SELF-COMPACTING CONCRETE: TECHNOLOGY

DEVELOPMENT AND MEASUREMENT OF WORKABILITY

J.P. Jooste1 and Y. Ballim2

1
Department of Civil Engineering, Technikon Witwatersrand, PO Box 17011, Doornfontein,
2028, Johannesburg, South Africa. Tel: +27 11 406 8121. E-mail: pjooste@twr.ac.za
2
School of Civil and Environmental Engineering, University of the Witwatersrand,
Private Bag 3, Wits, 2050, Johannesburg, South Africa. Tel: +27 11 717 7103.
E-mail: ballim@civil.wits.ac.za

ABSTRACT

Self-compacting concrete (SCC) is becoming a popular form of concrete usage in a range

of applications throughout the world. This new concrete type has also found application
in South Africa and has been used with great success in contracts such as the Nelson
Mandela Bridge in Johannesburg. The first formal application of SCC occurred in Japan
around 1986. This initiative was motivated by a shortage of skilled workers to place
concrete at the job sites and the resultant inadequate compaction and finishing of the
concrete. The response was to develop a concrete that flowed easily around obstacles
into all the corners of the formwork without the need for compaction. This made the
construction of concrete structures much quicker with fewer skilled construction
workers.

This paper considers the development of the technology and use of SCC internationally,
addressing the benefits as well as the disadvantages. Importantly, the paper aims to
highlight the opportunities for using SCC in South Africa.

A total of nine SCC mixes were tested using the Tattersall two point tester, the L-box, the
V-funnel and the slump flow tests to define the flowability and self-compactability. The
three aggregate types used where Jukskei granite, Eikenhof andesite and Olifantsfontein
dolomite. All the mixes had the same cement (Cem II AM 42.5) and cement to water ratio.
The admixtures and the percentage fly-ash were varied to determine the effect.

From the test results it was found that SCC can be made using South African materials,
but much more testing and research is needed. Of the aggregates used in this project the
Jukskei granite preformed the best, but required more superplasticizer and cement than
the Olifantsfontein dolomite. The flowability and self-compactability of all the mixes was
enhanced with the use of a bigger percentage fly-ash. The results from the Tattersall two
point tester gave additional information about the flowability which correlates well with
the L-box, the V-funnel and the slump flow tests.

Paper presented at the Conference: Developing Concrete to Serve Practical Needs Hosted by:
ISBN Number: 1-920-01717-8 The Industrial Development Engineers Association (IDEA)
13 – 14th October 2004, Midrand, South Africa. The Cement and Concrete Institute (C&CI)
CD produced by: Document Transformation Technologies 207 The Concrete Society of Southern Africa (CSSA)
1. INTRODUCTION

Self compacting concrete (SCC) is a specialized concrete designed to flow freely around
obstacles, completely fill formwork and enclose all reinforcing bars without segregation or
bleeding.[1] As the name indicates, this concrete type requires no external consolidation effort
while still fulfilling all the requirements of conventional concrete.

The idea of a concrete that flows and compacts under self weight only, originated in Japan.
Thereafter, the use of SCC spread through Asia into Europe and many parts of the world
including South Africa. The development and application of SCC technology across the world is
rapidly increasing.

The main benefits in using SCC are:

• Reduced energy requirement in compacting and finishing concrete
• Reduced noise in built-up areas
• Reduced construction time
• Improved final product performance
• The opportunity for innovative construction systems

As a contribution to the development of SCC technology in South Africa, this paper presents
details of a test programme that was undertaken to measure the opportunities and potential for
using local concrete-making materials to produce SCC. The paper presents a brief review of the
development of this technology as well as the basic principles and measurements of rheology of
these materials. In particular, the Tattersall two-point test apparatus was used and the principles
of its operation are explained in some detail to clarify the measuring process and interpretation
of the results.

To determine the properties of fresh SCC made from South African materials, nine different
mixes were tested. The flowability and filling ability of these mixes were measured with the
slump flow, L-box and V-funnel tests. The through-flow ability and segregation resistance are
assessed with the L-box and V-funnel tests. The Tattersall two point test apparatus was used to
determine the basic rheological parameters of the concretes.

The work presented in this paper represents the initial phase of a larger project aimed at
characterising the properties of SCC in South Africa. . The results are based on concretes which
have not yet been perfected as SCC concretes. With a better understanding and knowledge,
SCC could be used more extensively to create durable structures in less time and with less
energy input.

2. DEVELOPMENT AND APPLICATIONS

SCC was first developed in the 1980s by Okamura at the Tokyo University and its use has
gradually increased.[2] From Japan, the use of SCC spread through Asia and, by the early
1990’s, it was also being used in Europe. The RILEM First International Symposium on SCC
was held in Stockholm, Sweden in 1999. Two years later the second symposium was held in
Kochi, Japan and the third was held in 2003 in Reykjavik, Iceland. At this last symposium there
were 108 contributions from 26 countries. In North America the use of SCC grew from an
insignificant amount in the year 2000 to more than a million cubic meters in total at the end of
2002. Even though much research has been done across the world, SCC is still in its infancy
stage and much more research are needed.[3]

208
One of the first big projects undertaken in Japan using SCC was the anchorage (83m long, 63m
wide and approximately 45m high) of the Akashi-Kaikyo Bridge. This project is a very good
example where the use of SCC contributed to reducing the total construction time from 30 to 24
months. Another project where the use of SCC reduced the construction time from 22 to 18
months was the 0.8m thick wall of the LNG (liquid natural gas) tank for the Osaka Gas
Company.[1] More recent applications of SCC in Japan are lattice work (thin ribs), casting
without a pump (discharging concrete from the truck and allow to flow freely to fill the formwork)
and tunnel linings. SCC is used in lattice work because ordinary concrete cannot be vibrated in
this manufacturing process. To prevent cold joints in tunnel linings, SCC is used because it
limits bleeding or laitance at joints.[2]

Sweden started to develop SCC in 1993 with a project where walls were cast using different
materials as fillers in the mix designs. In 1998 a bridge was constructed using SCC. This was
the first bridge outside Japan where SCC was used for the whole structure.[1] Since then, SCC
has been used in monolithic frame bridges, box tunnel monoliths, rock linings, tunnel entrances,
headwalls, foundations and frame supports. The current use of SCC in the pre-cast and ready
mix concrete industry is about 10% of the total concrete use.[4]

The development of SCC is particularly favored in the precast concrete industry. Some precast
concrete producers in the Netherlands only use SCC in the manufacture of their products.[5]
Because of this extensive use of SCC, much experience has been gained and SCC is now used
in pre-cast slabs, beams, walls, columns, arches and bridge elements. SCC has also been used
in situ but only in special cases. The first major project was the façade of the National Theater in
The Hague where only SCC could be used to fill the tiny ribs (8mm deep). In some tunnel walls
SCC was used because of congested reinforcement and the possibility of remote casting
techniques. At the Rotterdam Zoo the heavily reinforced walls of a large fish pond was done
with SCC to ensure a homogeneous watertight structure. The design and shape of the bridge
piers for the “South Tangent” traffic connection between Haarlem and Amsterdam was of such a
nature that only SCC could be used. In this project 1800 m3 of SCC was used. The most resent
development in the Netherlands is self-compacting fiber reinforced concrete. Self-compacting
fiber reinforced concrete is used to produce floor elements that is thinner and lighter than
conventional elements[5].

Further examples of SCC applications are the steel form columns at the Toronto International
Airport and the outrigger columns at Wall Centre in Vancouver (North America). A more
interesting application was houses in Houston where the exterior walls and slabs were cast
monolithically out of SCC. The walls are textured and stained on the outside to resemble brick
and have a polystyrene foam core for insulation. These houses can withstand tornados and
hurricane winds in excess of 218 km per hour.[6]

SCC has also been used in South Africa in various projects with great success. The best
example is the use of SCC in the pylons of the Nelson Mandela bridge. These pylons are
constructed from 20mm and 40mm steel plate rolled to produce 1,35m diameter steel pipes
which was filled with 50 Mpa concrete to provide the required stiffness. The pylons are
respectively 31,1m and 43,9m high which create difficulty with concrete lifting and placing due
the free fall limits. Mechanical vibration was impossible due to limited access and external
vibration was inappropriate because of the large amount of energy needed to overcome the
pylon inertia. To overcome these problems it was decided to pump SCC from the bottom. Since
this process couldn’t be stopped at any stage it was required that 60% of the required concrete
to be on site and the balance dispatched and en route to site before the pumping process could
commence.[7]
209
3. BENEFITS OF SCC

The most important benefit when using SCC is that no compaction of the fresh concrete is
required. This leads to reduced energy requirement in the placing and finishing of the concrete.
Because placing is quicker and easier, the construction time is reduced and workers can be
used more effectively. The high flowability of SCC makes alternative placing methods possible,
like pumping the concrete continuously from the bottom of the structure or from remote areas.

The high flowability and elimination of the need for compaction make the use of special designs
and shapes possible. With traditional concrete, designs were restricted to shapes where
concrete can be placed manually and where compaction equipment can reach. The Science
Centre in Wolfsburg, Germany,[8] the façade of the National Theater in The Hague[5] and the
pylons of the Nelson Mandela bridge are examples where the use of ordinary concrete would
not have been possible. SCC lends itself to creative shapes and innovative construction
systems. Designs with very congested reinforcing are also acceptable, since SCC can flow
around these and external compaction is not required.

With the reduction of the noise levels (about 93 dB when compacting ordinary concrete) the
working environment is safer and the noise is reduced in built-up areas. When using SCC the
noise level can be brought done well below 80 dB. Intensities higher than 80 dB can cause
deafness, stress and fatigue.[5] With lower noise no ear protection is needed and communication
on site is easier. Vibration above 0,25 m/s2 cause pain and stiffness in limbs, back and neck.[5] A
more serious ailment caused by continuously using the poker vibrator (vibration levels from 0,75
to 4 m/s2) is known as “white fingers” which affects the blood circulation of the vibrator
operator.[8]

With well compacted concrete the possibility of air voids are reduced which increase the
strength and density of the concrete. The bond between the concrete and the reinforcing is
improved and there is a reduced chance of bleed water lenses beneath reinforcing and
aggregate.

The off shutter finish when using SCC is also very good. The changes of honeycombing and
blow holes are very slim. The use of admixtures assures thorough mixing since all the cement
particles are dispersed evenly throughout the mix, resulting in a more homogeneous concrete.

The properties of SCC are well suited to produce good quality precast elements reducing
energy consumption in the production process. The energy required is not just the power to
operate the plant, but also labour and equipment efficiency. Cycle time of the moulds is also
shorter because the admixtures used in the mix accelerate the hydration process which
accelerates strength development. There is also less wear and maintenance on the mixing
equipment.[9]

4. DISADVANTAGES

The biggest disadvantage in using self compacting concrete is the cost involved to make this
type of concrete. The material cost is higher since admixtures must be used. The aggregate
also needs to be smaller and a size that in not commonly used is required. The mixture requires
a large percentage of fines and filler material. SCC is also sensitive to variation the aggregate
and this needs to be well controlled for consistent quality and size. The initial cost to set up the
mixing plant can also be significant.

210
The material sensitivity of SCC means that strict quality control is necessary at the batching and
mixing operation. The material used in the mix need to conform to a very narrow specification.
Mixer operators must be well trained and always aware of the sensitivity of this product.

Furthermore, special formwork is required when using SCC. The formwork must be stronger to
support the concrete at early ages since form pressure is higher than with ordinary concrete.
Formwork needs to be near watertight to prevent loss of fines from the mixture.

5. MIX DESIGN REQUIREMENTS

More care is required in the design of self compacting concrete than for normal concrete. The
factors influencing the design are flowability, filling ability, flow-through ability and segregation
resistance.[10] These factors are governed by the aggregate, filler quantity, water to cement ratio
(w:c) and the use of admixtures. Okamura suggests that the coarse aggregate content should
be fixed at 50% of the solid volume and the fine aggregate content to be fixed at 40% of the
mortar volume.[4] The w:c also has an influence on the design, since this ratio still determines
strength and needs to be appropriately selected. SCC is very sensitive to changes in the water
content and too much water will render the mixture unfit for use. The reason for this sensitivity is
that the supporting effect of individual grains is lost in this high-fines content material.[11]

Superplasticisers disperse the cement particles and allow more cement particles to be covered
with water. The surface area of the separated cement particles are smaller and therefore require
less water and still provide very good flowability. The dosage of superplasticizer depends on the
aggregate and cementitious type and is required to ensure self-compactability.

The choice of aggregate plays an important role in the mix design of the concrete mix and
should therefore be consistent. A maximum stone size of 19 mm is recommended to minimize
blocking. The shape of the stone should be round and chunky. This shape is easier to cover
with the paste matrix and also flows better. A high percentage of fines is used to increase the
volume of the paste matrix and improve flowability. In Europe, the filler material used is usually
powdered limestone. Limestone is not as freely available in South Africa but can be substituted
with fly-ash. Fly-ash also contributes to the flowability of the mix due to the spherical shape of
the particles.

6. PRINCIPLES AND MEASUREMENTS OF RHEOLOGY

Since the flowability of self-compacting concrete is so important, relevant tests are needed to
describe this property. It is appropriate to describe flowability through the concept of rheology.
With a concrete that is sensitive to the proportions of the ingredients, like self compacting
concrete, at least two rheological parameters, shear strength and plastic viscosity are needed to
describe the workability.[12] The slump flow test can also be used but, as with the normal slump
test, only one parameter of concrete behaviour is described.

Rheology is the science used to describe the flow and deformation of a material. To measure
the rheology of concrete, the two-point test, a dynamic test, was developed to measure shear
resistance at two shear deformation rates. The yield stress (τ0) and plastic viscosity (µ) can
therefore be measured and used in the Bingham equation (Equation 1) to determine the shear
resistance:[13]

211
 τ = τ0 + µγ (1)

where τ is the shear stress, τ0 is the yield stress, µ is the plastic viscosity and γ is the shear
strain rate. [13]

The rheology of concrete is best measured with the use of a rheometer. There are a number of
rheometers available around the world with significantly different design and operation
parameters. The five most commonly used rheometers were compared at the Laboratoire
Central des Ponts et Chaussees (LCPC) facility in Nantes, France in 2000. The rheometers
used in this comparison were the BML from Iceland, the BTRHEOM and CEMAGREF-IMG
coaxial rheometer from France, the IBB from Canada and the Tattersall Two-Point tester from
UK. All these rheometers are designed to effectively describe the rheology of concrete[14].

The only rheometer currently available in South Africa suitable to test concrete rheology is a
Tattersall Two-Point tester which was donated for research purposes by the Cement and
Concrete Institute of South Africa. For this reason the Tattersall Two-Point tester was used in
this research project to measure and define the rheological properties of the concrete tested.

The Tattersall Two-Point tester originated because of impracticalities encountered with the
original coaxial cylinders viscometers. The coaxial cylinders viscometers had to be very large to
satisfy the requirements for measuring the rheology of a material like concrete. The
requirements state that the gap between the two cylinders should be 10 times the size of the
largest particle and the ratio of the outer to the inner cylinder radius less than 1,2 to ensure
reliable measurements.[15] Tattersall initially used a Hobart food mixer to measure the electrical
power input when mixing a standard quantity (25 kg) of concrete in the mixing bowl and also
when empty. These measurements were taken at the three rotation speeds of the stirring hook.
The difference between the two power inputs (P) is divided by the speed (N) to determine the
torque (T). The torque was plotted against time and this gave a linear or very near linear
relationship. This concept was developed further and a bigger apparatus with special impellers
and reduction gearboxes was created to produce more effective results[13].

Even though the Tattersall Two-Point tester has been used in more than 20 industrial and
research laboratories and a number of construction sites around the world, there is no standard
test procedure.[16] It is therefore necessary to explain the testing and calibration procedure as
well as the calculation of the results.

Before testing can be done the apparatus must be set up to be level, the reduction gearbox and
the hydraulic unit should be filled with the appropriate oil and the hydraulic unit should be free of
entrapped air. The impeller is then allowed to turn freely for 30 minutes at the recommended
speed of 3 rev/s to allow the oil in the drive unit to reach equilibrium temperature.

After the warm up, the procedures for testing is as follows:[17]

1. The bowl is raised with the rack and pinion to the working position where the clearance
between the impeller and the bottom of the bowl is 60 mm.
2. Fill the bowl gradually with concrete to about 75 mm from the rim while the impeller rotates
at approximately 0.7 rev/s.
3. Increase the speed to approximately1.3 rev/s and allow the pressure to stabiles.
4. Read the speed by tachometer.
5. Read the pressure gauge and record the average position of needle for the small
oscillations. Ignore the large oscillations due to trapping of aggregate.

212
6. Repeat procedures 4 and 5 at speeds of 1.2, 1.0, 0.9, 0.7, 0.5, and 0.3 rev/s.
7. Disconnect the impeller and record the idling pressures at the same speeds used in the
measurements on the concrete.

The results may be plotted on a graph relating net pressure to speed and the intercept and
reciprocal slope is then determined from this plot. Since it has been shown that the flow
properties of concrete conform to the Bingham model[17], the flow curves are taken as linear, the
intercept on the torque axis and the reciprocal slope can be calculated using the least squares
method.[17] The values of g and h were calculated by multiplying the intercept and slope by the
torque/pressure calibration constant. To determine the torque/pressure calibration constant, a
plummer block, fixed so that the impeller drive shaft passes through it, is used. The desired
level of frictional force is obtained by varying the tightness of the plummer block bolts. A lever is
attached to the plummer block base on the one side and a spring balance on the other side
fixed to the frame to measure the retarding torque.[13] The least squares method is also used to
determine the slope of the line representing the relation between the spring balance readings
and the pressure. The slope is then multiplied by the length of the lever and gravitational
acceleration to calculate the calibration constant.[18] (Because of time constraints the torque
calibration was not done and a calibration constant of 0,272 Nm/pressure unit, taken from
similar apparatus[17], was used to interpret test results.) Table 1 shows examples of calculations
for g, h, τ0 and µ from experimental data obtained from the two-point apparatus.

Table 1. Example of experimental data obtained from the two-point apparatus.

TATTERSALL READINGS RESULTS

Calibration constant: 0.272 Nm/pressure unit Yield Plastic
Mix Speed Pressure Torque Speed Speed X value viscosity
Number (rev/s) Total Idling Net (Nm) SQ Torque g h G K το µ
1.45 48 14.5 33.5 9.11 2.10 13.21 1.5 4.63 0.0587 7.1 181.4 78.88
MIX 4 1.3 41 14.2 26.8 7.29 1.69 9.48
(EIK01) 1.1 37 13.8 23.2 6.31 1.21 6.94
0.88 32 13.2 18.8 5.11 0.77 4.50
0.68 28 12.5 15.5 4.22 0.46 2.87
0.5 26 11.8 14.2 3.86 0.25 1.93
0.24 23 11 12 3.26 0.06 0.78

An experimental error will be included since g and h were obtained from the best straight line of
the experimental points. The correlation coefficient for the seven measurements shown in Table
1 was calculated as 0.99.

The flow properties of concrete are represented by the following equation:

T = g + hN (2)

Where T is the torque at speed N, g is a measure of the yield value and h is a measure of
plastic viscosity. [17] To relate g and h to the Bingham parameters, yield stress (τ0) and plastic
viscosity (µ) in fundamental units, the Tattersall two point tester had to be calibrated. For the
calibration, a silicone oil with a viscosity of 28 Pas at 22°C can be used as a Newtonian fluid
and an aqueous solution of carboxymethyl cellulose can be used as the power law fluid.[14]

213
The viscosity (η) of the Newtonian fluid is then obtained with the use of a Brookfield viscometer.
This value, as well as the linear relationship between T and N can be used to determine the
constant of proportionality (G) in:

T / N = Gη (3)

The power law fluid is used to determine constant K by comparing the power law relationship
between T and N in:

T = pNq (4)

obtained in the two point apparatus with the flow curve determined in the Brookfield viscometer
as:

τ = rγs (5)

(p,q and r are constants and s is the slope of the log stress against log shear rate line[13] ). If the
range of shear rates in the Two-Point test are similar to those in the viscometer then:

K = (p / rG)1/(s-1) (6)

It is assumed that there is an average effective shear rate in the apparatus and this is given
by[17]

γave = KN (7)

With the values of K and G, g and h can be expressed in terms of τ0 and µ using

τ0 = (K/G)g (8)

µ = (1/G)h (9)

Due to time constraints and difficulties to obtain the correct fluids for the calibration, the
calibration of the Tattersall two-point tester has not yet been successfully completed and values
for G and K were taken as 0.0587 m3 and 7.1 respectively.

According to Wallevik[19] SCC should have a very low yield value if the plastic viscosity is over
70 Pa.s and if the plastic viscosity is around 40 Pa.s the yield value must be significant. In
countries like Japan, Sweden and The Netherlands the plastic viscosity is very high due to a
high water-powder ratio and the yield value is close to zero. In Norway and Switzerland, where
the aggregate is of good quality, the plastic viscosity is very low and the yield value between
20 and 40 Pa.[19]. Table 2 presents a scan of the range of rheology parameters for concretes in
a range of countries.

214
 Table 2. Estimation of typical SCC properties in different countries.[19].

Power Water Yield value Plastic viscosity

(kg/m3) (kg/m3) (Pa) (Pa.s)
Sweden >550 180 0-30 50-100
The Netherlands >550 190 0-10 60-120
Japan >550 170 0-30 50-120
France ? ? 0-10 >60
Switzerland <450 200 0-50 10-20
Norway <450 170 10-50 30-45
Iceland <450 180 10-50 20-40
Denmark <400 160 30-60 <40
UK >500 210 10-50 50-80
Germany >500 180 0-10 60-90
US >500 190 0-20 40-120

7. LABORATORY PROCEDURE

7.1. Materials and Concrete Mixtures

In the development of a local SCC mix, aggregates from the Gauteng area were used to
determine the most suitable type as well as the best blend of aggregates. The aggregate types
used to compare and determine the best mix were granite from the Jukskei quarry, andesite
from Eikenhof and dolomite from Olifantsfontein. To achieve the required flowability with no
segregation, superplasticisers and high-range water-reducing admixtures were used at
appropriate dosages in relation to the demands of the mix design. The method of mix design
used in this project was adapted from the method of mix design used for ordinary concrete in
South Africa.

Table 3 shows the mixture proportions of the different concretes subjected to rheology testing.

Table 3. Mix proportions of the concretes assessed.

MIX 1 (JUK00) MIX 2 (EIK00) MIX 3 (EIK010)
Material kg/m³ Material kg/m³ Material kg/m³
Cem II AM 42.5 204 Cem II AM 42.5 196 Cem II AM 42.5 196
Fly-ash 87 Fly-ash 84 Fly-ash 84
Jukskei Crusher 698 Eikenhof Crusher 804 Eikenhof Crusher 804
Bothma Filler sand 372 Bothma Filler sand 400 Bothma Filler sand 400
13mm Jukskei 810 13mm Eikenhof 810 13mm Eikenhof 810
Water 190 Water 183 Water 183
Super plasticizer 5.238 Super plasticizer 5.275 Super plasticizer 5.275
Viscosity modifier Viscosity modifier Viscosity modifier 0.06
W:C 0.65 W:C 0.65 W:C 0.65

215
 Table 3. Continued.
MIX 4 (EIK01) MIX 5 (JUK01) MIX 6 (OLI01)
Material kg/m³ Material kg/m³ Material kg/m³
Cem II AM 42.5 196 Cem II AM 42.5 204 Cem II AM 42.5 183
Fly-ash 84 Fly-ash 87 Fly-ash 79
Eikenhof Crusher 804 Jukskei Crusher 698 Olifants Crusher 823
Bothma Filler sand 400 Bothma Filler sand 372 Bothma Filler sand 409
13mm Eikenhof 810 13mm Jukskei 810 13mm Olifants 835
Water 183 Water 190 Water 171
Super plasticizer 5.275 Super plasticizer 5.238 Super plasticizer 4.716
Viscosity modifier 0.03 Viscosity modifier 0.03 Viscosity modifier 0.03
W:C 0.65 W:C 0.65 W:C 0.65
MIX 7 (EIK02) MIX 8 (JUK02) MIX 9 (OLI02)
Material kg/m³ Material kg/m³ Material kg/m³
Cem II AM 42.5 196 Cem II AM 42.5 204 Cem II AM 42.5 183
Fly-ash 138 Fly-ash 143 Fly-ash 130
Eikenhof Crusher 804 Jukskei Crusher 698 Olifants Crusher 823
Bothma Filler sand 400 Bothma Filler sand 372 Bothma Filler sand 354
13mm Eikenhof 810 13mm Jukskei 810 13mm Olifants 835
Water 183 Water 190 Water 171
Super plasticizer 6.293 Super plasticizer 6.246 Super plasticizer 5.634
Viscosity modifier 0.03 Viscosity modifier 0.03 Viscosity modifier 0.03
W:C 0.65 W:C 0.65 W:C 0.65

7.2 Test Methods

7.2.1 The tattersall two point test

The test method is described earlier in this paper. The same method was used with minor
changes to the speeds as indicated in Table 1. This test is used to measure shear resistance at
two shear deformation rates. The yield stress (τ0) and plastic viscosity (µ) can therefore be
measured and used in the Bingham equation (Equation 1) to determine the shear resistance.

Figure 1. Tattersall two point tester.

216
7.2.2 The slump flow

The slump flow test is used to evaluate the flowability, deformability and stability of SCC.
Included in this test is the T50 value which describes the viscosity. A normal slump cone is used
in this test. The test is usually performed on a flat horizontal base plate with a 500mm φ circle
drawn on the surface for the measurement of the T50 time.

The test procedure is as follows:[16]

• Dampen the interior of the slump cone and the surface of the base plate.
• Make sure the base plate is flat and horizontal and place the cone in the centre of 500mm
circle, on the base plate.
• While pressing the cone down firmly, fill the cone continuously with SCC to the top, without
consolidating the concrete and level off.
• Remove the slump cone immediately and perpendicular to the base plate, starting the
stopwatch as the lifting begins.
• Record the time the concrete takes to reach the 500mm φ circle (T50).
• Measure the final diameter of the concrete as soon as it stops flowing. Assess the concrete
for segregation and bleeding.

Figure 2. The Slump flow test[1].

7.2.3 The v-funnel test

This test is used evaluate the through-flow ability and segregation resistance of SCC.

The following procedure is used to determine the flow speed:[16]

• Dampen the interior of the V-Funnel and place on a level surface with the gate close and a
container placed underneath the opening.
• Fill the V-Funnel continuously with SCC to the top, without consolidating the concrete.
• Wait one minute for the concrete to settle and observe for segregation and bleeding.
• Open the gate and start the stopwatch simultaneously.
• Record the time when the concrete has flowed out of the V-Funnel (flow time = t0)
• If blocking occurs, this indicates instability of the SCC mix.
• Repeat the test a few times with separate samples within 5 minutes and take the average.
• If there is segregation resistance, repeat the test procedure but wait 5 minutes before
opening the gate. This flow time is recorded as t5.

217
 Figure 3. V-Funnel.

7.2.4 The L-Box test

The L-Box test was based on the L-Flow test developed in Japan for underwater concrete.
Peterson developed the L-Box test to assess the through-flow ability and filling ability of SCC.[16]
The L-shaped box is 700mm long and 600mm high with reinforcing bars placed in front of the
gate.

The test procedure is as follows:

• Clean and dampen the interior of the L-Box and place on a level surface with the gate close.
• Fill vertical section of the L-Box continuously with SCC to the top, without consolidating the
concrete.
• Wait one minute for the concrete to settle and observe for segregation and bleeding.
• Open the gate and start the stopwatch simultaneously, allowing the concrete to flow into the
horizontal part.
• Measure the time it take the concrete to reach the 200mm (T20) as well as the 400mm (T40)
markings.
• Measure the H1 and H2 distances as soon as the concrete stops flowing.

The blocking ratio, H2/H1 should be between 0.8 and 1.0

218
 Figure 4. L-Box. Figure 5. L-Box dimensions.

8. RESULTS

Table 4 presents the test results obtained using the four test methods on the concrete mixes
listed in Table 3. The results were obtained in trial-and-error fashion in that, weaknesses or
errors identified in one series of tests were corrected or improved in the next series of tests.

Table 4. Test results.

Tattersall results Slump flow V-Funnel L-Box
Final Flow
MIX T50 diameter Flow speed T20 T40 H1 H2
Number g h το µ (sec) (mm) time (m/s) (sec) (sec) (mm) (mm) H2/H1
MIX 1
(JUK00) 2.5 0.8 304.1 14.1 4 680 5.44 0.38 3 9 160 50 0.3
MIX 2
(EIK00) 1.2 2.0 141.5 34.5 2.3 620 6 0.34 1.6 2.8 140 60 0.4
MIX 3
(EIK010) 3.4 5.2 412.3 87.8 4 580 40 0.05 2 4 240 43 0.18
MIX 4
(EIK01) 1.5 4.6 184.3 78.9 4 680 52 0.04 1 4 120 70 0.58
MIX 5
(JUK01) 0.4 5.6 43.2 95.2 3 690 13 0.16 2 7 500 50 0.1
MIX 6
(OLI01) 1.5 3.8 181.8 64.8 4 680 10 0.21 1.3 3 100 70 0.7
MIX 7
(EIK02) 1.4 3.1 171.7 53.5 5 620 11 0.18 1 3 120 70 0.58
MIX 8
(JUK02) 0.2 4.7 24.7 80.1 3 715 25 0.08 1 4 90 80 0.9
MIX 9
(OLI02) 1.1 2.9 129.5 50.0 4 610 6 0.34 1 2 120 70 0.58

219
In interpreting the results shown in Table 4, it is useful to refer to the “Specifications and
guidelines for Self-compacting concrete”, which require that the materials conform to the range
values shown in Table 5 [10].

Table 5. Specifications for workability tests on SCC[10].

Slump flow – Abrams 650 – 800 mm

T50 slump flow 2 -5 sec
V Funnel 8 – 12 sec
L Box H2 / H1 = 0.8 – 1.0

Mixes 1, 2 and 3 were the very first mixes to be tested and was used as a basis to start from
and to check the admixture dosage. A comparison with Table 5 shows that these mixes don’t
conform to the performance specifications for a good SCC mix. The yield values are too high
and the plastic viscosity too low, which is confirmed by the workability tests. In the observation
of the mixes, it was found that there was severe segregation, bleeding and blocking in the
L-box. This problem was rectified with the change in the admixture type and dosage and also
the inclusion of a viscosity modifier to make the mix more cohesive. The dosage of viscosity
modifier included in mix 3 was too high, (60 g/m3), as shown in all the results obtained when
testing this mix. It was therefore decided to use a dosage of 30 g/m3 of viscosity modifier for
mixes 4 to 9.

Mix 4 (Eikenhof andisite) also performed poorly as an SCC with a yield value that is too high. It
also blocked in both the V-funnel and L-box test and showed signs of bleeding. The better mix
of the three stone types used was Mix 5 (Jukskei granite), with a better yield value and an
acceptable plastic viscosity. The V-funnel and L-box results indicate that the yield value and
plastic viscosity are still too high and that blockage occurred. The mix containing Olifantsfontein
dolomite (mix 6) did not conform to the SCC specifications either, but much less superplasticiser
was required which make this mix cheaper. If the same amount of superplasticiser was used as
for mix 5, it could have had better rheological properties since the aggregate shape is better.
The yield value is unacceptably high, which is confirmed by the V-funnel and L-box results.

Mixes 7, 8 and 9 were the same as the previous three mixes, except for the extra 20% fly-ash
included in each mix. The extra fly-ash showed an improvement in the yield value, but mixes 7
and 9 are still not acceptable for a good SCC. Mix 8 (Jukskei granite) was the best mix tested
with a low yield value and a reasonable plastic viscosity. It also had the best slump flow and
L-box values of all the mixes. The L-box value of 0.9 indicates that blockage was not a problem
either.

Figure 5 presents a summary of the results from the Tattersall Two Point Tester, showing the
straight line relationship between shear stress (τ) and shear rate (γ). The lines in Fig.5 are
created from the yield stress (τ0) and plastic viscosity (µ) readings substituted into the Bingham
equation (equation 1) to produce a linear relationship line. This figure indicates that Mix 8 (thick
solid line) shows the best characteristic as a SCC because the yield stress is close to zero.

220
Figure 5. Bingham lines for six of the concrete mixes tested based on the measured plastic viscosity and
yield stress values from the Tattersall Two Point Tester.

9. CONCLUSION

The test results indicate that SCC can be made with the materials used in this investigation, but
careful mix design is required. The Jukskei granite was the most suitable aggregate in the mixes
used, but requires more admixture and cement than the Olifantsfontein dolomite. Because of
the high flakiness of the Eikenhof Andesite these mixes was susceptible to blocking. The
workability and flowability increase with increasing fly-ash content in the mix. The presence of
fly-ash also limits the possibility of segregation occurring.

The results provide a basis for further research work on the development of SCC in South
Africa. Full scale trials will also need to be done to determine the cost implications involved
when using SCC.

10. REFERENCES

[1] Billberg, P. Self-compacting concrete for civil engineering structures – The Swedish
experience. CBI Report 2:99, Swedish Cement and Concrete Research Institute, SE-100
44 Stockholm.
[2] Okamura, H. and Outchi, M. Applications of Self-compacting concrete in Japan,
Self-compacting concrete, third international RILEM symposium, Reykjavik, Iceland,
Aug. 2003, p3.
[3] Skarendahl, A. The present – The future, Self-compacting concrete, third international
RILEM symposium, Reykjavik, Iceland, Aug. 2003, p6.
[4] Utsi,S. Self-compacting concrete: Properties of fresh and hardening concrete for civil
engineering applications, Licentiate thesis, Lulea, Sweden: Lulea University of
Technology, 2003.

221
[5] Walraven, J. Structural aspects of Self-compacting concrete, Self-compacting concrete,
third international RILEM symposium, Reykjavik, Iceland, Aug. 2003, p15-22.
[6] Hurd, M.K. Self-compacting concrete. Can you fill your forms without vibrating? Concrete
Construction, Jan 2002, pp.44-50.
[7] Correia, E. Private communication, Johannesburg: Chryso, Feb. 2004.
[8] Gazendam, M. Durability of Self-compacting concrete, BIng thesis, Cape Town:
University of Stellenbosch, 2003.
[9] Corradi, M. Innovative Technology to improve precast processes, Zero energy system
workshop, Treviso, Italy, Oct. 2001.
[10] EFNARC2002, Specification and guidelines for self-compacting concrete, Surrey, UK,
Feb. 2002, pp 7-9.
[11] Cronje, H. Private communication, Johannesburg: Rocla, Feb. 2004.
[12] Goltermann P., Pade C., Thaulow N. High performance concrete. Rheology: from the lab to
the site, proc. Nordic concrete research meeting, Espoo, Finland, 1996, pp. 221 – 223.
[13] Tattersall, G.H., Banfill, P.F.G. The rheology of fresh concrete, 1st ed. Boston, Pitman,
1983, p 15-115.
[14] Ferraris, C.F., Brower, E.B. Comparison of concrete rheometers: International tests at
LCPC (Nantes, France), 2000, NIST. Pp 33-36.
[15] Banfill, P.F.G. The rheology of fresh cement and concrete – A review. 11th International
congress on the chemistry of cement, 2003, pp 50.
[16] Bartos, P.J.M., Sonebi, M. and Tamimi, A.K. Workability and rheology of fresh concrete:
Compendium of tests. France: RILEM, 2002, p103.
[17] Tattersall, G.H. Workability and quality control of concrete, 1st ed. London, E and FN
SPON., 1991, pp 54-77.
[18] Tattersall, G.H. and Bloomer, S.J. Further development of the two-point test for workability
and extension of its range, Magazine of Concrete Research, vol.31, no.109, Dec. 1979,
pp 202-210.
[19] Wallevik, O.H. Rheology – A scientific approach to develop self-compacting concrete,
Self-compacting concrete, third international RILEM symposium, Reykjavik, Iceland,
Aug. 2003, p23-31.

222
 SELF-COMPACTING CONCRETE: TECHNOLOGY
DEVELOPMENT AND MEASUREMENT OF WORKABILITY

J.P. Jooste1 and Y. Ballim2

1
Department of Civil Engineering, Technikon Witwatersrand, PO Box 17011, Doornfontein,
2028, Johannesburg, South Africa. Tel: +27 11 406 8121. E-mail: pjooste@twr.ac.za
2
School of Civil and Environmental Engineering, University of the Witwatersrand,
Private Bag 3, Wits, 2050, Johannesburg, South Africa. Tel: +27 11 717 7103.
E-mail: ballim@civil.wits.ac.za

Biography

Petrus Jooste

Petrus Jooste is a lecturer at The Department of Civil Engineering, Technikon Witwatersrand

where he also obtained the National diploma in Civil Engineering as well as the B-Tech degree
in Civil Engineering (Structural). He is currently doing his MSc (full dissertation) under the
supervision of Prof. Yunus Ballim at The University of the Witwatersrand. His research interests
include concrete rheology and self-compacting concrete. He is registered with ECSA and is a
Member of the CSSA.

223