Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Behaviour of Fibre Composite Walkways and Grating

Download as pdf or txt
Download as pdf or txt
You are on page 1of 102

University of Southern Queensland

FACULTY OF ENGINEERING AND SURVEYING

Behaviour of fibre composite walkways and grating

A dissertation submitted by

Mr. Lachlan Keith Nicol

In fulfilment of the requirements of

Bachelor of Engineering (Civil)


Abstract
Throughout modern engineering, there has been a push to research, develop and
implement new and innovative building materials as a substitution for the materials
currently being used which have showed various forms of deterioration and faulting.
This project specifically focuses on the field of construction of boardwalks and
walkways.

The aim of this research project was to investigate how fibre reinforced polymer (FRP)
grating behaves mechanically while being subject to various types of static loading. A
numerical simulation using 3D modelling software was also undertaken to compare
simulation results with results found during the physical testing. A series of full scale
and sample sized tests were undertaken to help in determining the mechanical properties
and behaviour of the FRP grating. The full scale testing included static line loading, and
two different concentrated loadings; central to the panel as well as off centre loading. 3
different sample sized tests were also undertaken to help gain an understanding of the
material properties. These sample sized tests included, flexural, compressive, as well as a
burn out test.

The failure of the full scale line loading test was observed as major cracking and slight
delamination of the grating directly underneath the loading bar. The two concentrated
loading cases showed very similar failure modes to each other which consisted of the
loading block sinking into the grid immediately surrounding the loading area while the
rest of the panel remained intact. The line loading cases reached a maximum of
approximately 56.16kN of force which resulted in 64.85mm of deflection. The centred
concentrated load was tested to 33.37kN for a maximum deflection of 49.14mm,
whereas the off centre loading reached 57.24mm for a maximum load of 34.93kN.

As part of the sample sized testing system, a burn out test was undertaken to estimate
firstly the density of the provided FRP grating as well as the glass to resin ratio. The
density of this material was calculated to be 1544kg/m 3 with a glass to resin ratio of 54%
glass fibre to 46% resin. As part of the sample sized tests, other material properties were
determined including a flexural modulus of 9.89GPa and a compressive strength of
approximately 69.84GPa. These results aided in assigning a material property to a model
as part of the FE analysis using the software PTC Creo for simulation. The results from
the various FEA simulations gave very closely comparable results to those in the
physical testing.

2
University of Southern Queensland

Faculty of Engineering and Surveying

ENG4111 Research Project Part 1 &

ENG4112 Research Project Part 2

Limitations of Use

The Council of the University of Southern Queensland, its Faculty of Health, Engineering &
Sciences, and the staff of the University of Southern Queensland, do not accept any
responsibility for the truth, accuracy or completeness of material contained within or
associated with this dissertation.

Persons using all or any part of this material do so at their own risk, and not at the risk of the
Council of the University of Southern Queensland, its Faculty of Health, Engineering &
Sciences or the staff of the University of Southern Queensland.

This dissertation reports an educational exercise and has no purpose or validity beyond this
exercise. The sole purpose of the course pair entitled “Research Project” is to contribute to
the overall education within the student’s chosen degree program. This document, the
associated hardware, software, drawings, and other material set out in the associated
appendices should not be used for any other purpose: if they are so used, it is entirely at the
risk of the user.

3
Certification

I certify that the ideas, designs and experimental work, results, analyses and conclusions set
out in this dissertation are entirely my own effort, except where otherwise indicated and
acknowledged.

I further certify that the work is original and has not been previously submitted for
assessment in any other course or institution, except where specifically stated.

Lachlan Keith Nicol

Student Number: 005 0101 493

___________________________

Signature

4
Acknowledgments
I wish to thank the following people:

I would firstly like to thank my family and friends for their support and patience through this
project and my time at USQ.

Dr. Allan Manalo for helping to supervise and guide me through this research dissertation.

Buchanan’s Advanced Composites (BAC) and Nepean Engineering & Innovation for the
supply of test materials and equipment.

I would also like to thank the Centre of Excellence in Engineered Fibre Composites at USQ
for the use of their testing facilities as well as the helpful staff who assisted with the testing
procedures.

5
Table of Contents
Abstract ................................................................................................................................................... 2
Limitations of Use ................................................................................................................................... 3
Certification............................................................................................................................................. 4
Acknowledgments................................................................................................................................... 5
Table of Contents .................................................................................................................................... 6
List of Figures .......................................................................................................................................... 8
List of Tables ......................................................................................................................................... 10
Nomenclature ....................................................................................................................................... 11
1. Introduction .................................................................................................................................. 12
1.1 Project topic ................................................................................................................................ 12
1.2 Project background ..................................................................................................................... 12
1.3 Project aim and objectives .......................................................................................................... 14
1.4 Justification of project ................................................................................................................ 15
1.5 Scope ........................................................................................................................................... 16
1.6 Resource requirements............................................................................................................... 17
2. Literature Review .......................................................................................................................... 18
2.1 Introduction ................................................................................................................................ 18
2.2 Fibre composites ......................................................................................................................... 19
2.3 Resin Matrices ............................................................................................................................. 20
2.4 Fibres ........................................................................................................................................... 22
2.5 Manufacturing of FRP Grid ......................................................................................................... 24
2.5.1 Introduction ......................................................................................................................... 24
2.5.2 Moulded FRP grating............................................................................................................ 25
2.5.3 Pultruded FRP grating .......................................................................................................... 26
2.6 Current uses of moulded FRP grating ......................................................................................... 27
2.7 Testing ......................................................................................................................................... 28
2.7.1 Mechanical testing ............................................................................................................... 28
2.7.2 Testing not included in this project ..................................................................................... 36
2.8 Risk Management ....................................................................................................................... 37
2.8.1 Introduction ......................................................................................................................... 37
2.8.2 Risk Identification................................................................................................................. 37
2.8.3 Risk Evaluation ..................................................................................................................... 38
2.8.4 Risk Control .......................................................................................................................... 39
3. Methodology................................................................................................................................. 41

6
3.1 Introduction ................................................................................................................................ 41
3.2 Preparation of test specimens .................................................................................................... 41
3.2.1 Full Scale Tests ..................................................................................................................... 41
3.2.2 Sample Sized Tests ............................................................................................................... 42
3.3 Testing Procedures...................................................................................................................... 46
3.3.1 Full Scale Tests ..................................................................................................................... 46
3.3.2 Sample Sized Tests ............................................................................................................... 50
3.3.3 Finite Element Analysis ........................................................................................................ 55
4. Results and Analysis ...................................................................................................................... 60
4.1 Introduction ................................................................................................................................ 60
4.2 Full Scale Testing Analysis ........................................................................................................... 60
4.2.1 Line Loading ......................................................................................................................... 60
4.2.2 Centred Concentrated Load ................................................................................................. 62
4.2.3 Off Centre Concentrated Load ............................................................................................. 64
4.3 Sample Sized Testing Analysis ..................................................................................................... 66
4.3.1 Flexural Loading ................................................................................................................... 66
4.3.2 Compressive Loading ........................................................................................................... 70
4.3.3 Burn Out Test ....................................................................................................................... 75
4.3.4 Finite Element Analysis ........................................................................................................ 77
5. Conclusions & Future Work .......................................................................................................... 83
5.1 Introduction ................................................................................................................................ 83
5.2 Failure Modes in Testing ............................................................................................................. 83
5.3 Loading Case Comparison ........................................................................................................... 84
5.4 Use of FEA ................................................................................................................................... 84
5.5 Summary ..................................................................................................................................... 85
5.6 Recommendations & Future Work ............................................................................................. 87
References ............................................................................................................................................ 88
Appendices............................................................................................................................................ 90
Appendix A – Test Results ................................................................................................................. 90
Appendix B – Project Offer (Allan Manalo) ..................................................................................... 100
Appendix C – Project Specification ................................................................................................. 101

7
List of Figures

Figure 1 – Deteriorated concrete .......................................................................................................................... 13


Figure 2 – Deteriorated steel ................................................................................................................................ 13
Figure 3 – Deteriorated wood/timber ................................................................................................................... 14
Figure 4 - Percentage value of materials in fibre reinforced composite market ................................................... 23
Figure 5 - Moulded FRP grid manufacturing process ............................................................................................ 26
Figure 6 - Pultruuded FRP manufacturing process................................................................................................ 27
Figure 7 - Flexural test specimens ......................................................................................................................... 43
Figure 8 - Compressive test specimens ................................................................................................................. 44
Figure 9 - Burn out test specimens........................................................................................................................ 45
Figure 10 - Arrangement of strain gauges ............................................................................................................ 47
Figure 11 - Line load testing setup ........................................................................................................................ 48
Figure 12 - Concentrated load block dimensions) ................................................................................................. 48
Figure 13 - Centred & off centred loading ............................................................................................................ 49
Figure 14 - Flexural sample during loading........................................................................................................... 50
Figure 15 - Compression testing for a 1x1 grid sample......................................................................................... 51
Figure 16 - Computer used to record compressive testing data ........................................................................... 52
Figure 17 - Weighing of the crucible and the grid sample ................................................................................... 53
Figure 18 - Thermolyne industrial benchtop furnace ............................................................................................ 54
Figure 19 - Flexure test 3D model ......................................................................................................................... 56
Figure 20 - Flexure 3D model ready for analysis ................................................................................................... 57
Figure 21 - Full scale line load ready for analysis.................................................................................................. 59
Figure 22 - Deflection results for line loading ....................................................................................................... 60
Figure 23 - Failure of line loading test panel ........................................................................................................ 61
Figure 24 - Deflection results for centred concentrated loading .......................................................................... 62
Figure 25 - Failure of centred concentrated load test panel ................................................................................. 63
Figure 26 - Deflection results for off centre concentrated load ............................................................................ 64
Figure 27 - Flexural deflection results at load points ............................................................................................ 66
Figure 28 - Example of shear failure for flexural loading ...................................................................................... 67
Figure 29 - 1x1 compressive loading results ......................................................................................................... 70
Figure 30 - Example of 'crushing' compressive failure for 1x1 block .................................................................... 71
Figure 31 - 2x2 compressive loading results ......................................................................................................... 72
Figure 32 - 3x3 compressive loading results ......................................................................................................... 73
Figure 33 - Burnt out FRP web containing 'filler material' and glass fibres .......................................................... 76
Figure 34 - Load-deflection simulation at 10% deformation of flexural model .................................................... 77

8
Figure 35 - Stress distribution for flexure analysis ................................................................................................ 78
Figure 36 - Strain distribution for flexure analysis ................................................................................................ 79
Figure 37 – Load-deflection simulation at 10% deformation for full scale line load............................................. 80
Figure 38 - Load-deflection simulation at 10% deformation for full scale centred concentrated load ................. 81
Figure 39 - Load-deflection simulation at 10% deformation for full off centre concentrated load ...................... 82

9
List of Tables
Table 1 - Typical characteristics of common unreinforced thermosetting resins used in composite materials.... 21
Table 2 - Full scale load testing specs. .................................................................................................................. 42
Table 3 - Line load test results .............................................................................................................................. 61
Table 4 – Centred concentrated load test results ................................................................................................. 63
Table 5 - Off centre concentrated load test results .............................................................................................. 65
Table 6 - Flexure test results summary ................................................................................................................. 68
Table 7 - Compressive loading test results ............................................................................................................ 74
Table 8 - Burn out test results ............................................................................................................................... 75
Table 9 - FRP grating mechanical properties ........................................................................................................ 85
Table 10 - Full scale summary ............................................................................................................................... 86

10
Nomenclature

AS –Australian Standard
C1x1 – Compressive loading for a 1 by 1 block
C2x2 – Compressive loading for a 2 by 2 block
C3x3 – Compressive loading for a 3 by 3 block
CEEFC – Centre of Excellence in Engineered Fibre Composites
CL – Centred concentrated loading
FEA – Finite Element Analysis
FL – Flexural loading
FRP – Fibre Reinforced Polymer
ISO – International Organization for Standardization
LL – Line loading
NZS – New Zealand Standard
OC – Off centred loading
PPE – Personal Protective Equipment
USQ – University of Southern Queensland

11
1. Introduction
This chapter will provide an outline of the general aims and objectives associated with the
completion of the project. The overall purpose of this research dissertation is to analyse the
behaviours of fibre composite grating under different loading cases to gain an understanding
on how this particular type of grating performs structurally to help promote its use throughout
the engineering and construction industries, primarily as a form of walkway.

1.1 Project topic


‘Behaviour of fibre composite walkways and grating’

1.2 Project background


For a very long time within the engineering and construction industries, there has been a
somewhat non varying selection of materials used in the construction of boardwalks and
walkways. The industry standard materials most commonly used today include concrete,
steel, and wood.

These traditional materials have posed various problems due to dilapidation and deterioration
over time as well as an ever increasing influence due to costing within the materials selection
process. Given the flaws of these modern engineering materials, there has been a call recently
for a more sustainable product which will assist in increasing the longevity and lifespan of
boardwalks throughout Australia and the world.

Fibre reinforced polymer (FRP) grating demonstrates numerous advantages over customary
materials for a large range of scenarios, particularly within harsh and corrosive environments
where traditional materials have proven to suffer.

Within the civil engineering industry today, the popularity of fibre composite walkways and
boardwalks is slowly gaining momentum. However, there is not enough information or
knowledge on the products available and their mechanical capabilities to be able to use it
more commonly as an essential material.

12
Fibre reinforced polymer composites or advanced composite materials are very attractive for
use in civil engineering applications due to their high strength-to-weight and stiffness-to-
weight ratios, corrosion resistance, light weight and potentially high durability (Lelli Van Den
Einde et al. 2003).

Figure 1 – Deteriorated concrete (A. Davies, 2003)

Figure 2 – Deteriorated steel (A. Davies, 2003)

13
Figure 3 – Deteriorated wood/timber (J. Hoath, 2006)

1.3 Project aim and objectives

To be able to reach the overall goal of analysing the mechanical behaviours of FRP grating
under various loading cases, various objectives need to be worked towards and completed. A
thorough investigation into fibre reinforced polymer grating needs to be conducted including
the analysis of various load cases including full scale static loading as well as sample sized
tests including flexural, compressive and burn out. Not only will FRP grid be physically
tested and recorded, but a computerised finite element analysis will also be conducted to help
verify and simulate results. Within this project, a great deal of research needs to be conducted
in order to gain a greater understanding of not only fibre composites in general but also
specific research into the developing knowledge surrounding FRP grating and walkways in
particular.

14
The research objectives to aid in the analysis of FRP grating as a form of walkway material
are itemized below:

 Complete an extensive literature review in regards to FRP grating as well as testing


and analysing of materials
 Develop an in depth test schedule and programme for the testing procedures to fully
establish the aims and objectives that are needing to be accomplished over the course
of this research project.
 Perform static load testing on a number of equal sized decking panels under line and
point loading.
 Perform sample sized testing including flexural, compressive, and burn out fibre tests.
 Analyse and compare results between testing to gain a better understanding of the
different physical behaviours/properties of FRP grating.
 Perform a Finite Element Analysis on the grating to help map out the stresses and
strains associated with the separate loading cases.
 Write and submit a final academic dissertation on the research and results from the
entirety of this project.

1.4 Justification of project


The driving factor for conducting this research project is the need for a more sustainable
material for walkways and boardwalks to act as a permanent substitute to the traditional
materials used today. Fibre composites can possibly offer an efficient alternative while
increasing the life span and minimising the deterioration due to varying corrosion
circumstances.

The use of fibre composite grating within Australia is relatively insignificant with minimal
knowledge regarding its physical and mechanical properties. There are some circumstances
where this form of grating has become very popular for example the grating can most
commonly be found as flooring in highly corrosive factories or plants such as desalination
plants or pumping stations. This form of grating was developed when simply no other
alternative would suffice for use over long period of time.

15
This project aims to help promote the use of FRP grating throughout engineering in Australia
by exploring and reviewing previous studies as well as conducting a number of related tests
to help increase the knowledge available for use of the product. The lack of specific
Australian standards for this particular type of product is also an influence as to why this
exact project was taken on.

1.5 Scope
The testing and analysis conducted for the entirety of this project is focused on the moulded
FRP grating supplied by Nepean engineering and innovation with regards to the mechanical
testing procedures: line load, centred/non centred concentrated load, sample compressive and
flexural tests, and a burn out fibre analysis. A finite element analysis was also conducted to
correlate physical results with an analytical sequence of simulations using the program Creo.
“Creo is a scalable, interoperable suite of product design software that delivers fast time to
value” (P.T.C Creo).

Since the samples tested were produced by external and unknown sources the raw materials,
exact manufacturing process, and curing procedures are unidentified and certain factors either
need to be assumed or investigated. The fibre burn out analysis aids in determining one
unknown; which fibre was used for the structural makeup of the grid being tested.

Throughout the mechanical testing programme there are certain limitations which were
accepted and noted to be considered in the analysis of results. One limitation was that the
load cell available at the Centre of Excellence in Engineered Fibre Composites was limited to
a serviceable 220kN force which needed to be considered if the question of what would
happen at higher loading was asked. Impact testing was not considered for this project,
however it would be a test which could provide you with a set of results to accompany the
outcomes found in this particular project.

During the review of literature, due to the lack of research and comparative testing with
regards to FRP grating specifically, similar products such as honeycomb cores and FRP flat
panels have been used. It was deemed acceptable to use such products for research as there
are numerous similarities between the materials used in production as well as their

16
manufacturing processes and finally the similarities in the applications within the
construction industry are comparable.

1.6 Resource requirements


Throughout this project, there are certain resources which are to be sourced and acquired to
be used for the testing programme with an available budget determined. The test specimens
themselves have been provided by Nepean engineering and innovation with a total of 14
panels supplied at the beginning of the project for testing purposes. These 14 test panels have
been supplied in a range of lengths and widths which will need to be considered for the
analysis of results. For analysis of stress and strain, strain gauges are attached to the test
specimens, both full scale as well as sample. To help reduce costing, the sample sized test
pieces were shaped from either undamaged sections of the test panels or the unused panels.
The strain gauges and concentrated load block used for testing have been generously supplied
by the second sponsor, Buchannan’s advanced composites. In regards to testing equipment,
since this project is of interest to the Centre of Excellence in Engineered Fibre Composites at
the University of Southern Queensland, all testing was conducted in the CEEFC’s test
laboratories using their supplied equipment. Given the potential costing for all of these
discussed materials and test equipment this project could have been very expensive, however
with the generous sponsorships provided there will be minimal out of pocket costs.

17
2. Literature Review

2.1 Introduction

This literature review will focus on the analysis and discussion of previously conducted
studies as well as currently published literature relevant to fibre composites and more
specifically Fibre Reinforced Polymer (FRP) grating. Information on the topic of FRP grating
has also been obtained from non-published sources in addition to the correspondence and
guidance from people such as supervisors and colleagues who have relevant experience in the
industry or familiarity with the subject matter. A brief historical background of the use of
fibre composites within the engineering industry will be provided with detail in regards to
polymers and fillers used as part of modern day fibre composites. The production process of
FRP grating will be included and will cover the various procedures involved in the
manufacturing of the different forms of FRP grid.

Gibson et. al. (2013) explain how composite gratings have been employed for lightweight,
corrosion-resistant grid-flooring on offshore platforms, ships, floating production systems,
drilling rigs and, elsewhere, for over 20 years. The selection of materials for use offshore is
commonly ‘performance-based’, so the case for choosing composites is generally made by
comparing key properties such as density and corrosion resistance to those of steel.

This leads to a suggestion that there is a substantial amount of recognized research with
respects to both the investigation of FRP grating for off-shore platform use as well as the
determination of density and corrosion resistance in comparison to traditional materials such
as steel or concrete. The need in the past for research development into these areas was a
necessity and has already been conducted, however as fibre composites are being looked at as
a more suitable alternative material within a growing number of industries, the need for
broader research and investigation is apparent. Particular attention needs to be paid to the use
of FRP grid in on land applications rather than simply for offshore use. This leaves a gap in
research which this project aims to help minimise by investigating numerous mechanical
properties and behaviours for not only off-shore use but also the use on land in boardwalks
and walkways.

18
2.2 Fibre composites

Lubin (1969) gave a common definition for a modern composite material as: A composite is
a system that is created by the synthetic assembly of two or more materials; namely, a
selected filler or reinforcing element and a compatible resin binder to obtain specific
characteristics and properties. More specifically, Fibre Reinforced Polymer (FRP) suggests
the use of a thermoset polymer matrix containing fibres, either synthetic such as glass or
natural such as bamboo fibre strands. Cooperation and synergy of the combined constituents
typically ensures a composite with enhanced material properties as the matrix and fibre work
together to form a combination of properties which are superior in comparison to the
materials in unaccompanied states.

While the composite industry is thought to be an emerging industry in modern engineering,


the concept of composites has in fact been thought to date back as long ago as the ancient
Egyptian and Mesopotamian settlers who used a mixture of straw and mud to construct
housing which proved to be the most durable and strongest buildings at the time (Moorey,
1994). This combination of straw and mud fits perfectly into the description of what
constitutes a composite material; mud, a natural matrix component being reinforced by the
straw fibre to create a material with structural properties surpassing that of the materials when
individually used.

Fibre Reinforced polymer as we know today was not developed until the 1930’s where it was
extensively researched with minimal commercial use and was not commercially used on a
mass level until the late 1950’s for use in the aerospace and motor vehicle industries. The
production and application of FRP grating is a relatively new material within the engineering
field, especially within Australia where very little is known about its property performance
and mechanical abilities. It is gaining popularity throughout Australia and the world for its
use in corrosive environments as well as the electrical industry for its non-conductive nature.
In Many applications, fibre reinforced polymers provide superior performance to other
materials of construction (Stevens, 2012).

The strength of FRP walkways is often dependent on the mechanical properties displayed in
both the fibres themselves as well as the matrix, the concentrations relevant to one another, as
well as the fibre length, orientation and direction set within the resin matrix.

19
2.3 Resin Matrices

Resin systems such as epoxies and polyesters have limited use for the manufacture of
structures on their own, since their mechanical properties are not very high when compared to
most metals (Cripps, unknown date). However, they are very desirable within engineering for
their ability to be easily moulded into complex shapes and forms which other standard
materials struggle to do. The functions of the polymer matrix are to transfer load, secure the
fibre reinforcement and to prevent any mechanical or environmental damage to the fibres
(Manthey, 2009). The polymers used as matrix materials can be divided into two primary
families, the natures of which depend on their molecular structures. These are thermosetting
and thermoplastic resins, for example, respectively, epoxy resin and polyamide (Nylon)
(Bunsell & Renard, 2005). Resins commonly used in FRP grid systems are, Isophthalic
polyester, vinyl ester, phenolic. The use of different types of resin can be influenced by the
need for different mechanical abilities required in the use of the grid system, this can include
increased fire resistance, different types of chemical resistance, as well as resins that can
perform in a more brittle or flexible state. Bunsell & Renard display a comparison between
various physical property performances of different resins as well as how resistant to
chemical attack they are in table 1.

20
Table 1 - Typical characteristics of common unreinforced thermosetting resins used in composite materials. (Bunsell & Renard, 2005)

Resin Physical Properties Chemical Resistance Properties

Flexural Tensile Tensile Failure Heat Deflection Normal Maximum Water Solvent Acid Alkali
Strength Strength Modulus Strain (%) Temperature (°C) Temperature Limit
(MPa) (MPa) (GPa) (°C)

Unsaturted polyester

Ortho-phthalate 100-135 50-75 3.2-4.0 1.2-4.0 55-100 80-100 fair poor fair poor

Iso-phthalate 110-140 55-90 3.0-4.0 0.8-2.8 100-130 100-130 good fair good poor/fair

Modified bisphenol type 125-135 65-75 3.2-3.8 0.9-2.6 130-180 130-180 very good fair good fair/good

Epoxy (bishpenol)

Aliphtatic polyamde cure 85-125 50-70 3.5 1.0-3.5 60-90 100 good fair/good fair/good fair/good

Boron trifluoride complex 110 85 3.0-4.0 1.0-2.5 120-190 90-150 good fair/good good good

Aromatic amine cure 80-130 60-75 3.0-3.5 1.5-3.5 85-170 120-180 excellent good fair/good good

Aromatic anhydride cure 90-130 80-105 2.65-3.5 2.0-2.5 130-200 150-220 poor/fair poor/fair good poor

Vinyl ester

Polyimide 110-130 70-85 3.3 1.0-4.0 90-125 90-125 good fair/good good good

Friedel-Crafts 75-130 50-120 3.1-4.7 2.0-3.5 250-360 250-360 low - - low

Phenolic furane 110-120 95-110 4.1 1.5-3.0 160-240 150-300 excellent good good fair/good

Silicone 100-120 60-75 2.5-3.5 0.5-1.0 180-220 250-300 good excellent good poor

21
2.4 Fibres

Composite materials owe their remarkable characteristics to the fibres which are used to
reinforce the matrix (Bunsell & Renard, 2005). The fibres used in composite technology
serve the purpose of providing reinforcement and strength for a material when combined with
a matrix, such as resin. The fibres act as the ‘bone structure’ in composites and are the source
of the strength generated when cohesion between the fibres and matrix phase is achieved. The
use of fibres also helps to reduce the overall cost of fibre composites as they are generally a
much cheaper commodity than the polymer matrix. The use of different types of fibres has
become very popular in modern composite development as different mechanical properties
can be achieved by using different sources as a fibre content.

Fibres can be categorized into two different forms, both synthetic and natural. As the name
would suggest, synthetic fibres are artificially developed materials which do not occur
naturally and are the most commonly used fibre for FRP gratings and walkways. The
synthetic fibre group being used within the composite industry have traditionally been petro-
chemical based materials. Traditional synthetic fibre reinforcements that are commonly used
in engineering applications can be categorised into three main groups; glass, carbon and
aramid fibres (Manthey, 2013). Natural fibres used in modern FRP design are most
commonly derived from plant sources but can also be sourced from animal by-products and
in some cases from minerals. Natural fibres are beginning to be further looked into for their
use ahead of traditional petroleum based fibres for their ability to be extracted from
renewable resources. With the diminishing supply of petroleum, a sustainable alternative is
becoming increasingly more desirable. Natural fibres that are gaining increasing popularity
within the engineered FRP industry include: hemp, flax, cotton, asbestos, and wool amongst
others.

22
Carbon fibres Aramid fibres
Polypropylene 4% 1%
5% Unsaturated polyester
Other thermoplastics
28%
7%

Polyamide
9%

Other thermosets
10%

Epoxy Glass fibres


10% 26%

Figure 4 - Percentage value of materials in fibre reinforced composite market (Bunsell & Renard 2005)

Fibres are commercially available in various forms suitable for different applications
(Agarwal et. al., 2006) including chopped strand mat, woven cloth, and woven roving.
Woven cloth is similar to common fabric as it is woven tightly to create a cloth type texture
with various weaving techniques used to provide different strength results. Chopped strand
mat consists of randomly oriented short strands of fibre meshed together to form a mat of
fibres. While chopped strand mat does not tend to hold as much strength as woven cloth, it is
generally used for tight curves and corners as it is easier to conform than the woven cloths.
Woven roving is the fibrous form used when producing FRP gratings given the
impracticability of using cloth or mat for the specific grid shape. Fibres are bundled and yarns
are orientated into two directions (Fibreglass Warehouse, 2013).

23
2.5 Manufacturing of FRP Grid
2.5.1 Introduction
In order to determine the requirement and need for fibre reinforced polymer grid it must first
be understood why grids and gratings are used throughout various industries. Grid or grating
structures are most commonly used for drainage systems and walkways frequently exposed to
high volumes of water. In recent history, FRP grating has excelled in its application for
corrosive, water prominent environments such as the marine industry, food preparation
shops/factories, chemical plants, production factories, and most commonly in recent times
throughout the mining industry of Australia which generally hosts a very corrosive
environment. Grating can also be used for smaller scale applications other than walkways,
including spillway covers, inspection hatches, and stairway treads. Not only is grid type
flooring used for drainage, but it also assists in the significant reduction of cost by providing
structural integrity over a substantial area while using minimal material to cover said area. In
turn, this reduction in material can often result in a reduction in weight which can be very
favourable in structures not designed to take very large loads, i.e. pedestrian walkways or
platforms.

Traditionally steel has been the most common material used for the construction of grid
walkways and decks and has served the construction industry well for a very long time.
However, problems with the corrosion and degeneration of steel grids has gained the
attention of many industrial leaders and governments for the possible replacement and
solution to the issue regarding corrosive and wet environments. The engineering group at
Kerr McGee Coal Company undertook an investigation into the fast deterioration of steel
walkways titled ‘Alternative to Metal Walkways and Handrails’. Within this investigation it
was found that in corrosive environments such as in the coal mine which was tested, steel
undergoes significant deterioration within two years and stainless steel within six years
(Strongwell, 2013). The result of the investigation gave results leading to the need to use FRP
grating in such a circumstance given its remarkable corrosion resistance. FRP grating poses a
potential and very tangible substitute for the traditional use of metal grid with its record of
being able to resist deterioration when challenged by corrosive environments and situations.

Glass Fibre Reinforced Polymer (GFRP) gratings are normally manufactured by moulding
and pultrusion process. It may consist of various combinations of glass fibre and resin (Habib
& Akbar, 2014). Both manufacturing processes greatly differ from each other, however they

24
can result in relatively similar products in terms of strength and possible uses. The different
manufacturing processes will be explained in detail in the corresponding sections but as a
general practice, moulded FRP grating is a process where fibres and the resin matrix is
combined together by hand in a set mould and left to cure to form a grid structure, whereas
pultruded grating has its fibres pulled through a machine where they are then infused with the
resin matrix and cooled and cut to the required sizing.

2.5.2 Moulded FRP grating


Moulded grating is produced in a complex open mould and the entire sheet is monolithically
moulded in one process (Meng & Lo, 2000). As the manufacturing process classification
would imply, moulded FRP grating is manufactured using moulds which form the ‘base’ or
profile of the grating to suit the desired dimensions. This moulded manufacturing process
involves the systematic laying of a fibre series, or commonly a roll of continuous fibreglass
roving into an open, heated mould in an alternating progression pattern. The moulding
process is constrained to the bounds of only being able to create FRP panels to a set size
limit, restricted by the mould parameters. Generally, the alternating layers are placed
longitudinally over the whole mould and then transversely to create an alternate cross pattern
which encourages enhanced strength and structural integrity over the lifespan of the grating.

The layered fibres are thoroughly wetted out with the desired resin matrix to resolutely bond
the fibrous material in place as well as form the fundamental shape of the grid as desired.
During curing, the mould is subjected to reducing heat to help restrict shrinking throughout
the curing process. Upon curing, the grating is cast out from the mould where post curing
applications can then be applied such as a grit coating for additional slip resistance. Since
moulded gratings are cast as one piece panels, there is an optimum distribution of load over
the whole span with no weak points caused by joints which can be found while analysing the
pultruded manufacturing process.

25
Figure 5 - Moulded FRP grid manufacturing process (ZarnitsaTeam, 2011)

2.5.3 Pultruded FRP grating


Pultrusion is a process of FRP manufacturing whereby fibreglass roving is drawn through a
resin wet bath to fully immerse the fibreglass in the resin bonding matrix and is then drawn
through a series of dies which form the desired shape of the finished product and helps in
forming the structural make-up of the pultrusion profile. Once the pulled roving/matrix
combination forms the general profile required, it is then pulled through a final set of forming
dies which are heated to cure the matrix and fibre combination. The process of pultrusion is
constant and required lengths are cut as the final part in the manufacturing chain. This
continuous pultrusion operation can be readily automated and allows for low labour
involvement (Compositebuild.com, 2014).

Since a profile such as grating is not easily manufactured using pultrusion method, the most
common practice is to manufacture a series of thin pultruded FRP bars and glue on joints (or
commonly cross-rods) to create a large grid superstructure panel. These pultruded bars are
most commonly manufactured as ‘T’ bars or ‘I’ bars to maximise the strength to weight ratio.
Unlike the singularly formed moulded grating, pultrusion can often fail due to unsatisfactory

26
cohesion between grid lengths and the glued joints. Pultruded FRP grids can most commonly
be located with various industry plants which are subject to high moisture or are located in
highly corrosive environments being used as cover for ‘flow away’ drains for any moisture
which might be found on the factory floor. The grid itself allows liquid to fall beneath the
floor level into the drains while still allowing workers to walk freely and safely on top of the
grid as a safe walkway area.

Figure 6 - Pultruuded FRP manufacturing process (Creative Pultrusions, 2014)

2.6 Current uses of moulded FRP grating


In 2006 the Queensland Government issued the “Fibre composites action plan – new
technology taking shape” initiative which was aimed at promoting the use of fibre composites
over a varied range of fields throughout Queensland. As part of this initiative, the
Government issued informational handbooks which included the varied uses and benefits
fibre composites could offer. In their handbook titled ‘Composites in industrial plants’, a
large number of uses for FRP grating was suggested by the Government for use in large
industrial plants or factories.

27
2.7 Testing
Testing is a vital element in analysing the properties of different types of fibre composite
walkways in order to determine the optimum grating system for particular situations,
requirements and environments. Since minimal testing has been recorded or reported on
publically with respects to moulded FRP grating specifically, the literature used for this
particular section was often related but not confined to the testing performed on what is
known as honeycomb sandwich panels. These sandwich panels are fairly similar to FRP
grating in their uses and can be formed to be of similar profiles and shapes to the panels
which are tested as part of this project. The literature that is available discusses various
testing techniques to gain results to help determine the tensile, compressive, impact, and
flexural properties of a particular material.

2.7.1 Mechanical testing


While mechanical testing often covers tensile, compressive, shear, impact, and flexural tests,
shear and impact testing will be neglected from the testing programme as they lie outside of
the scope of this project. The mechanical testing which will in fact be conducted will include
the full scale static loading tests which form the line load test, centred concentrated load test,
and the off centre concentrated load test. Sample sized analysis will also be conducted and
the sample sized testing procedures which will be completed during this project include
flexural, burn out, and several compressive tests.

2.7.1.1 Full scale static loading tests


Line load test

Line load testing of FRP grating can be closely comparable to the three point bending tests of
sandwich panels conducted by Ugale et al. (2013). In Ugale et.al ‘s paper, a three point
bending test setup is described on a relatively small scale compared to our full scale panel
tests, however the basic principles can be applied on a larger scale as required. Further
guidance for test setups was found in the Australian standard AS 4155 – 1993 ‘Test methods
for general access floors’ with correct test setup, test procedure/methodology, and recording
procedures listed and discussed with respect to line loading.

28
Another test paper which was used as guidance for the line loading test was undertaken by
A.I.Habib and I.Akbar in their investigation into the ‘effect of accelerated ageing process on
the strength of moulded and pultruded GFRP gratings’ in 2014. The aim of Habib and
Akbar’s research was to try and determine the effect ageing of grating had on its mechanical
properties. The concept of the testing procedure and methodology of Habib and Akbar’s three
point full scale bend test was closely followed for the line loading tests.

The line load was chosen not only for the importance of the results which can be determined
but also the simplicity and ease of preparation and set up of the tests. The line loading test
system can provide results which are of great importance to the mechanical analysis of a
panel. Most notably the modulus of elasticity is determined by load-deflection measurements
at stresses below the proportional limit (E. Shapiro, 2000).

The modulus of elasticity in bending is used to define the stiffness of a material and aids to
determine the behaviour of said material while subject to loading by using the ration of stress
to strain in flexural deformation. This modulus is calculated by using the gradient of the load
vs deflection increments when load is applied to the midspan of the test piece following the
formula:

Modulus of Elasticity:

(1)

Where:

L = Length of supported span (mm)


b = Width of test specimen (mm)
d = Thickness of test specimen (mm)
ΔP = Load increment (N)
Δδ = Deflection increment at midspan (mm)

The elastic modulus is calculated in the form Pascals (Pa) or more commonly for ease of
representation can be presented as Gigapascals (GPa).

29
Concentrated Load Tests

Awad, Aravinthan & Zhuge (2011) conducted an experimental analysis of a newly developed
GFRP sandwich panel, where a fibre composite sandwich panel system was subject to a
number of testing methods and analytical simulations. One of the testing processes used for
analysis was the concentrated (or point) loading tests which displayed an important set of
results for the analysis of mechanical properties. Similar to the line loading cases, an
Australian standard was used as a guide for the testing setup and procedure as well as the
proper recording techniques employed for a successful testing programme. While AS 4155.8
– 1993 ‘Test methods for general access floors – test for 25mm x 25mm concentrated load’ is
not a standard generally associated with the testing for this particular project in reference to
FRP grating, in the absence of a more relevant standard, this was used as a general guide for
employing general procedures for this form of testing.

30
2.7.1.2 Sample sized tests
Flexural load test

A four point bend test was chosen for the flexural testing and helps to determine the peak
flexural stress (σf) of the grating material in a singular grid strip. As well as the peak flexural
strain, the flexural testing will also help to determine the peak flexural strain (εf) and the
flexural modulus (Ef).

Flexural Stress equation:

(2)

Where:

L1 = Length of supported span (mm)


L2 = Length between load bars (mm)
b = Width of test specimen (mm)
d = Thickness of test specimen (mm)
P = Maximum applied load (N)

Flexural stress (flexural strength) is calculated/measured in Pascals (N/m2) but is more


commonly reported in megapascals (N/mm2) for ease of recording. The peak flexural stress is
originally derived from the bending moment equation which relates to the load at failure.

31
Flexural Strain equation:

(3)

Where:

L = Length of supported span (mm)


d = Thickness of test specimen (mm)
D = Maximum deflection

The flexural strain as indicated by equation 3 is an indication of the deformation occurring at


the point of failure of the test specimen in relation to the applied load. Flexural strain is
directly related to the displacement change between pre and post loading. Strain will always
give a dimensionless result as it gives the ratio of change rather than a dimension of the
change itself.

Flexural Modulus equation:

(4)

Where:

L = Length of supported span (mm)


b = Width of test specimen (mm)
d = Thickness of test specimen (mm)
m = Gradient of load/deflection curve

Flexural modulus is the ratio of stress to strain in flexural deformation (i.e. the tendency for
the specimen to bend). It is primarily a factor of the gradient produced by the load deflection
curve produced during a flexural load test.

32
Since a 4 point bend test records the deflection at the point of applied loading instead of in
the centre, it is necessary to calculate the deflection in the centre to be able to compare results
from the FE analysis as well as gain an understanding of the maximum deflection for the
flexure test. Equation 5 uses the recorded deflection at the points of loading to calculate what
the approximate deflection at the centre of the specimen would be.

(5)

Where:

P = Maximum applied load (N)


L1 = Length of supported span (mm)
L2 = Length between load bars (mm)
E = Modulus of elasticity (GPa)
I = Moment of inertia (mm4)

33
Compressive load test

Compressive Stress formula:

(6)

Where:

P = Maximum applied load (N)


A = Total loaded area (mm2)

The compressive stress equation (equation 6) is a direct relation to its compressive strength
for the load at failure. The units used for compressive strength are N/m2 or more commonly
N/mm2 (megapascals (MPa)). Compressive stress is the stress of the test specimen in the
uniaxial direction i.e. the force is applied by ‘compressing’.

Burn out test

Glass Content formula:

(7)

Where:

m1 = Initial mass of crucible (g)


m2 = Initial mass of crucible plus specimen (g)
m3 = Final mass after calcination (g)

ISO 1172:1996 - ‘Textile-glass-reinforced plastics’ provides a guide on the proper


procedures of how to successfully conduct a fibre burn out test. Within this standard there is a
proper operating procedure, recording procedure, as well as the formula to determine the
fibre content of a particular specimen. This formula is stated here in equation 7 above. The

34
equation for Mglass gives a percentage of weight that the glass contributes to the overall
weight of the specimen. From this glass percentage, and indication of the glass to resin ratio
can be calculated. Since the result for Mglass does not have any formal units, only a
percentage, the percentage value can be used to determine the physical weight of glass by
applying the ratio of glass to the overall weight of a piece of grating.

Another material property which could be obtained during the burn out test is the density of
the FRP grid being tested. It is always important to be able to compare densities of products
and would be essential for final recommendations and analysis. Density (ρ) is simply
calculated by dividing the weight of a sample by its volume as shown below in equation 8.

Density Formula:

(8)

Where:

ms = Sample mass (g)


V = Volume of sample (mm3)

The results of this density calculation will give g/mm3 but for easy comparison between other
materials it will be converted to kg/m3.

35
2.7.2 Testing not included in this project
Due to the restraints on this project such as time, number of personnel and, resources, some
testing procedures have been excluded from the scope of this project but are essential in the
full determination of all grid properties and are very important in the recommendation of the
use of fibre composite grating as a viable alternative material for use in walkways. Other tests
which could be conducted as part of further analysis could include but not limited to;
corrosion resistance testing, impact tests, and fire resistance testing.

A cost analysis may also be an essential factor in determining the best possible material to
replace traditional materials with a more corrosion resistant alternative. In today’s market,
costing tends to be the driving factor ahead of any other property and needs to be considered
to fully determine the best alternative material in construction of walkways and boardwalks.

36
2.8 Risk Management
2.8.1 Introduction
Throughout this project there is a degree of risk which needs to be noted and control methods
need to be enforced to ensure no accidents or incidents occur during the course of this project
which could have been prevented. The aim of this risk management and assessment is to
determine, analyse, and document any risks which may be related to the works done
throughout this project. Risk management is generally a case of if the risks are not thought
about with a solution known to the operator then they will usually occur. However, if all risks
and hazards are properly analysed and the correct working procedures are followed then the
possibility of an incident occurring is significantly reduced.

2.8.2 Risk Identification


The major risks associated with this project are primarily found throughout the entirety of the
testing programme. The project risks can be characterised as sample sizing, housekeeping,
test setup, sample testing, and workplace restoration. The risks associated with sample sizing
are quite hazardous as the operator is subject to running electric saw blades, airborne dust
particles, the possibility of flying fragments, and high noise exposure. The injuries associated
with these risks can range from as little as skin irritation to a major incident such as loss of
limbs.

While the risks found through improper housekeeping are easy to avoid, they could possibly
create quite a hazardous environment if the risks are not avoided. Such risks can include
tripping hazards, slippery or wet surfaces, and unmaintained equipment.

There is a degree of high risk associated with the testing setup process which includes
working at elevated heights, working with heavy tools above head height, weights which can
cause pinching or crushing, as well as more general risks which may cause hand and feet
injuries.

The testing process creates risks which can be associated with not only sample destruction in
the form of airborne fragments, but also operator error. Due to the risks involved with the test
processes, injuries may occur if the operator is either not confident, not trained, or a

37
combination of the two. These injuries have the potential to be quite major such as the
crushing of body parts if they were caught under the test piece while loading was occurring.

Workplace restoration is simply the risks which may be produced during the entirety of my
project which may create a hazardous work environment for people in the future either using
the same test equipment or sharing the same workspace. Such risks may include any debris
left after testing, the proper disposal of any waste or finished test pieces, and reduction of any
particles which may become airborne during sample sizing.

2.8.3 Risk Evaluation


Risks associated with this project can be identified into separate categories to analyse the
immediate risk importance and consequence outcome. The consequences due to the stated
risk are labelled from levels one to five and are as follows:

1. Insignificant
2. Minor
3. Moderate
4. Major
5. Catastrophic
The risk consequences are scaled from insignificant where there are no formal injuries and
only minor delays, to catastrophic which could include death as well as possible destruction
to the external environment. The risks are also rated on their probability of occurring and
ranging from rare which is very unlikely to happen and would only generally occur when
hazard management is avidly avoided, to almost certain where day to day risks occur and can
generally be avoided.

The highest probability for a serious injury which the operator may come upon in the sample
sizing stage. This process also includes the largest array of possible hazards which can range
from insignificant to even a major risk. The minor hazard consequences include skin irritation
which is a minor inconvenience but can still be avoided during sample sizing. The
consequences then continue to increase upwards until possible appendage loss due to
improper use of the electric saw; the other associated risks and consequences can include,
minor cuts and scratches, flying fragments caused by sample pieces breaking apart during
cutting are a risk to the operator’s eyes and exposed skin, as well as long term and short term
breathing problems associated with airborne particles.

38
The risk is regards to housekeeping are relatively minimal as the CEEFC follow a strict
maintenance regime where the workspaces and environment remain clean and tidy to reduce
any risk on a day to day basis. These housekeeping risks can be classified as almost certain to
occur but are easily fixed providing all users of the CEEFC’s facilities accepts the possibility
of risk and acts to avoid any potential hazards.

Test setup also has the potential to generate fairly substantial consequences if the risk
management is not properly followed as there are a number of heavy works at elevated
heights. Within the process of adjusting the load frame for this particular test, there is a
degree of risk associated with using a ladder to undo the bolts which hold the frame in place.
There is potential for the ladder to be unstable and cause the operator to fall to what could
become a possible injury. There is also work with undoing bolts above head height which
could cause serious head injury if a bolt or tool was to fall. An unlikely but still quite
dangerous risk could be the crushing weight of the load frame if the chain block were to give
way while adjusting the height if somebody was caught underneath. The loading frame has a
total mass of 1000kg which would cause very serious injury if it were to crush somebody.

With a fully confident and trained operator, the testing procedure offers fairly low risks
providing the proper procedures are followed. The only non-human risk which may pose as a
potential hazard is during sample destruction where fragments and debris may become
airborne and potentially impale people within the immediate surrounding area. If the operator
is not confident or properly trained there is a large risk of potential crushing or pinching by
the load cell during loading.

The risks which are caused by lack of workplace restoration can include debris being left on
the ground, test equipment not properly secured and/or packed up, or improper disposal of
waste and test pieces.

2.8.4 Risk Control


Risk control is a straight forward process and firstly involves the identification and
classification of all risks into the appropriate consequence and probability categories. Onc the
risks are assessed based on possible consequence and likelihood of occurring, the necessary
steps in helping prevent such an incident from occurring can be put in place to either reduce

39
the risk of an accident arising or to ensure the operator of each particular task is fully trained
and confident with the particular safe work method relevant to the task at hand.

If the operator is deemed to be fully trained in a particular task, it becomes their


responsibility to provide the knowledge and direction as to what safe working in a specific
task consists of. Some examples of the questions a fully trained operator need to ask
themselves before beginning could include;

- What hazards are associated with performing the task at hand?


- What is the necessary personal protective equipment (PPE)?
- What steps can be taken to avoid an incident or accident?
- Am I confident with completing this task safely?
What could potentially be the most important step in a successful risk management plan, the
effectiveness monitoring of the risk control methods is as simple as following some questions
to be able to review and determine if the methods put into action are deemed successful.

 Have the chosen control measures been implemented as planned?


- Are the chosen control measures in place?
- Are the measures being used?
- Are the measures being used correctly?
 Are the chosen control measures working?
- Have any of the changes made to manage exposure to the assessed risks given a
successful result?
 Are there any new risks?
- Have the implemented introduced any new problems?
- Have the implemented control measures resulted in the worsening on any existing
problems?

40
3. Methodology
3.1 Introduction
This chapter details the specimens used as well as the testing methods and processes used in both
full scale and sample sized mechanical testing of FRP grating. Included is the preparation of the
testing specimens, including the shaping and sizing as well as the sizes of testing samples
associated with each form of testing procedure. Also covered in this chapter is a guideline into the
finite element analysis methods and procedures used to computationally investigate/correlate the
grating properties and data.

3.2 Preparation of test specimens


3.2.1 Full Scale Tests
The Full scale load tests consisted of three (3) specimens for each type of test, resulting in a total
of nine (9) test specimens being used for the line load, as well as the centred and off centred
concentrated load tests. To accurately compare each specimen the profile (thickness) and grid
alignment needed to remain consistent over all tests. One investigation as part of the burn out test
involved comparing materials between the different colours of grating to determine whether or
not the colour symbolised different materials or if it was purely for cosmetic purposes. The two
colour types of grating supplied were green and yellow, as much as possible, the colours were
divided evenly between each test to ensure a minimum of at least one (1) of each colour was
tested across all three testing procedures.

The specimens used in the full scale line load and concentrated load tests were provided by
Nepean Building & Infrastructure as part of their Weldlok Fibreglass Grating & Handrail
collection. The grating was supplied in pre-cut sizes with a collection of both green and yellow
colours. The associated sizing and colour indication as well as their respective test is found in
table 2. Since these test pieces were supplied by Nepean, the moulding and curing processes had
already been completed prior to this research project, the test pieces were ready for use
immediately. One problem associated with the provision of test materials is the lack of
information supplied with regards to the materials and chemical combinations used in the
manufacturing process. The glass to resin ratio as well as the fibre/resin materials are unknown.
The methodology behind testing for consistency of materials is explained in 3.2.2 Burn Out Test.

41
Table 2 - Full scale load testing specs.

Specimen number Length (mm) Width (mm) Thickness (mm) Colour


Line Load
LL-1 830 920 40 Yellow
LL-2 835 920 40 Yellow
LL-3 1020 920 40 Green
Centre Concentrated Load
CL-1 1020 920 40 Yellow
CL-2 1220 1040 40 Green
CL-3 830 920 40 Yellow
Off Centre Concentrated Load
OC-1 970 920 40 Green
OC-2 1200 920 40 Green
OC-3 1220 790 40 Yellow

3.2.2 Sample Sized Tests


Since only full scale testing pieces were provided, the smaller sample sized testing pieces
needed to be individually shaped and prepared from the extra grating sections not used as part
of the full scale tests. The three sample tests included flexural, compressive, and burn out; for
each of these tests, a different type of sample portion of the larger panels was produced.

During the shaping of all sample sized test pieces, appropriate PPE in the form of safety
glasses, work boots, gloves, dust mask, and earmuffs was worn to ensure maximum safety.
The samples were shaped using dangerous power tools and all proper procedures and
instructions of use were acknowledged and followed. In regards to dust, the stationary
machines used such as an electric saw have in built dust extractors which were used, if
extractors were not a part of the tool used, all means of trying to alleviate the risk of the fibre
dust becoming airborne in the workshop was taken.

42
Flexural Test

The flexural testing specimens consisted of five (5) thin lengths of grating approximately
500mm in length. The width of the samples was determined by cutting webs of the larger
panel directly through the centre. The final specimens needed to contain one (1) fully in tact
row of grating and half webs on either side. It is estimated that the two lengthwise strips took
all of the flexural loading and the smaller perpendicular webs were only counted as
‘connectors’ to hold the lengthwise strips in place while the load was applied. The specimens
can be viewed in the corresponding figure 7 below.

The flexural test pieces were cut using a diamond tipped electric saw at Buchanan’s
Advanced Composite Technologies (BAC). The accuracy of measuring and cutting of these
samples was paramount as to ensure the accuracy of results was not hindered by the simple
task of sample shaping. A selection of five (5) test pieces was decided to ensure an
appropriate amount of results could be compared for the flexural loading.

Figure 7 - Flexural test specimens (Nicol, Lachlan. 2014)

43
Compressive Test
The compressive load test samples consisted of creating a range of square blocks containing a
ranging number of full grid pieces. The samples created comprised of three (3) specimens for
1x1, 2x2, and 3x3 grid blocks giving a total of 9 compressive samples. The arrangement of
the final test specimens for the compressive tests are displayed in figure 8. Similar to the
flexural test specimens, the compressive samples were cut to size using a diamond tipped
electric saw with proper PPE being worn whilst shaping.

Figure 8 - Compressive test specimens (Nicol, Lachlan. 2014)

Burn Out Test

For the burn out test, the requirement for very small samples was important to be able to fit
the test pieces in the appropriate crucibles to be placed in the oven. It was determined that
singe webs of the grating would suffice for such a test. The investigation of materials used
after the burn out test was completed involved determining whether or not the colour
difference was a result of different materials between the green and yellow panels. It was

44
important to ensure that a minimum of one (1) sample for each colour was tested to be able to
investigate any differences between the colours.

Since the sample sizes were so small, it was determined to be too dangerous to use the same
cutting procedure as the flexure and compressive test samples. With this in mind, a diamond
tipped drop saw with a thin blade was used with the samples being cut fully secured using
clamps so that they did not flick out of the saw and become airborne with the possibility to
injure someone.

Figure 9 - Burn out test specimens (Nicol, Lachlan. 2014)

45
3.3 Testing Procedures
3.3.1 Full Scale Tests
All three forms of full scale load testing, line load and the two concentrated loading cases
were all conducted using the same test setup. This setup included the use of two large metal
round bars spaced at 830mm centres acting as the supports of the testing panels. These round
support bars were in turn connected to large, heavy ‘I’ beams which ensured no movement of
the support width during testing. The accuracy of the test results is paramount and movement
in the support width would give incomparable and unusable data, with this in mind, the
support bars and large ‘I’ beams provided a solid support for test panels to ensure maximum
accuracy of results. The test pieces were then placed evenly onto the support bars with a
distance laser placed underneath all specimens directly in the centre of the applied loading.
Specific loading procedures and tool set ups are clarified in the following sections: Line
Loading, and Concentrated Loading (Centred & Off Centre). Figure 11 shows the testing
frame and set up for the line loading test, the frame setup is identical for all three testing
types with the exception of the load application specifics.

In order to record the data for stress and strain, strain gauges were attached to a number of
test panels to in turn be connected to a computer for recording of the stresses and strains
observed during each loading case. These strain gauges were positioned on the centre webs in
either a vertical or horizontal manner, across the longitudinal or transverse webs. The
longitudinal and transverse orientation is determined by the positioning of the loading,
longitudinal being parallel with the load and transverse being perpendicular to the load.
Below in figure 10 is a depiction of how the strain gauges were attached to the testing
samples.

46
Figure 10 - Arrangement of strain gauges (Nicol, Lachlan. 2014)

Line Loading

The line loading setup involved attaching a 200kN hydraulic load cell to the large scale
testing frame and adjusting the frame height to ensure the load applicator was as close to the
test panel as possible. There was a restriction on the distance the load cell could be moved
towards the test panel and was a constraint governed by the amount of bolt holes in the
testing frame, the frame itself could not be safely lowered any further which resulted in a
very large gap between the load cell and the testing panel. The load cell was then connected
to a large ‘I’ beam which reduced the gap between the load cell and the test panel
significantly and meant the loading bar could be positioned within a few millimetres of the
load cell (i.e. ‘I’ beam plus the load cell). Between the ‘I’ beam and the test panel, a
rectangular hollow metal bar was positioned directly in the centre of the testing width. This
rectangular bar would essentially become the applied loading area for the line loading tests.

Once set up was completed, the force was slowly applied to the load cell with the data for
applied force, deflection, stress & strain being recorded by the computer for investigation and
analysis. The load was steadily applied until the FRP panels reached final failure or the load
cell reached its maximum load limit, i.e. the load plateaued and would not significantly
increase. The load was then released and the data recorded and stored.

47
Testing Frame

Load Cell

Spacing ‘I’ Beam

Load Bar

Test Panel

Supports

Figure 11 - Line load testing setup (Nicol, Lachlan. 2014)

Concentrated Loading (Centred & Off Centre)

The concentrated loading cases both involved the use of the 500kN load cell and a
concentrated load block 60mm square by 50mm depth. This load block would simulate the
effect a concentrated load has while being applied to molded FRP grating panels. The load
block consisted of solid steel welded to a base plate which would in turn be bolted to the
loading cell and positioned ready for testing.

Figure 12 - Concentrated load block dimensions (Nicol, Lachlan. 2014)

48
For the centred loading case, the load block was positioned directly in the centre of the test
panel, both longitudinally and transversely. However, the off centre load was positioned at a
distance of 230mm from the support. This distance was determined to ensure the
concentrated load was far enough away from the centre of the panel to provide a difference in
results to be compared to the centred loading, as well as being sufficiently far enough away
from the end support so that the deflection recording would not be hindered by being too
close to the support itself.

The distance recording laser was positioned so that it was directed exactly in the centre of the
applied load, this would ensure the maximum deflection during loading was recorded, it also
guaranteed accuracy between the different samples so that the set up was equivalent for each
case. Similar to the line loading, once the setup for the test panels was complete, load was
applied steadily to the specimen until either full failure was observed or the load cell reached
its maximum operative limit.

Figure 13 - Centred & off centred loading (respectively) (Nicol, Lachlan. 2014)

49
3.3.2 Sample Sized Tests
Flexural Test

The flexural sample testing of the five test pieces previously outlined in 3.2.2 Preparation of
Test Specimens entailed the use of the MTS Insight Electromechanical 100kN load testing
machine in a four point bend test. Two fixed supports were spaced 450mm apart which would
comfortably support the 500mm flexural samples prepared earlier. Since this particular form
of testing is a four point bend test, the load is applied using a set of two round bars spaced at
150mm centres, is attached to the loading crosshead to create the four points (2 supports, and
2 load bars). Once the specimen is in place, the crosshead was lowered to approximately
1mm from the panel surface. Once the test setup is complete and the specimen is ready, the
load is applied automatically with data such as crosshead distance and applied load being
recorded on to the computer for interpretation and analysis at a later date. An example of one
of the flexural test specimens in the process of being loaded is shown below in Figure 14.

Figure 14 - Flexural sample during loading (Nicol, Lachlan. 2014)

50
Compressive Test

The SANS Hydraulic Compression Testing Machine – 2,000kN was used as the compressive
force application machine in the series of sample compressive tests. The series of
compressive tests consisted of 1x1, 2x2, and 3x3 grid blocks to compare the difference of
results between singular grid blocks and the possible strength increase of blocks kept in larger
combinations. A quick brush down of the test bearing pad was required to clean away any
remnants from previous tests which could potentially hinder the accuracy of the results
during testing. The test samples were place directly in the centre of the bearing pad with the
crosshead manually adjusted to move no more than 1-2mm away from the test specimens.

Once these setup procedures were competed, an input of load rate for the test was entered
into the computer to determine how fast (in seconds) the load would be applied, this would be
a factor of both accuracy and time available for testing, the load rate of 3mm/min was
selected as it would give relatively accurate results for this testing while not taking too long.
If extremely accurate results were needed, the load rate could have been slowed down to
ensure higher accuracy was achieved, however for this particular set of tests, increased
accuracy was not necessary as the results will only be examined to a certain tolerance and the
difference in accuracy of test results would not have been worth the extra time the tests would
have taken.

Figure 15 - Compression testing for a 1x1 grid sample (Nicol, Lachlan. 2014)

51
The output data which is used to analyse the test samples compressive strength include the
play time and the applied load. The compressive strength is determined by dividing the peak
load before failure by the cross sectional area of the samples.

Figure 16 - Computer used to record compressive testing data (Nicol, Lachlan. 2014)

Burn Out Test

The burn out test is an examination of the fibre content within a FRP product and involves
the burning away of the resin matrix which surrounds the fibres. In this particular case, the
material content of the FRP grating is unknown as the grating was supplied by a third party
(Nepean Engineering) so the results of the burn out test give a good indication of exactly
what type and amount of fibre that was used in production of the grating. Not only is the burn
out test a comparison between both transverse and longitudinal contents but also a
comparison between the colours of the grating. It is unknown if the fibre contents between
the green and yellow samples is consistent or if the colour itself could be an indication of the
fibre percentage it contains.

52
The test itself is very easy to set up and conduct as it is a matter of weighing all specimens
prior to heating and in turn re-weighing the samples post heating. The specimens used in the
burn out test are a collection of webs from both green and yellow panelling in both transverse
and longitudinal directions for each. The testing procedure is in accordance with ISO
1172:1996 ‘Textile-glass-reinforced plastics’ with particular attention directed towards
chapter 7. The method involves firstly weighing of the crucibles fully dried and entirely
empty. The small FRP webs are then placed in the crucibles and re-weighed to give a value
for the weight of the crucible plus the weight of the sample piece itself.

Figure 17 - Weighing of the crucible and the grid sample (Nicol, Lachlan. 2014)

The burn out specimens were placed in a Thermolyne industrial benchtop furnace and heated
to 600°C for a total of 4 hours to allow the resin matrix to be fully burnt away thus leaving
solely the fibres.

From the remaining fibres left in the crucibles, an analysis is conducted to compare any
differences of fibre content between the longitudinal and transverse webs as well as a
comparison between the yellow and green samples to determine if there is in fact a difference
with respects to the colour of the grating.

53
Figure 18 - Thermolyne industrial benchtop furnace (Nicol, Lachlan. 2014)

54
3.3.3 Finite Element Analysis

A finite element analysis (FEA) simulation was developed to aid in the prediction of the
flexural behaviour of full scale FRP grip panels by using the results obtained from the sample
sized testing. The concept of the FEA was to use values such as flexural modulus and density
taken from the sample sized tests to be used for material properties as part of a model
simulation in predicting deflection and the stress and strain which a full scale panel may be
subject to in the different loading cases.

To be able to justify the use of the FEA as a prediction method of full scale panel mechanical
properties, the results found using the program Creo need to match closely to the results
found in the various tests conducted throughout this research paper. Models of both the
flexural specimens and full scale panels were developed using the 3D modelling software
SolidWorks, these models were then imported into Creo ready for analysis. The first model
developed was the flexure sample. The reasoning behind the analysis of the flexure sample
first is that if the results of the FEA simulation using the flexural mechanical properties
corresponds closely to the actual results from testing then it can be said the properties used
are indeed correct and successfully simulate the physical testing.

During the physical tests for flexure, dimensions of the grating were taken of all 5 sample
pieces, the average dimensions of these tests were used in creating the model ready for FEA.
Figure 19 shows the model created in SolidWorks with a few critical dimensions labelled.
Since the support span of the flexural test was 450mm, the model in the FEA was used as a
total 450mm with the fixtures acting at either end of the model. It was found that the average
web thickness of samples was 7.87mm with an internal spacing of 30mm.

55
Figure 19 - Flexure test 3D model (Nicol, Lachlan. 2014)

In order to successfully simulate the tests via Creo, certain material properties need to be
obtained and are assigned to the 3D model. Some of these properties include but are not
limited to: flexural modulus (Ef), density of the material (ρ), and compressive strength. The
values for these different material properties are established via the sample sized testing
procedures discussed above in the methodology with the values taken from the results
obtained during the physical testing. Figure 20 shows the flexure model ready for analysis in
Creo, this figure shows all of the features needed for simulation including material
assignment, fixed constraints, and the applied load spaced at 150mm to simulate the load
from the physical flexure test. While the flexure test included two bars for the applied loads,
Creo required the load to be fully applied to a surface on the model, hence why figure 20
shows 4 loads being applied. These four load points essentially represent the two loading bars
being applied across the flexure samples.

56
Figure 20 - Flexure 3D model ready for analysis (Nicol, Lachlan. 2014)

57
Once the flexure model was verified and the results of both the FEA and the physical testing
were closely comparative, a full scale analysis was undertaken. Similar to the flexure analysis
the 3D model of the full scale panel was developed in SolidWorks and imported into PTC
Creo with all required features being added. For the line load analysis, the load was applied
directly in the centre of the panel and covered the entire span, this closely simulates the setup
which was used for the full scale line load testings.

A mesh is applied to the 3D model to help define the geometry which is to be analysed. The
finer the mesh is, the more accurate Creo can interpret the geometry of the FRP grid panel
and therefore provide a more accurate analysis with a more precise set of results. A mesh size
of 3mm was used and applied for the flexure simulation alone, this mesh size was deemed
unusable for the full scale simulation due to the amount of computing time needed to obtain a
solution. Therefore, a mesh sizing of 7mm was used for the full scale analysis which was
determined to give an accurate result while still being computed within a practical timeframe.
The material properties used for the full scale simulation remain the same to that used in the
sample sized simulations.

The analysis setup for the two concentrated loadings was identical to that of the line loading
case, however the applied loads were positioned differently, in the centre as well as 230mm
from the edge of the supports. Instead of the load following a line across the span of the
panel, the two concentrated load analyses had an applied load which simulated the loading
block of 60x60mm square. From all loading cases, a coloured model was produced for the
displacement, stress, as well as the strain observed during simulation and analysis.

58
Figure 21 - Full scale line load ready for analysis (Nicol, Lachlan. 2014)

59
4. Results and Analysis
4.1 Introduction
This chapter contains an in depth analysis and discussion on the results found from all forms
of testing performed in this research project. The testing analysed includes the full scale
testing (line load and concentrated loading), sample sized testing (flexural, compressive, and
burn out testing) as well as a finite element analysis which compares the results obtained
from physical testing with a set of simulation results obtained from 3D modelling analysis.

4.2 Full Scale Testing Analysis


4.2.1 Line Loading

Figure 22 - Deflection results for line loading

60
Figure 22 displays the deflections until physical failure for the line loading cases of the three
separate test specimens. The figure displays a peak deflection range prior to failure between
approximately 60-68mm. The deflection due to increasing loading displays an almost linear
relationship from commencement of loading up until failure of the test specimens.

The load applied to both LL-1 and LL-2 when failure occurs is within the region of 53kN
whereas LL-3 displays a much higher peak load of approximately 64kN. The determination of
failure associated with the physical results ascertained from the recording equipment can be
explained by the rapid decline in applied load i.e. the breaking of the linear pattern. As an
inspection observation taken during the testing procedure, the physical failure is described as
major cracking and delamination on the bottom face and side of the grating.

Figure 23 - Failure of line loading test panel (Nicol, Lachlan. 2014)

Table 3 - Line load test results

Test ID Deflection Max (mm) Load Max (kN) Support Length (mm) Width (mm)
LL-1 60.00 52.90 830 920
LL-2 64.67 53.90 830 920
LL-3 69.88 61.68 830 920
Average: 64.85 56.16
61
4.2.2 Centred Concentrated Load

Figure 24 - Deflection results for centred concentrated loading

Similar to the deflection found for the line loading, the load to deflection ratio is relatively
linear for the centred concentration load tests. The exception is the slight changes close to the
failure point, the change associated with CL-1 is observed as a slight drop of applied load at
approximately 28kN, the change is CL-3 is not a drop as such but more of a slight change in
direction of the linear load/deflection pattern. These minor differences could be due to a
number of factors including the strength loss as a result of the development of minor cracking
of the polymer resin matrix or a slight pressure error caused by the load application. CL-2
remained relatively linear throughout the test until it reached physical failure status.

The peak applied loads for all three test specimens fall within a range of 31.5-35kN before
reaching their respective failure limits. The maximum deflection for centred concentrated
loading appears substantially lower than the line load tests with a maximum out of the three

62
tests (CL-1) only reaching 53mm with CL-2 and CL-3 deflecting to approximately 47.3mm
and 46.5mm respectively. The failure determined by observation for concentrated loading in
the centre was the cracking of laminates within the immediate surrounds of the loading block
with the block appearing to ‘sink’ into the test panels. Once the load was released, the area
which the load block sunk into the grating in fact rebounded back to near its original profile
with only approximately 2-3mm of permanent depression of the grating webs.

Sinking of load block


during mechanical failure
of grating

Figure 25 - Failure of centred concentrated load test panel (Nicol, Lachlan. 2014)

Table 4 – Centred concentrated load test results

Test ID Deflection Max (mm) Load Max (kN) Support Length (mm) Width (mm)
CL-1 52.26 33.52 830 920
CL-2 49.00 34.94 830 1040
CL-3 46.15 31.64 830 920
Average: 49.14 33.37

63
4.2.3 Off Centre Concentrated Load

Figure 26 - Deflection results for off centre concentrated load

As observed in figure 26, the off centre concentrated load deflection did not remain linear
after about 30kN for OC-1 & OC-3, and the variation away from a linear pattern was
observed after approximately 25kN for the OC-2 test sample. The load to deflection ratio
begins to deviate fairly dramatically until the complete failure of the specimens is detected at
55-60mm of deflection. One interesting point to take note of from the failure limits of this
particular test is that the deflection slightly rebounded when the mechanical failure limit was
hit whereas the failure in the centred load tests did not rebound but in fact slightly increased
the deflection of the samples.

The physical failure itself was observed to be the same as the centred concentrated load in
that the loading block sunk into the grating and in turn recovered to near original profile once
the applied load had been released.

64
Table 5 - Off centre concentrated load test results

Test ID Deflection Max (mm) Load Max (kN) Support Length (mm) Width (mm)
OC-1 59.02 32.61 830 920
OC-2 55.78 36.41 830 920
OC-3 56.92 35.77 830 790
Average: 57.24 34.93

65
4.3 Sample Sized Testing Analysis
4.3.1 Flexural Loading
12

10

8
Load (kN)

0
0 5 10 15 20 25
Deflection (mm)

FL-1 FL-2 FL-3 FL-4 FL-5

Figure 27 - Flexural deflection results at load points

The load deflection relationship for the five flexural loading samples is shown in figure 27.
The figure shows that the load capacity retains a relatively linear relationship with deflection
for the entirety of the loading until the observed failure limits. The failures observed are clear
on the graph as the sudden decline in loading on the test specimens. The maximum applied
loads for the flexural tests lie within a range between approximately 7.3 and 10.6kN with
deflections between 17-24mm at the load points. The collection summary of the flexural
loading results is shown in table 6.

It is interesting to note that flexural test 3 (FL-3) displays slight drops of applied loading at
approximately 5.5 and 6kN, whereas the other four tests remain constant over the entirety of
the test until failure. This could possibly be deduced as slight cracking or delamination of the
specimen which immediately reduces the load capacity of the flexural test piece. Another
interesting point is that FL-3 was also witnessed to be the smallest deflection until failure,
therefore giving a smaller applied load as well as the smallest deflection before complete
failure (17.87mm & 8.09kN).

66
Figure 28 - Example of shear failure for flexural loading (Nicol, Lachlan. 2014)

The failure modes of the flexural tests were observed as shear failures, where longitudinal
cracking of the specimens lead to complete delamination of entire fibre layers creating
separation. Figure 28 shows this particular delamination where it can be said that cracking
occurred between two fibre layers which broke the bond of the resin matrix. This would
suggest that the thin flexural samples are weaker in the lengthwise directions (i.e. weaker
over the 450mm span).

As discussed in section 2.7.1.2 in regards to the calculation of deflection at the centre of the
flexure specimens, the deflection at mid span i.e. maximum deflection needs to be calculated
to create a comparable set of results. Equation 5 was used to develop an approximation of the
deflection at mid span for all 5 specimens. The average maximum deflection equated to be
approximately 31.84mm

67
Table 6 - Flexure test results summary

Test ID Deflection Load Max m Flexural Stress Deflection at Flexural Strain Flexural Modulus
Max (mm) (kN) (MPa) Centre (mm) (Gpa)

FL-1 23.73 10.425 0.46 186.28 34.96 0.04143 10.40281032

FL-2 20.73 8.587 0.375 153.44 35.32 0.04186 8.480551888

FL-3 17.87 8.09 0.446 144.56 27.98 0.03316 10.08620305

FL-4 18.72 7.33 0.376 130.98 30.07 0.03564 8.503166693

FL-5 19.68 10.59 0.529 189.23 30.88 0.03660 11.96323186

Average: 20.15 9.00 0.4372 160.90 31.84 0.03774 9.89

68
Table 6 as viewed above, shows the results of the 5 separate flexural loading cases with
worked calculations for equations 2 – 5. Also included in table 6 is the calculation for
deflection at the mid-point since the recorded data only displays deflection at the applied
loading points. Over the 5 different tests, an average value for flexural modulus was obtained
of 9.89GPa. This modulus was added to the material properties used in Creo for the FE
analysis simulation process. The calculation of the flexural modulus required a value of m
which was the gradient of the load deflection curve and was taken directly from the graph
displayed in figure 28.
The calculation of deflection at mid span rather than at the load points now gave a range
between 27 and 36mm and an average centre deflection of 31.84mm. The average flexural
stress of all test specimens was calculated to be 160.9MPa, ranging between 130-190MPa.
This range of stress is rather large with a 37.5% of variation which is relatively substantial
considering the products are all manufactured out of the same material.

69
4.3.2 Compressive Loading

1x1 Compressive Loading


120

100

80
Load (kN)

60 CL1x1-1
CL1x1-2
40 CL1x1-3

20

0
0 20 40 60 80 100 120

Play Time (s)

Figure 29 - 1x1 compressive loading results

Figure 29 illustrates the combined compressive load test results for all three 1x1 blocks. The
behaviour of the three specimens appears relatively consistent and comparable for the
majority of the loading over the observed time, i.e. the load to play time curves are all very
similar and follow a consistent trend. It’s not until after approximately the 60kN mark do the
three samples vary from one another. As observed, CL1x1-1 rises to a maximum of 76.24kN
before rapidly dropping in applied load. CL1x1-3 on the other hand maintains its complete
compressive strength up to an applied load of 70kN where a slight drop in load is detected
before the compressive strength picks back up again and continues to rise constantly to 79-
82kN. At this point, the applied load over time starts to taper off and begins to plateau while
displaying small load drops which indicates slight compressive failure of the block and is a
direct contribution from the internal cracking of the resin matrix while the fibres still manage
to hold the grid structure together. This ‘plateau’ of loading occurs over about 35 seconds
before the eventual decline of the loading which indicates full compressive failure of the
singular block.
70
Similar to CL1x1-3, the second test block (CL1x1-2) rises consistently for 42 seconds before
displaying a small load drop. The applied load rises again, noted as a rise not quite as steep,
until approximately 100kN where the results then show a decline of load including a number
of compressive strength load drops until a play time of 80 seconds (or approximately 90kN).
At this time, the load suddenly drops which again is an indication of complete compressive
failure of the 1x1 sized block.

Figure 30 - Example of 'crushing' compressive failure for 1x1 block (Nicol, Lachlan. 2014)

71
2x2 Compressive Load
250

200

150
Load (kN)

CL2x2-1
CL2x2-2
100
CL2x2-3

50

0
0 10 20 30 40 50 60 70 80 90
Play Time (s)

Figure 31 - 2x2 compressive loading results

The results displayed in figure 31 of the 2x2 grid block tests show very similar trends to that
observed in the CL1x1-2 & CL1x1-3 results, where there is a comparable rise of loading over
time with a point in time reaching failure stage and showing a plateau of loading with small
load drops before an eventual decline of load. CL2x2-3 was the only test out of the three which
showed an initial drop of load at around the 43 second mark. The other two tests (CL2x2-1 &
CL2x2-2) rose constantly until the compressive strength began to decline. CL2x2-1 displays a
rather large and sudden drop of approximately 15kN before reaching the compressive plateau
and displaying the constant load drops as expected.

In the figure of given results, CL2x2-2 can be descried as an outlier as the results show a fairly
large variation of the load to play time ratio. CL2x2-1 & CL2x2-3 follow an almost identical
loading trend up until initial compressive failure and both show a maximum loading of 230-
230kN whereas the outlier (CL2x2-2) only ever reaches a maximum loading of 210kN. It is an
observation that both CL2x2-1 & CL2x2-3 have a larger total compressive strength compared to
CL2x2-2.

72
3x3 Compressive Loading
600

500

400
Load (kN)

300 CL3x3-1
CL3x3-2
200 CL3x3-3

100

0
0 20 40 60 80 100 120
Play Time (s)

Figure 32 - 3x3 compressive loading results

The results of the 3x3 grid block tests show and array of very consistent load/play time
trends. All three specimens slowly increase in load over time until they eventually begin to
display the small load drops which has come to be expected for these particular types of tests.
Different to the previous tests (1x1 & 2x2) however, the transition into compressive failure is
relatively smooth with no obvious point in time where failure begins. One slight exception in
the otherwise smooth set of results is the slight decrease in load early on in the CL3x3-1 test.
Since this is observed so early into the loading, it can be assumed that it is simply the failure
of the non-slip coating on the top of the test piece. Since all grid specimens are coated on top
with a non-slip sand and resin coating, it is assumed that at this load drop the sand ‘settled’
and the entire surface of the loading block began to be loaded. The range of maximum loads
applied to all three of the 3x3 grid tests was observed as 433-482kN including an average
load of 460kN.

73
Table 7 - Compressive loading test results

Test ID Length (mm) Width (mm) Web Thickness (mm) Total Area Under Loading (mm²) Max Load (kN) Compressive Strength (GPa)
C1x1 -1 45.94 45.86 7.98 1210.41 76.24 62.987
C1x1 -2 46.24 45.69 7.75 1184.67 99.31 83.833
C1x1 -3 45.26 46.02 7.78 1178.20 90.05 76.430

C2x2 -1 83.36 83.74 7.96 3420.09 231.37 67.649


C2x2 -2 83.78 83.92 7.90 3412.80 210.40 61.650
C2x2 -3 83.26 83.8 7.91 3401.22 234.35 68.902

C3x3 -1 122.22 122.44 7.84 6689.09 433.36 64.786


C3x3 -2 121.73 122.1 7.76 6605.00 481.41 72.886
C3x3 -3 122.2 121.85 7.91 6720.65 466.58 69.424

Average: 69.838

74
4.3.3 Burn Out Test

Table 8 - Burn out test results

Colour Green Green Yellow Yellow Average


Orientation T L L T
m1 20.6133 21.3045 20.1489 23.3442
m2 43.6617 41.3639 44.1289 45.2253
m3 32.9777 32.5350 33.0859 34.8854
Sample Mass (g) 23.0484 20.0594 23.9800 21.8811 22.24
mglass (%) 53.6454 55.9862 53.9491 52.7451 54.08
mglass (g) 12.3644 11.2305 12.9370 11.5412 12.02
Volume (m^3) 1.4399E-05 1.44E-05 1.44E-05 1.44E-05
Density(kg/m^3) 1600.69983 1393.115 1665.399 1519.631 1544.71

The results displayed in table 8 show the formulation of equation 7 from the literature review
section 2.7.1.2 ‘Burn Out Test’. This formula is made up of the difference factor between the
initial mass of a test specimen and the final mass of the remaining fibrous content. The results
of the burn out test and applying equation 7 give a fibre content ranging between 52-56%. As
an average over the four (4) separate tests, it can be said that as an average, the FRP grating
has a glass content percentage of approximately 54%. This means that in total, the
combination of glass and resin is at a 54% to 46% respectively. This ratio can be applied to
much larger panels to still gain a good understanding of the fibre ratio.

For full sized panels, the total weight of glass which is included in the total weight can be
calculated by finding 54% of the total weight of panel. The reverse can be done if the weight
of resin is required. The reasoning behind having different sample types (i.e. both green and
yellow as well as both transverse and longitudinal webs) was to compare and determine any
differences there may be between both the colour of the panels and direction of fibres. Since
the resulting standard deviation is approximately only 2, there is minimal difference between
all of the specimens as they all display a similar amount of fibres. A larger investigation into
the fibre to resin ratio would have needed to be undertaken if results were not so closely
comparable. However, since the results of the burn out test do indeed display very similar

75
results between specimens, it is safe to assume that the percentage of fibre remains relatively
consistent over all panels.

A small amount of what can be described as filler material was found once the resin had been
burnt away. Since the amount of filler was relatively minimal its weight was neglected in the
analyis and calculation of the glass to resin ratio. The filler merely retained the grating web
structure once the resin matrix was burnt away and did not contribute to the weight or
strength enough to conduct a different separation test.

Figure 33 - Burnt out FRP web containing 'filler material' and glass fibres (Nicol, Lachlan. 2014)

As well as calculating the fibre ratio within the FRP grating, it was important to calculate the
density of the product to aid in the finite element analysis. Given the average volumes and
weights of the 4 specimens analysed, the average density was calculated to be 1544.71kg/m 3.
As a theoretical comparison, the common density for glass fibres is 2580kg/m3 and resin
density is commonly 1200kg/m3. Using the fibre-resin ratio evaluated earlier, a combined
density should equate to approximately 1947kg/m3. This value is however purely theoretical
using common densities for both glass fibres and resin matrices and is only used to ensure the
calculated density is somewhere within the expected range, which is in fact the case.

76
4.3.4 Finite Element Analysis

Flexure Analysis

From the collection of results obtained from the sample sized testing results previously
discussed, mechanical properties and loads for use as part of the finite element anaysis were
added to the flexural model in Creo ready for analysis. The values used include:

ρ = 1544.71 (kg/m3)

Ef = 9.89 (GPa)

σc = 69.84 (GPa)

P=9 (kN)

Figure 34 - Load-deflection simulation at 10% deformation of flexural model (Nicol, Lachlan. 2014)

The load-deflection behaviour of the flexural analysis is displayed in figure 34 with a


deflection magnification of 10%. The deflection predicted using Creo’s simulation processes
was estimated to be approximately 32.83mm using a load of 9kN applied in a 4 point bending
pattern spaced evenly at 150mm (i.e. 75mm from the centre).

77
To successfully proceed to full scale analysis, the FEA on the flexure specimen needs to
correspond closely with the results obtained from the physical flexure tests undertaken. Since
the testing undertaken gave a set of 5 separate results, and the FEA gave only one set of
results, and average of all 5 physical tests was used in the comparison between theoretical
and physical results. As an average, the physical testing specimens deflected approximately
31.84mm at mid span and the FEA gave a resulting deflection estimation as 32.83mm. This is
a variation if only 1mm (or a 3% difference) which is extremely similar and therefore
comparable.

Given the similarity in results between the physical testing and the finite element analysis, it
can be assumed that the material values used are correct and can be used in the accurate
simulation of the full scale tests. Figure 35 & 36 show the stress and strain distribution
relationships for the flexure analysis. The results found using PTC Creo simulation process
gave rather extreme results of both the stress and strain maximum and minimums, this can be
described as stress/strain rises at the points of loading and at the fixtures. For applicable
analysis, these values were neglected and the legend range modified to display the stress and
strain distributions closer to the actual results i.e. for the stress distribution shown in figure
35, 4062.87MPa was neglected and the maximum stress of 160MPa was used in its stead,
thus providing a new range between 10 and 160MPa.

Figure 35 - Stress distribution for flexure analysis (Nicol, Lachlan. 2014)

78
With the legend range adjusted for accurate analysis, figure 35 shows the maximum stress at
the centre of the specimen to be of yellow colouring, this corresponds to a flexural stress on
the legend to be approximately 141.25MPa. As discussed in section 4.3.1 for the results of
the physical flexural testing, an average flexural stress was calculated to be 160.9MPa over
the 5 separate samples. This equates to a difference between the theortical and physical
results of approximately 13% which is deemed to be an acceptable percentage of difference.

Figure 36 - Strain distribution for flexure analysis (Nicol, Lachlan. 2014)

Similar to the flexural stress FEA results, the legend in figure 36 needed to be adjusted to
display the strain distribution which could be compared to the physical test results. The
colour which displays the largest strain reading (excluding the values disregarded due to
strain rises at loads and fixtures) is the green colour drectly below the yellow which
corresponds to a strain value of 0.03025. This reuslts in a percentage difference of
approximately 22% between the FEA and the physical testing results of 0.03774.

79
Full Scale Analysis

Since the flexural finite element analysis results were closely comparable with the physical
test results, the full scale test panels used the same material properties listed in the flexural
analysis section with the only variation being the maximum applied load and the loading
location/type.

As discussed in section 4.2.1 the average maximum deflection which was observed for the
line loading test panels was at approximately 64.85mm. The resulting maximum deflection at
the centre for the FEA simulation gave a relatively similar resulting deflection of
73.0277mm. The percentage difference between these two values was calculated to be 11.8%,
which can be considered relatively accurate for a FEA result comparison. As can be viewed
in figure 37, the largest deflection occurs in the centre of the test panel and slowly decreases
as it nears the supports, which correlated directly to what would be expected and what was
observed in the physical testing.

Figure 37 – Load-deflection simulation at 10% deformation for full scale line load (Nicol, Lachlan. 2014)

80
Figure 38 displays the deflection generated due to a point load of 33.37kN taken from the
average maximum load found in the centred physical tests and is applied directly in the centre
of the analysis model. The results from the FEA give a maximum deflection at the centre of
the model as 47.63mm which corresponds relatively closely with the deflection found in the
physical testing of 49.14mm. A difference percentage of only 3% between the two values was
found which can be classed as extremely accurate for an FEA comparison. As expected, the
deflection is greatest in the centre of the test panel and progressively decreases towards the
supports and edges of the panel.

Figure 38 - Load-deflection simulation at 10% deformation for full scale centred concentrated load (Nicol, Lachlan. 2014)

81
Consistent with the centred loading case, the off centre loading in figure 39 shows a
deflection which is greatest directly underneatht the applied load which gradually reduces
closer to the edges and the fixtures. The reduction of deflection is spread over a greater
distance on the side furthest away from the support i.e. the longer distance to the support.
Whereas, the transition from maximum to minimum deflection on the shorter distance to the
support is relatively rapid. The average maximum deflection taken from the physical tests for
off centre concentrated loading was calculated to be 57.24mm. The deflection calcualated as
part of the FE analysis was approximately 53.82mm.

Figure 39 - Load-deflection simulation at 10% deformation for full off centre concentrated load (Nicol, Lachlan. 2014)

82
5. Conclusions & Future Work
5.1 Introduction
This project has investigated the mechanical behaviours of fibre reinforced polymer grating
for use in walkways and boardwalks. The investigation has included the physical testing of
both full scale and sample sized specimens as well as the theoretical simulation using the FE
analysis function in PTC Creo. The physical testing included three full scale tests, line load,
centred concentrated load, and an off centre concentrated load, in addition to the full scale
tests, sample sized tests were also undertaken which included tests for flexure, compressive
strength, and fibre content as part of a burn out test. The panels used for the testing were
manufactured and supplied for investigation by Nepean Engineering. Buchanan’s Advanced
Composites provided strain gauges to be attached to the full scale panels to aid in the
investigation of mechanical properties.

The various properties which have been calculated using the results of the physical testing
and the use of the equations listed in section 2.7 of the literature review aided in the
development of material properties for use in the finite element analysis comparison. The use
of FEA in Creo as a way to simulate results was verified by comparing the results of the
sample sized tests and what was predicted using Creo’s simulation function.

5.2 Failure Modes in Testing


The failure modes observed in the physical testing varied depending on the different loading
cases. The full scale line loading showed major cracking of the grating on the bottom side of
the panel which could be described as failure due to excessive tension. While the major
cracking and breaking occurred on the tension plane of the FRP grid, minor cracking was also
observed on the compression side, i.e. directly below the applied line load. The failure of the
two different concentrated loading cases, centred and off centred, was observed as the
loading block sinking into the grid structure. This sinking is predicted to be caused by the
failure of the grid structure immediately surrounding the load block, while the rest of the
panel remained undamaged.

The failure mode for the five separate flexural tests was observed as major cracking along the
longitudinal direction which led to full separation and delamination along the length of the

83
specimen. This type of failure can be classified as a form of shear failure, where the weakest
layer of glass & resin completely failed causing the specimen to break into two separate
pieces. The compressive loading cases displayed failure similar between the three different
sizes, 1x1, 2x2, and 3x3. This compressive failure was observed as the crushing of the
specimens and collapse of the resin matrix, this failure corresponds to the decline of applied
load using the computer recording software for compressive loading. While the burn out test
essentially did not have a failure mode, the experiment was directly aimed at the investigation
of the internal glass content of the FRP grating as well as calculation of the density of the
product.

5.3 Loading Case Comparison


The principle aim of testing three different loading cases for full scale panels was to
distinguish any differences as well determine which load case could be considered the most
critical. The three line loading tests gave an average maximum load of 56.16kN for a
maximum deflection of 64.85mm, whereas the concentrated loading cases both gave
maximum applied loads of approximately 34kN. The centred concentrated loading was
recorded to reach a maximum deflection of 49.14mm whereas the off centre loading reached
57.24mm for a similar applied load.

5.4 Use of FEA


The theoretical model simulation and prediction which has been undertaken using Creo’s
simulation function provides a very good approximation for the flexural behaviour of the
results found in the physical testing. The results obtained from the flexural specimen analysis
was closely relatable with the deflection and stress/strain relationship results observed as an
average of the 5 physical flexure tests undertaken. The confidence gained by closely
comparing the flexural physical and theoretical results justified the use of the FEA software
and material properties for full scale analysis and prediction. The confidence in the outcomes
of the FEA was furthered with deflection results of all three full scale load cases being very
closely relatable. Over the three load types, a percentage of difference between theoretical
and physical ranged from 3-11% which is classed to be extremely accurate and deemed
useable for any further predictions/analysis.

84
5.5 Summary
As a summary of the important results of this research and investigation into the mechanical
propertied of FRP grating, a list of mechanical properties is supplied as well as a summary
table which displays the maximum deflections and loads for the three different full scale
loading cases. The summary table can be used as a quick reference to show the maximum
loads as well as expected deflections for individual loading cases.

Mechanical properties found for FRP grating supplied by Nepean Engineering:

Table 9 - FRP grating mechanical properties

Property Value Units

Density 1544 kg/m³

Glass Content 54 %

Resin Content 46 %

Flexural Strength 160.9 Mpa

Flexural Strain 0.03774 -

Flexural Modulus 9.89 GPa

Compressive Strength 69.8 GPa

85
Table 10 - Full scale summary

Test ID Deflection Max (mm) Load Max (kN) Support Length (mm) Panel Width (mm) Thickness (mm) Panel Weigth (kg)

LL 64.85 56.16 830 920 40 18.29

CL 49.14 33.37 830 920 40 18.29

OC 57.24 34.93 830 920 40 18.29

Where:

LL – Line load

CL – Centred concentrated load

OC – Off centre concentrated load

86
5.6 Recommendations & Future Work
The results of this research and investigation project show great possibilities for the further
use of FRP grating as a common material in the construction of walkways and boardwalks
throughout not only Australia, but the world. The results show that this particular type of
grating is best suited when a line load (or more commonly a uniformly distributed load) is
applied compared to the application of a point load, either in the centre or closer to the
supports. The added strength found closer to the supports allows a point load to be more
resistant to flexural failure than if it was applied directly in the centre of the panel.

This research is primarily a starting platform for the full investigation of FRP grating for its
suitability for use within the construction industry. Further investigation will need to be
undertaken to fully conclude the viability the product has within the industry. Further
investigation and research may be focused on a number of other testing procedures and
methods such as the effect of impact testing or how resistant to damage the grating is under
the influence of fire, UV, weather, and so forth. An economic investigation is also a necessity
to be able to fully evaluate if the use of FRP is not only mechanically viable but also feasible
on a cost based comparison, concentrating on the comparison between FRP grating and
traditional materials such as concrete, steel, timber etc.

Other testing which will be useful for the full understanding of FRP grating may be the
changing of certain parameters such as the thickness, fixings, or geometry of FRP grating
panels. Given that the FE analysis provided accurate results, a FEA simulation program may
be used within the industry for quick and accurate estimations for the behaviour of panels of
varying sizes without the need to test the individual sizes. The use of Creo as a modelling
software may not be the ideal software available globally however provided very accurate
results for what was required in this project. An alternate simulation software may be used for
higher accuracy or possibly even the development of a stand-alone program which predicts
the behaviour of FRP grating.

Once the additional testing is completed, table 10 can successfully be expanded to show the
predicted loads and deflections for a range of varying sized panels as well as the effects of
other testing methods such as impact or fire resistance. The completed table could then be
used as a quick reference for engineers to estimate the behaviour they may well expect when
using FRP grating as a form of walkway.

87
References
Bunsell & Renard, A.r & J, 2005. Fundamentals of Fibre Reinforced Composite Materials.
1st ed. France: Institute of Physics Publishing

Manthey, N W, 2009. Environmentally Friendly Natural Fibre Composites With QLD Based
Vegetable Oils. Undergrad. Toowoomba: University of Southern Queensland

Habib, A, & Akbar, I, 2014. Effect of accelerated ageing process on the strength of moulded
and pultruded GFRP gratings. In the 7th International Conference on FRP Composites in
Civil Engineering. Vancouver, Canada, August 20-22 2014

Van Den Einde, et al., 2003. Use of FRP Composites in Civil Structural Applications.
Construction and Building Materials, 17, 389-401

Fibreglass Warehouse. 2013. Your #1 Online Resource for Fibreglass Priducts. [ONLINE]
Avaliable at: https://www.fiberglasswarehouse.com/. [Accessed 22 May 14]

B.D.Agarwal, L.J.Broutman, K. Chandrashekhara, 2006. Analysis and Perfomance of Fibre


Composites. 2nd ed. John Wiley & Sons inc. Hoboken, New Jerseey

ZarnitsaTeam, 2011, video recording, Dura Composites Marina Deck production Process

Lubin, G, 1969. Handbook of fibreglass and advanced plastic composites. Springdale


Research & Developmental Center, Springdale, Pennsylvania

Mazumdar, S K. 2002, Composites manufacturing. CRC Press LLC

Meng, X. 2000, Advances in FRP moulded grating in Asia. RP Asia 2000 conference, Kuala
Lumpar, Malaysia

Islam, I & Aravinthan, T, 2010. Behaviour of structural fibre composite sandwich panels
under point load and uniformly distributed load. Composite Structures. Volume 93. Pages
206-215

Queensland Government, 2006. Composites in Industrial Plants.

Saafi, M. 2000. Design and fabrication of FRP grids for aerospace and civil engineering
applications. Journal of aerospace engineering. Pages 144-149

88
Awad, Z, Aravinthan, T, & Zhuge, Y. 2012. Experimental and numerical analysis of an
innovative GFRP sandwich floor panel under point load. Engineering Structures, Volume 41,
Pages 126-135

Ugale, VB et. al. 2013. Experimental studies on thin sandwich panels under impact and static
loading. Reinforced plastics and composites. Volume 32. Pages 420-434

Fan, H et. al. 2013. Flexural failure mechanisms of three-dimensional woven textile sandwich
panels: Experiments. Composite materials. Volume 48. Pages 609-620

Stocchi, A et. al. Manufacturing and testing of a sandwich panel honeycomb core reinforced
with natural-fibre fabrics. Materials and design. Volume 55. Pages 394-403

MTS Insight Electromechanical Testing Systems Brochure. [Accessed 22 April 6]

Pokharel, N. 2003. Behaviour and design of sandwich panels subjected to load buckling and
flexural wrinkling effects. A thesis for QUT ‘School of civil engineering, Queensland
University of Technology’.

Gibson, AG et. al. 2014. Post-fire integrity of composite gratings for offshore platforms.
Reinforced plastics and composites. Volume 33. Pages 543-555

Biddah, A. 2005. Structural reinforcement of bridge decks using pultruded GFRP grating.
Composite structures. Volume 74. Pages 80-88

Standards:

AS 4155 – 1993 - ‘Test methods for general access floors’

AS 4155.8 – 1993 - ‘Test methods for general access floors – test for 25mm x 25mm
concentrated load’

ISO 1172:1996 - ‘Textile-glass-reinforced plastics’

AS/NZS 1170 – 2002 – ‘Structural design actions’

89
Appendices
Appendix A – Test Results
Line Load

Line Load Deflection


70000

60000

50000

40000
Force (N)

LL - 1
30000
LL - 2
20000 LL - 3
10000

0
-10 0 10 20 30 40 50 60 70 80
-10000
Deflection (mm)

Line Load deflection results (3 records)

Test ID Deflection Max (mm) Load Max (kN) Support Length (mm) Width (mm)
LL-1 60.00 52.90 830 920
LL-2 64.67 53.90 830 920
LL-3 69.88 61.68 830 920
Average: 64.85 56.16

Summary of Line Load test result

90
Centred Concentrated

Centre Concentrated Load Deflection

40000

35000

30000

25000
Load (N)

20000 CL-1
CL-2
15000
CL-3
10000

5000

0
-10 0 10 20 30 40 50 60
-5000
Deflection (mm)

Centred concentrated load deflection results (3 records)

Test ID Deflection Max (mm) Load Max (kN) Support Length (mm) Width (mm)
CL-1 52.26 33.52 830 920
CL-2 49.00 34.94 830 1040
CL-3 46.15 31.64 830 920
Average: 49.14 33.37

Centred Concentrated load summary of test results

91
Off Centre Concentrated

Off centre concentrated load deflection results (3 records)

Test ID Deflection Max (mm) Load Max (kN) Support Length (mm) Width (mm)
OC-1 59.02 32.61 830 920
OC-2 55.78 36.41 830 920
OC-3 56.92 35.77 830 790
Average: 57.24 34.93

92
Flexure Test

12

10

8
Load (kN)

0
0 5 10 15 20 25
Deflection (mm)

FL-1 FL-2 FL-3 FL-4 FL-5

Flexure test deflection results (5 records)

Test ID Deflection Max (mm) Load Max (kN) m Flexural Stress (MPa) Flexural Modulus (Gpa) Deflection at Centre (mm) Flexural Strain
FL-1 23.73 10.425 0.46 186.28 3.08 34.96 0.04143
FL-2 20.73 8.587 0.375 153.44 2.51 35.32 0.04186
FL-3 17.87 8.09 0.446 144.56 2.99 27.98 0.03316
FL-4 18.72 7.33 0.376 130.98 2.52 30.07 0.03564
FL-5 19.68 10.59 0.529 189.23 3.54 30.88 0.03660
Average: 20.146 9.0044 0.4372 160.90 2.93 31.84 0.03774

93
Compressive load

1x1 Compressive Loading


120

100

80
Load (kN)

60 CL1x1-1
CL1x1-2
40 CL1x1-3

20

0
0 20 40 60 80 100 120

Play Time (s)

1x1 compressive load results (3 records)

2x2 Compressive Load


250

200

150
Load (kN)

CL2x2-1
CL2x2-2
100
CL2x2-3

50

0
0 10 20 30 40 50 60 70 80 90
Play Time (s)

2x2 compressive load results (3 records)

94
3x3 Compressive Loading
600

500

400
Load (kN)

300 CL3x3-1
CL3x3-2
200 CL3x3-3

100

0
0 20 40 60 80 100 120
Play Time (s)

3x3 compressive load results (3 records)

Test ID Length (mm) Width (mm) Web Thickness (mm) Total Area Under Loading (mm²) Max Load (kN) Compressive Strength (GPa)
C1x1 -1 45.94 45.86 7.98 1210.41 76.24 62.987
C1x1 -2 46.24 45.69 7.75 1184.67 99.31 83.833
C1x1 -3 45.26 46.02 7.78 1178.20 90.05 76.430

C2x2 -1 83.36 83.74 7.96 3420.09 231.37 67.649


C2x2 -2 83.78 83.92 7.90 3412.80 210.40 61.650
C2x2 -3 83.26 83.8 7.91 3401.22 234.35 68.902

C3x3 -1 122.22 122.44 7.84 6689.09 433.36 64.786


C3x3 -2 121.73 122.1 7.76 6605.00 481.41 72.886
C3x3 -3 122.2 121.85 7.91 6720.65 466.58 69.424

Average: 69.838

Compressive results summary

95
Burn Out Test

Colour Green Green Yellow Yellow Average


Orientation T L L T
m1 20.6133 21.3045 20.1489 23.3442
m2 43.6617 41.3639 44.1289 45.2253
m3 32.9777 32.5350 33.0859 34.8854
Sample Mass (g) 23.0484 20.0594 23.9800 21.8811 22.24
mglass (%) 53.6454 55.9862 53.9491 52.7451 54.08
mglass (g) 12.3644 11.2305 12.9370 11.5412 12.02
Volume (m^3) 1.4399E-05 1.44E-05 1.44E-05 1.44E-05
Density(kg/m^3) 1600.69983 1393.115 1665.399 1519.631 1544.71
Burn out test summary (4 records)

FEA Analysis

Flexure FEA deflection

96
Flexure FEA stress

Flexure FEA strain

97
Line Load FEA deflection

Centred concentrated load FEA deflection

98
Off centre concentrated load FEA deflection

99
Appendix B – Project Offer (Allan Manalo)

PROJECT: Behaviour of fibre composite walkways and gratings

AUTHOR: Allan Manalo

KEYWORDS: Civil engineering, fibre composites, walkways

ABSTRACT:

The Queensland Government has long been promoting the use of fibre composites in various
industry sectors. In the introductory guide they released in 2009 entitled “Composites in
Industrial Plants” (http://www.industry.qld.gov.au/documents/FibreComposites/Composites-
in-Industrial-Plants-Introductory-Guide.pdf), they identified walkways and drainage grating
as some of the many applications of fibre reinforced polymer (FRP) materials in the mining
and minerals and chemical processing plants due to its highly corrosion resistance.
Furthermore, this type of FRP structures also found many applications in marine and other
highly corrosive environments. Similar to other FRP materials used in many engineering
applications, the limited understanding on the overall behaviour and the lack of designers'
experience place FRP materials at a disadvantage when considered against traditional
construction materials. Thus, it is important to investigate the behaviour of these FRP
systems in order to gain a detailed understanding of their structural performance. When this is
achieved, essential technical information can be provided to Standards Australia to establish
design standards for this type of construction and could expedite the implementation of FRP
in civil, marine and mining infrastructure through a more functional and economical design.

In this study, the behaviour of FRP walkways and grating will be investigated. The test
procedures for this type of structures are documented in AS1657-1992 but the performance
requirements for a fibre composite walkway gratings are not available. In addition, the effect
of different restraint/fasteners on the structural behaviour of the FRP walkway under point
load test (corner and midspan) will be investigated.

Suitability: For On-Campus (and for External who are able to travel to CEEFC on a regular
basis) students.

100
Appendix C – Project Specification

University Of Southern Queensland

ENG4111/4112 Research Project

Project Specification

For: Lachlan Keith Nicol

Topic: Behaviour of fibre composite walkways and grating

Supervisor: Allan Manalo

Sponsorship: - Buchanan’s Advanced Composites

- Nepean Engineering & Innovation

Project Aim: The aim of this project is to analyse the behaviour of fibre composite grating
under static loading in order to gain an understanding on how this type of grating performs
structurally to help promote its use throughout the construction industry, primarily as a form
of walkway.

Programme:

1. Complete an extensive literature review in regards to FRP grating as well as testing and
analysis of materials.
2. Develop an in depth test schedule and programme for the testing procedures to fully
establish the aims and objectives that are needing to be accomplished over the course of
this research project.
3. Perform static load testing on a number of equal sized decking panels under line and point
loading.
4. Analyse and compare results between testing to gain a better understanding of the different
physical behaviours/properties of FRP grating.
5. Perform a Finite Element Analysis on the grating to help map out the stresses and strains
associated with the separate loading cases.
6. Analyse the behaviour of different sized FRP grating to evaluate the possible changes in
behaviours due to sizing variations.

101
7. Submit a final academic dissertation on the research and results from the entirety of this
project.

If time permits:

8. Analyse the effect on the overall behaviour of the deck system due to changes of geometry
i.e. change of thickness, depth, mesh sizing, or support conditions.

102

You might also like