Concrete Institute of Australia (2001) Conference Proceedings

QFR Academy, Live Fire Training Pad

Steel Fibre Reinforced Concrete Pavements
Mr Satyajit Datar, B.E. (Hons), MIEAust. CPEng, Design Manager, GHD Brisbane.

Introduction
GHD were the Project Managers and Design Consultant for a new $15M training Academy for the Queensland Fire and
Rescue Authority (QFRA) in Brisbane. The Academy began operation in September 2000 and is located on two sites:
● Exercise Yard
The Exercise Yard or Drill Ground is located on approximately 3.5 hectares of land leased from the Port of Brisbane
Corporation at Whyte Island, near the Brisbane River. This site provides training for:
– Technical Rescue which comprises Vertical Rescue, Confined Space, Trench and Urban Search and Rescue
(USAR) and Road Accident Rescue.
– Firefighting Operations and Management.
– Community Education and Basic First Response Training.
A 1600m2 Education Centre has been constructed and includes a 100 seat auditorium and multipurpose lecture rooms
for general training and incident command exercises.

● Live Fire Training Pad

The Live Fire Training Pad is 1.6 hectares of concrete pavement within the Caltex Refinery at Lytton. This site was
developed under a partnership agreement with Caltex and is located approximately 0.5 km from the Exercise Yard,
adjacent to mangrove wetlands.
The Pad contains a number of structures specifically built for burning. These structures are called Props or Simulations
and are used for both basic and advanced training in:
– Compartmental firefighting including fire phenomena such as flashover and backdraft.
– Specific fires such as aircraft, marine, petroleum and chemical incidents.
A Fire Investigation Research Unit (FIRU) is located here with a purpose built 2 storey, fully concrete Burnhouse for
both training and research into:
– Analysis and determination of fire behaviour.
– Product testing.
– External analysis.
– Community awareness programs.
The project outcome is to develop a facility which supports the training of emergency management. The physical
deliverables from the project support the training of emergency services groups and in particular QFRA, in enhancing
service delivery to Queensland communities.
This paper presents the structural design basis for the 1.6 hectare concrete pavement that is the Live Fire Training Pad.

Precedences In High Temperature Applications For Concrete Pavements

Prior to commencing design work for the Pad in early 1999, a review was undertaken of previous industry experience in
high temperature pavements. A brief summary of the review outcome is given below:

● Country Fire Authority, Victoria – Flammable Liquids Training Centre, Fiskville

This Centre contained a live fire training pad of plan dimensions, 70mx80m. It was designed by GHD in 1998 as a
continuously reinforced concrete slab with no joints crossing the 80m direction, and with tied construction joints at
approximately 10 metre centres in the 70m direction. The slab was heavily reinforced with a steel area of 0.67% of
gross concrete area in both directions.
Expansion joints were not used, to avoid the effects of fire and petrochemicals on the joint sealant. To minimise
volume changes due to entrapped water, the concrete was specified with a low water-cement ratio. No unconventional
materials were specified for the concrete mix. A cooling water system was used to prevent concrete spalling at the
surface. Cooling water pipes were cast into the top layer of the slab along the boundary of each prop and the crown of
the main access way to provide a thin film of water over the slab.

● Existing ‘Hot Fire Pad’ at the Caltex Refinery, Lytton (Brisbane)

An existing concrete pad, approximately 10m x 20m in area, supporting one gas fired prop, had been in use by Caltex
and QFRA for over 20 years. Engineering details of the pad construction were not available. Visual inspection of the
concrete surface revealed a moderate amount of cracking; but no more than would be expected on any typical external
reinforced concrete pavement of similar size and location.
The existing pad had survived intact, presumably due to the abundant use of fire fighting water for cooling.

● Refractory Materials
For many years, throughout the world, refractory materials have been used as linings for kilns, smelters, jet engine
testing cells and aircraft pavements. Steel fibre reinforced refractory toppings have been used for military aircraft

477
 Concrete Institute of Australia (2001) Conference Proceedings

pavements (Wu, G. 1987). Specialised concrete mixes such as air-cooled slag aggregate and high alumina (HA) cement
based concrete toppings and linings have been used for kilns and smelters.

Design Of The Pad - Macro Issues

The range of issues identified at the commencement of engineering design of the concrete Pad included the following:
● Integrity of the Concrete in High Temperatures
– mass loss
– spalling
– cracking/fracture
– stiffness loss
– temperature gradients
● Strength – Flexural and Compressive
– Design wheel loads;
● 22 tonne fire truck; 20,000 repetitions
● Austroads T44 load; 1000 repetitions
● Austroads W7 load; 100 repetitions
– fracture and impact toughness
● Durability and Exposure
– extreme temperatures
– corrosion of reinforcement
– spalling of concrete
– abrasion resistance (high pressures hoses)
– coastal environment (salt laden air)
– moisture and high humidity
– chemical attack (fuels)
– repeated and frequent drenching and submergence in water and fuels
● Serviceability
– slab curling (hogging)
– allowance for thermal expansion
– jointing and sealants
– shrinkage cracking
● Environmental
– prevention of leakage to subgrade (proximity to sensitive mangrove wetlands)
● Costs
– choice of structural system (continuously reinforced, prestressed or jointed)
– choice of materials - concrete (aggregate, cement, water cement ratio)
- reinforcement (prestressed steel, conventional steel, steel fibres)
- refractory toppings
- cooling water
● Program
– materials availability, lead times
– speed of construction
– structural system
– jointing
The principal issue that differentiated the design of the Pad from other external industrial pavements, was the repeated
exposure to extreme temperatures. In this regard, the first task was to ascertain the likely temperatures within the
concrete slab sections. This is discussed below.

Concrete Slab Temperature

The proposed live fire simulations and props were of numerous shapes, sizes and heights above the concrete surface
level. The gas and fuel discharge points were to be typically 25mm in diameter, and could be located anywhere from
several metres above to within centimetres of the surface. Compartment fires simulating conditions in rooms of
buildings would also create direct heat and flame contact with concrete floor and wall surfaces in the FIRU Burnhouse.
Some simulations were to be in the order of 10-20 metres in length, such as the aircraft and industrial props. Some were
to be limited to the size of small shrubs, and shaped like trees with branches made of gas fuelled pipes. Many props
were to be made from used shipping containers, some up to four storeys high.
The concrete surfaces were expected to experience instantaneous extreme temperature shocks in local hot spots, as well
as a high temperature rise across several square metres of area.
The likely instantaneous maximum air temperature above the pad surface was estimated to be in the order of 1200 C,
based on the literature review (Krampf, L. & Haksever, A. 1986); previous measurements undertaken by the QFRA for
liquid petroleum gas “training” fires, and historical “real life” fires. The air temperature was expected to fluctuate quite
dramatically under the influence of air currents and cooling waters.

478
 Concrete Institute of Australia (2001) Conference Proceedings

The literature review identified research based on laboratory tests of heated concrete beam specimens with internal
thermocouples. Likely concrete temperature profiles under fire durations of 15 to 30 minutes (which were expected to
be the typical durations of gas fire training exercises) and 60 to 90 minutes for comparment fires; were in the range
from 500 to 800 C at the surface exposed to the fire; to 100 to 400 C at the middle of sections of depths varying from
200 to 300mm (Krampf, L. & Haksever, A. 1986, and Desai, S.B. 1998).
It was recognised that under most training conditions, the fires, within minutes, would be subjected to high intensity
water drenching, particularly for gas and fuel fires; or low volume, high water surface area, mist sprays for
compartment fires for backdraft training. Both forms of fire fighting were expected to quickly reduce or limit the air
and concrete temperatures.
Supplementary cooling water applied prior to the commencement and for the full duration of each fire event, to both the
prop structures and the Pad surfaces, was also intended to be used for temperature reduction. Research papers
quantifying the temperature reductions within the concrete sections, due to cooling water, were not found.

Micro Issues
To address the above macro issues, principally the issue of high temperature, the starting point was a literature search
and review of relevant papers and discussions with researchers such as Dr H. Trinh Cao of the CSIRO, Sydney. A
number of material properties and micro-issues were identified as being critical to the design of the Pad for high
temperatures:
– Thermal Expansion – rate of change of volume.
– Thermal Conductivity – rate of flow of heat.
– Specific Heat – amount of heat required to raise temperature.
– Diffusivity – index of the facility with which concrete will undergo temperature change.
– Mass Loss – based on the absolute temperature or melting point of concrete, and would be related to the specific heat.
Thermal Expansion and Mass Loss were considered to be the most important criteria in the selection of a concrete mix.

● Thermal Expansion
– Differential expansion of the constituents of concrete; principally, cement and aggregate was acknowledged. Cement’s
coefficient of thermal expansion , (averages approximately 13 x 10-6/ C) is greater than that of aggregate. However
the aggregate occupies about 70-80% of the total volume of hardened concrete. The coefficient of thermal expansion of
concrete is almost proportional to that of the aggregate (Zoldners, N.G. 1971).
– Mineral composition of aggregate; principally the proportion of quartz present in the rocks. The following coefficients
were identified for typical rocks. (Zoldners, N.G. 1971) :
high quartz content (quartzite, sandstone) 12 x 10-6/ C;
medium quartz content (igneous rocks – granite, basalt) ranges from 5 to 8 x 10-6/ C;
low quartz content (limestone, carbonates, calcites) 5 x 10-6/ C.
– Other tests undertaken (Berwanger, C. & Sarkar, A.F. 1973) have shown the following ranges for :
limestone aggregates range as low as 2.5 x 10-6/ C;
sandstone to range from 6 to 12 x 10-6/ C;
quartzite to average at 10 to 13 x 10-6/ C;
Feldspars found in igneous rocks such as basalts, average at 6 to 7 x 10-6/ C.
Cement sand mortars exhibited average coefficients approximately 1 x 10-6/ C greater than the corresponding
aggregates alone, where the sand was made from the crushed aggregate.
It was concluded that limestone and feldspar aggregates had the best thermal properties, with limestone usually having a
lower .
– Thermal expansion at very high concrete temperatures (above 800 C) may increase sharply for reinforced concrete
compared to plain unreinforced concrete, because of the presence of steel which continues to expand (Lie T.T. & Kodur
V.K.R. 1996).

● Mass Loss
Mass Loss is not affected by the presence of reinforcement (steel fibre or bars) up to temperatures of 1000 C (Lie, T.T.
& Kodur, V.K.R. 1986). Mass loss is affected by the type of aggregate.
Carbonate aggregate mass loss is significant, up to 30%, above 600 C. (Kodur, V.K.R. and Sultan, M. 1998).
Thermogravimetric tests carried out by Queensland Cement Limited on carbonate, siltstone and basaltic aggregates, on
behalf of the Bechtel Minenco Joint Venture, Boyne Smelter Expansion Project at Gladstone, Queensland in 1995
(Basford, G. 2000), indicated the following:
– Siltstone with less than 7% calcite, showed negligible mass loss, of only 1% - 1.5%, at temperatures of about 600 -
750 C.
– Carbonate aggregate with (99% calcite) lost approximately 15% of its weight, between the temperature range of 650 C
- 850 C. This is consistent with the temperature at which limestone is heated to, in the production of lime.
– Basaltic aggregate underwent virtually no mass loss up to a temperature of 1000 C.

479
 Concrete Institute of Australia (2001) Conference Proceedings

● Thermal Properties of Reinforcement

The types of reinforcement considered were as follows:
– high strength cold drawn prestressing steel
– conventional cold worked deformed reinforcing bars and wire mesh
– steel fibres - slit sheet
- continuously deformed
- deformed or flared at ends
Prestressing steel contains a higher percentage of carbon and is usually cold drawn for high strength. When subjected
to high temperatures, it may become annealed and permanently lose a large part (up to 50%) of its strength and modulus
of elasticity.
Prestressing steel begins to lose its properties from temperatures of about 300 C. Conventional reinforcing steel
properties remain almost unchanged up to temperatures of about 600 C after which they lose strength and ductility
(Troxell, G.E. 1961).
The yielding temperature for steel fibres is higher than for bar reinforcement. (Kodur, V.K.R. & Sultan, M. 1998).

● Crack Growth and Propagation

The presence of reinforcement reduces crack widths and propagation. A fully prestressed concrete crack free slab under
each fire prop would have been considered ideal. However, due to the risks of loss of prestress and tensile strength,
prestressing was not adopted.
Conventional mesh and steel fibre reinforced concrete offer similar overall benefits in shrinkage crack control at normal
temperatures (Mindess, S. 1994). Fibres are distributed throughout the thickness of the slab including the top few
millimetres. Fibres were thus considered more advantageous than mesh reinforcement, in preventing surface cracking
and spalling under thermal and mechanical loads.
Fibre reinforced sections were also considered to be more ductile and better able to redistribute stresses beyond the
elastic range, and perhaps also increase the ultimate strain of the concrete, compared to mesh reinforced sections. (Lie,
T.T. & Kodur, V.K.R. 1996).
At high temperatures, up to 1000 C, fibres have been shown to increase the compressive strength of plain concrete.
(Lie, T.T. & Kodur, V.K.R. 1996).

● Reinforcement Bond Strength at High Temperatures

Publications regarding this issue were very limited in number. However, there was enough evidence based on pull-out
tests on heated cylinders, indicating that the bond strength to concrete of deformed bars is greater than for plain round
bars for all temperatures in the range 0 - 700 C. This included deformed prestressing steel (Diederichs U. & Schneider
U. 1981).
Amongst the various types of steel fibres available in Australia, it was difficult to distinguish which type had the highest
bond strength. Intuitively, fibres with deformations, either at their ends or along their full length, would be expected to
have a superior bond than plain fibres.

● Concrete Moisture Content

The effect of moisture content on spalling under high temperatures is a very complex issue. The cement matrix in
concrete contains water in many different forms; within capillary pores, adsorbed upon gel surfaces and
‘intercrystalline water’. There is evidence to suggest that spalling is more likely in very high strength concretes
(50 MPa or higher), as the dense cement matrix hinders the escape of vapours through capillary pores to the surface
(Gowripalan, N. et al, 1997).
There is evidence to suggest that low strength concretes with high water to cement ratios are prone to spalling due to the
higher free water contents.
Most spalling occurs at temperatures between 100 C and 250 C (Gowripalan, N. et al, 1997).
A reasonable choice of strength, given the complexities, was 32 to 40Mpa, with a low water to cement ratio.

Adopted Design
The adopted design attempted to address the macro and micro issues listed above, as follows:

● Integrity of the Concrete in High Temperatures

A carbonate based aggregate in lieu of basalt was originally specified because of its marginally lower coefficient of
thermal expansion. The risk of mass loss with carbonates was expected to be minimal due to the proposed use of
cooling water. However a feldspar based basaltic aggregate offered by the Contractor was accepted and sourced from
the Queensland Cement Limited, Glasshouse quarry in South East Queensland.
The coefficient of thermal expansion of feldspars was considered to be only marginally greater than that of carbonates,
and the feldspars offered a lower risk of mass loss. Suppliers were able to offer a more reliable source of feldspars than
carbonate aggregates.

480
 Concrete Institute of Australia (2001) Conference Proceedings

The application of a thin film of surface cooling water across each pad was considered to be an essential and economic
means of limiting concrete temperatures and protecting the joint sealants.

● Strength, Durability and Serviceability

A 40 MPa mix was adopted with 280kg/m3 of GP cement, 120kg/m3 of Fly Ash and a water : binder ratio of 0.44. A
water reducer was also used at a rate 2000 mL/m3. For mechanical loads, the slab thickness was determined to be
150mm with 30kg/m3 of steel fibre reinforcement, based on a CBR 35 subgrade. The steel fibres were 35mm long,
0.8mm diameter, cold drawn undulating wire, continuously deformed, with a minimum yield strength of 1000 MPa
(Combe, M. 2000).
A conventional, dowel jointed slab was adopted with joint spacings of approximately 6 to 8 metres in both directions, to
minimise thermal and shrinkage movements.
40MPa concrete with steel fibres was considered appropriate for the coastal environment and repeated exposure to
drenching and high pressure water abrasion.
Chemical attack of the slab and joints by the frequent presence of fuels, was considered to be minimised by operational
procedures such as frequent washdown cleaning.
The slab thickness could have been reduced with a higher cement and fibre content, however, it was considered that
150mm was the minimum thickness desirable for ease of dowel jointing and prevention of slab hogging or curling. A
fire-rated joint sealant was used, with the expectation, that sealant maintenance would be an ongoing responsibility of
the QFRA.
The FIRU Burnhouse reinforcement and concrete mix was similar to that for the pavements.

● Environmental
The subgrade of the Live Fire Training Pad was designed to be encapsulated within impermeable clays at the bottom
and sides. The original subgrade comprised mostly hydrocarbon sludge, which was completely excavated and removed
down to bedrock level, approximately 3 to 5 metres below finished surface levels.
A clay bund of 5 metres width at the berm, and founded in the bedrock was placed around the perimeter of the Pad. A
clay foundation was then placed across the site, over the bedrock, for a minimum thickness of 1 to 2 metres. A CBR10
material was placed over the clay. A 150mm thick subbase of CBR35 material was placed over the CBR10 material.
All layers were compacted to their maximum densities at their optimum moisture contents.
Polythene sheeting was placed over the subbase prior to placement of the concrete pavement. This sheeting was also
used as a bond breaker between the pavement and subbase, to facilitate shrinkage and expansion. Water quality
monitoring bores and spears were installed through the pavement at strategic locations, such as at the downstream end
of service trenches, to monitor any leakage of contaminated water through the pavement.
The entire Pad surface was bunded and drained to large collection tanks to contain stormwater and washdown runoff.
The collection tanks were designed with 3 compartments to isolate and retain for long periods, specific washdown
waters, such as that containing Aqueous Film Forming Foam, which was to be used for fire fighting.
During extreme rainfall occurrences, the Pad and collection tanks were designed to overflow to an adjacent storm and
wastewater retention basin within the Caltex Refinery.

● Costs and Program

A jointed steel fibre reinforced concrete slab was estimated to be similar in price to a jointed mesh reinforced concrete
slab. Both were estimated to be cheaper than a post-tensioned or continuously reinforced (with deformed bars) slab,
even allowing for the reduction of joints. The fibre reinforced slab was considered to be quicker to construct, due to the
absence of placement of mesh reinforcement.
The potential costs of future joint sealant replacement was considered a disadvantage for the jointed slab system.
However, due to capital cost restraints, the conventionally jointed system was more favourable than a continuously
reinforced system with fewer joints.
The costs of refractory toppings or brick paver overlays, would have exceeded the project budget if used on a large
scale over the 1.6 hectare Pad, but have not been ruled out from being introduced at selected locations, in the future, if
funds are available. HA cement was not adopted due to high costs, risks of delays in delivery, and risks of topping
delamination and strength degradation in the frequent presence of water.
Surface cooling water was able to be provided at a minimum cost, because extensive ‘fire’ water infrastructure and
reticulation was always going to be provided for fire fighting purposes. Cooling water branch lines were run to each
prop from the ring main.
The basaltic aggregate was readily available at the time from the concrete suppliers quarry, at virtually no cost penalty
to the project.

Performance Of The Pad

The concrete pavement was poured during the middle months of 2000. Live fire training on the Pad commenced in
April 2001. At the time of writing this paper (June 2001) the performance of the concrete pavement has been excellent.
There are a few localised areas of visible but fine cracking, generally limited to areas not subjected to fires, such as

481
 Concrete Institute of Australia (2001) Conference Proceedings

roadways and kerbs. There are localised areas of 1 to 2m2, where surface spalling of 1 to 2mm depth has occurred; but
only where cooling water was not used during commissioning trials.
The proximity of the fuel and gas discharge points to the slab surface, and the initial intensity of the fires upon ignition,
appear to be the main factors in causing spalling.
The joint sealant is performing well and without discoloration.
There is no evidence of adverse effects of thermal expansion, slab hogging, curling or problems due to lateral restraints
at the dowelled joint intersections.
Some minor chipping of kerbs and pitting of the slab surface has occurred, due to extreme impact loads, such as heavy
metallic objects and shipping containers being dropped or dragged over the surface.
Concrete section temperature profiles are not being measured at present, however, in the near future, thermocouples and
probes, may be installed by the QFRA for research in this matter. The QFRA would welcome any support or input
from research organisations or companies with an interest in the performance of concrete in fires.

References
Abrams, M.S. “Compressive Strength of Concrete at Temperatures to 1600 F” ACI Publication SP-25 “Temperature and Concrete”,
1971.
Basford, G. “Report on Thermal Properties of CSR Aggregate used in Concrete Production at Murarrie Plant”, CSR Construction
Materials, 2000.
Berwanger, C. & Sarkar, A.F. “Effect of Temperature and Age on Thermal Expansion and Modulus of Elasticity of Concrete”, ACI
Publication SP-39-1 “Behaviour of Concrete Under Extreme Temperatures”, 1973.
Combe, M. “Design of Steelcrete Steel Fibre Reinforced Fire Pad Slab”, Steelcrete Australia, Pty Ltd. February 2000.
Desai, S.B. “Design of Reinforced Concrete Beams Under Fire Exposure Conditions”, Magazine of Concrete Research, Vol. 50 No.
1, p75-83, Thomas Telford Services Ltd, 1998.
Diederichs, U. & Schneider, U. “Bond Strength at High Temperatures”, Technical University of Brunswick, Federal Republic of
Germany, Magazine of Concrete Research : Vol. 33, No. 115, June 1981.
Gowripalan, N., Salonga, P. & Dolden, C. “Residual Strength of High-Performance Concrete Subjected to High Temperatures” ACI
Publication SP 172-10, 1997.
Kodur, V.K.R. & Sultan, M. “Thermal Properties of High Strength Concrete at Elevated Temperatures”, ACI Publication SP 179-29,
1998.
Krampf, L. & Haksever, A. “Possibilities of Assessing the Temperatures Reached by Concrete Building Elements During a Fire”,
ACI Publication SP-92-7 “Evaluation and Repair of Fire Damage to Concrete”, 1986.
Lie, T.T. & Kodur, V.K.R. “Thermal and Mechanical Properties of Steel-Fibre-Reinforced Concrete at Elevated Temperatures”,
Canadian Journal of Civil Engineering, V23, p.511-517, 1996.
Mindess, S. “Fibre Reinforced Concrete – Myth and Reality”, Proceedings of an Engineering Foundation Conference, “Advances in
Cement and Concrete”, ASCE, 1994.
Troxell, G.E. “Fire Resistance of Prestressed Concrete’ ACI Publication SP-5, 1961.
Wu, G. “Steel Fiber Reinforced Heat Resistant Pavement” ACI Publication SP-105-18 “Fiber Reinforced Concrete Properties and
Applications”, 1987.
Zoldners, N.G. “Thermal Properties of Concrete Under Sustained Elevated Temperatures”, ACI Publication SP-25-1 “Temperature
and Concrete”, 1971.

482