dynamic cone penetrometer tests were preformed to evaluate
their relationship with the pressuremeter parameters.
SDPMT engineering parameters were consistent and comparable to published
results. The rapid SDPMT testing was completed in less than 3-minutes and
provided reliable data, indicating that with further refinements, these test could
compliment and possible replace nuclear density testing.
It was observed that the resulting stress – strain response from the
incremental and continuous 6 and 12 inch tests resembles those of the standard
PENCEL pressuremeter curve. This study also demonstrates that SDPMT data
match common pressuremeter parameters for sands. This study showed p0 is the
least useful engineering parameter of those generated during testing.
The E0/pL ratio for incremental SDPMT tests ranged from 7.0 to 16.1 for
both 6 and 12 inch probes, which is similar to Briaud’s (2005) published
relationship of 7 to 12 range. Therefore the test quality is acceptable. E0/pL for
rapid or continuous SDPMT tests ranged from 10.9 to 33.2 for the 6 and 12 inch
probes. Although this is higher than the upper range of 12, the higher values are
attributed to the increased strain-rate applied to the soil during rapid testing.
Data collected during this research shows an excellent correlation between
densities and both E0 and pL for all test configurations. The excellent E0 verses
pL R2 from the 12 inch incremental testing suggested that engineers may be able
to predict pL from E0. Logarithmic relationships were observed between densities
and E0 and pL.
Good to excellent stiffness correlations exist between SDPMT moduli and
LWD moduli as nearly all R2-values exceeded 0.70. SDPMT E0 - LWD moduli
correlations were higher for the Dynatest LWD than the Zorn LWD.
An excellent logarithmic correlation between the Clegg impact value and the
SDPMT initial moduli for both 6 inch and 12 inch incremental tests.
No correlation was found between the DCP Index and the SDPMT p0, pL or
E0.
Thesis (Ph.D.) - Florida Institute of Technology, 2018
Sun, 01 Jul 2018 00:00:00 GMThttp://hdl.handle.net/11141/25462018-07-01T00:00:00Z