Content uploaded by Matt Blacka
Author content
All content in this area was uploaded by Matt Blacka on Mar 07, 2016
Content may be subject to copyright.
Assessment of the use of Concrete Hanbar Armour
Units on NSW Breakwaters
Matthew J Blacka1, James T Carley1, Ron J Cox1, and Indra Jayewardene2
1Water Research Laboratory, School of Civil and Environmental Engineering, University of New South Wales
2Manly Hydraulics Laboratory, Department of Commerce
Abstract
Over the past 40 years, many of the breakwaters and revetments on the New South Wales coast have been
constructed, repaired, or upgraded with concrete armour units. Since the early 1980s the most widely used armour
unit has been the Hanbar. Whilst several applications of the Hanbar unit have been physically modelled, there has
been no broad research into the hydraulic response of the units to wave attack. As the units are unique to NSW, there
are no published properties in standard coastal engineering reference texts.
Physical modelling research has recently been undertaken at the University of New South Wales, Water Research
Laboratory (WRL) that investigated various parameters regarding the use of Hanbar armour units. This research
targeted the stability of the Hanbar units under different wave and water level conditions. The effect that the
placement method of the units has on stability, wave runup, and placement density was also investigated.
This paper summarises the results of the previous modelling studies of Hanbar applications to specific breakwaters
and seawalls, and presents the results of the recent research program completed at WRL. The relationships between
the stability of the armour units and the incident hydraulic conditions have been established, and guidelines allowing
more effective use of the units developed. By looking at previous breakwaters constructed with Hanbar units, and
investigating the effects of different placement methods, more efficient use of the units can be adopted in the future,
potentially reducing the construction and repair costs of structures.
Keywords: Concrete Armour Unit, Breakwater, Hanbar, Physical Model
1 Introduction
1.1 Background
The Hanbar concrete armour unit was initially
proposed and used by the NSW Public Works
Department in the late 1970s, on breakwaters along the
NSW south coast. The Hanbar is a three legged
concrete armour unit (Figure 1, Section 2), that has
been found to be easier to construct, and more durable
than many other concrete armour units. The Hanbar
provides a means of gaining stable armour on coastal
structures where it is impractical to quarry rock, and
has been designed to be produced more economically
than many other concrete armour units.
On the south coast of NSW, 15 tonne Hanbars were
first used on the Port Kembla coal loader seawall in
the early 1980s, and shortly after were also used on the
extension of the Eden breakwater. Since then, Hanbars
have been placed on Wollongong southern breakwater,
Bellambi breakwater, and Ulladulla northern and
southern breakwaters.
Hanbars were also more recently used in the upgrade
of several breakwaters on the NSW north coast.
Ballina southern breakwater was upgraded in 2000,
followed by the Coffs Harbour eastern breakwater in
2002, where 15 tonne Hanbars were placed along the
trunk of the structure and 28 tonne Hanbars on the
head. In late 2003 Forster southern breakwater was
also upgraded with Hanbar units. In total, Hanbars
have been used at eight locations along the NSW
coast, and are considered a likely option for future
breakwaters where concrete armour units are required.
1.2 Study Objectives
Previous to this study, physical modelling of
breakwaters with the use of Hanbars has been
completed on a case by case basis. Although Hanbars
have been used extensively on breakwaters in NSW,
only approximately half of the structures have been
physically modelled. The modelling that has been
completed was targeted at assessing the applicability
of a certain sized Hanbar unit, for use on a specific
breakwater design. There has been little to no
investigation or data published regarding the general
use of the unit, with regard to placement methods and
hydraulic performance.
The major objective of this study was to investigate
the effectiveness to which the Hanbar armour unit has
been used in the past. This was achieved through:
Site visits to compare modelled and actual
placement methods of the units
Researching and summarising the data
available from previous investigations using
Hanbars
Although several studies have been completed in the
past that have investigated the use of Hanbars, the
results of the studies have not been generally available.
A secondary objective of this investigation was to
bring together all of the previous physical modelling
studies, to summarise the information from them, and
to undertake additional modelling of alternative
placement methods for the units. It is thought that the
information obtained from the previous investigations,
along with the results of this study, would provide
increased guidance for future use of Hanbars.
2 Hanbar Unit Details
Geometric details of the Hanbar unit are shown in
Figure 1, with the dimensions of the unit for several
different masses shown in Table 1. The dimensions
and equations shown have been determined and scaled
from values given in PWD (1979a) and PWD (1979b).
The purpose behind the Hanbar design was to create
an unreinforced concrete armour unit, that was simpler
and hence more economical to construct, than other
concrete armour units such as Dolosse, Tetrapods, and
Tribars, that were being used around the world at the
time. The largest increase in economical production
with the Hanbar comes from the simple formwork
required for casting the unit. The form is an open
ended single draw mould that is filled from the top,
and able to be removed in a significantly shorter time
than the formwork used for other armour units. The
mould for the Hanbar has no base, with the formwork
typically being set up on a flat ground surface,
meaning that the form can simply be lifted off. This
leaves the unit sitting on the ground, ready to cure and
be placed, resulting in a much larger rate of
production, meaning that shorter lead times are
required for casting the units.
It is likely that the Hanbar unit also provides
significantly greater structural strength than most other
slender concrete armour units. Although only limited
in situ drop testing of the unit has been completed, it
can be seen from the design of the Hanbar, that the
legs are significantly stockier than units such as
Dolosse. This along with the large area of connection
between the legs and the base block, appears to give
the unit increased structural strength. Very few
structurally damaged or broken armour units were
identified in site visits to any of the locations where
they have been used, even on structures where
Hanbars have been in place for over 20 years such as
Eden breakwater and Port Kembla coal loader seawall.
Figure 1 Geometric Details of the Hanbar Armour Unit
Table 1 Hanbar Armour Unit Dimensions
Mass of Unit Approximate Dimen sion (mm)
(t) A B C D E F G H
5 800 1560 595 860 1010 745 745 940
10 1005 1965 750 1085 1275 935 935 1185
12 1070 2090 795 1150 1355 995 995 1260
15 1150 2250 855 1240 1460 1070 1070 1360
20 1265 2475 940 1365 1605 1180 1180 1495
22 1310 2555 970 1410 1660 1215 1215 1545
25 1365 2670 1015 1470 1730 1270 1270 1610
28 1420 2770 1055 1525 1795 1320 1320 1675
30 1450 2835 1080 1560 1840 1350 1350 1710
Ratio of Length A 1.00A 1.95A 0.74A 1.08A 1.27A 0.93A 0.93A 1.18A
Ratio of Mass M 466.83M1/3 912.61M1/3 346.97M1/3 502.55M1/3 591.93M1/3 434.24M1/3 434.24M1/3 550.86M1/3
3 Previous Modelling Studies
Although there has been little to no information
published that advises on either design or construction
using the Hanbar armour unit, there have been several
physical modelling studies completed. To date there
is known to have been approximately seven
modelling studies undertaken at Manly Hydraulics
Laboratory (MHL), that have all investigated specific
breakwater designs and upgrades using the Hanbar.
PWD (1979a) and PWD (1979b) were both
undertaken to investigate the design of the Port
Kembla coal loader seawall. PWD (1981) and
Lawson and Treloar (1984) were undertaken to
investigate the extension and upgrade of the Eden
breakwater. MHL (1997) was undertaken to
investigate the upgrade of the Ballina breakwater, and
MHL (1999) investigated the upgrade of the Coffs
Harbour eastern breakwater. MHL (2004) was
undertaken to investigate the upgrade of the Forster
southern breakwater.
All of the modelling studies listed here, were
completed to investigate and improve the use of
Hanbars on specific breakwater designs or upgrades
that had been proposed for certain locations. Little
research has been done to optimise the general use of
the Hanbar unit, with none of the studies investigating
the possibility of alternative placement methods of the
armour units to increase stability (damage coefficient)
and/or decrease runup.
3.1 Placement Methods and Details
The geometry of the Hanbar unit lends itself to many
different placement methods, all with potentially
different hydraulic performances; however, the
method used in the past to pick the armour units up,
has dictated the way in which they were placed. In all
previous cases where Hanbars have been used, lifting
hooks have been cast into the top of the unit during
moulding, meaning that the units are always lifted
from the top leg, and placed so that their flat base
comes into contact with the structure first. While this
method of construction has been simulated in the
physical modelling studies undertaken in PWD
(1979a), PWD (1979b), PWD (1981), and Lawson
and Treloar (1984), the method of placement used in
the later modelling studies of MHL (1997, 1999, and
2004) is unclear.
The placement densities used and recommended in
previous modelling studies, have been well
documented. Table 2 shows a summary of these
placement densities for a variety of Hanbar masses.
Typically the units have been placed in a double
layer, with approximately 60% of the units in the
bottom layer, and the remaining 40% in the top layer.
There is no data available that suggests placement
densities that have been achieved in actual field
placement of the units. It is however, suggested in
PWD (1981), that the placement densities used in
PWD (1979a) were successfully implemented in the
prototype construction of the Port Kembla coal loader
seawall.
Table 2 Previous Recommended Placement Densities
Reference Location
Hanbar
Mass
(t)
Placement Density
(units/m2 of sloped
face)
PWD (1979a) Port Kembla 15 0.235
PWD (1979b) Port Kembla 12 0.273
PWD (1981) Eden 15 0.290
15 0.280 - 0.290
Lawson and
Treloar (1984) Eden 10 0.328
MHL (1997) Ballina 15 0.300
MHL (1999) Coffs Harbour 28 0.170
12 0.240
MHL (2004) Forster 16 0.120
Note: Some of the values shown in Table 2 were not directly
quoted in the listed reference, but were instead expressed as a total
number of units that were used on a model structure. In these cases,
placement densities were calculated from the geometry of the
model structure and the number of units that were used.
3.2 Damage Coefficients
The method used to document the observed damage
in the previous physical modelling studies has varied
from one study to the next, making direct
comparisons between the results difficult. Adding to
the problem is the differing structural design and
hydraulic conditions tested (such as crest design and
overtopping). In attempting to draw some conclusions
regarding the effectiveness of the Hanbar as an
armour unit, the results from the previous modelling
studies have been reviewed, with calculations
completed that allow some basic comparisons to be
made. In keeping with previous literature, a common
index to compare the stability of different armour
units has been the Hudson damage coefficient (Kd)
recommended in SPM (1984), see Equation 1. Figure
2 is a plot of the damage coefficient (Kd) versus the
percent damage observed, from the results of previous
modelling studies. These results should be treated
with care, and it is advised that the original references
be viewed to fully understand the varying test
conditions under which the results were obtained.
The scatter in the results shown in Figure 2 can be
attributed to several factors, with the most significant
of these being variations in the following test
conditions:
Wave breaking regime (plunging, spilling,
surging, etc)
Random or monochromatic waves
Wave period
Trunk or head testing
From Figure 2 it can be observed that a conservative
value of the damage coefficient (Kd) for the 5%
damage level, for the Hanbar armour unit is
approximately 7.
M = Armour Unit Mass
H = Incident Wave Height
Kd = Hudson Damage Coefficient
ρr = Armour Unit Density (1)
∆ = Relative Submerged Density
α= Structure Slope
cotα
3
∆
d
K
3
H
r
ρ
M=
0
5
10
15
20
25
0 5 10 15 20
Percent Damage (%)
K
d
PWD (1979a)
PWD (1979b)
PWD (1981)
L & T (1984)
MHL (1997)
MHL (1999)
Figure 2 Damage Coefficients Recorded in Previous
Modelling Studies
4 Physical Modelling Investigation
Having assessed the manner in which the Hanbar
armour unit has been used in the past, through field
investigations and a desktop analysis of previous
model studies, a physical modelling investigation
program was established. The main aim of the
physical modelling program was to compare the
effectiveness of the Hanbar unit in the manner it has
previously been used, to other possible placement
methods. To do this a series of tests were performed,
that assessed the stability and wave runup over two
differently placed Hanbar armoured slopes, under
incrementally increasing wave heights.
4.1 Model Construction and Procedure
4.1.1 Testing Facilities
All physical modelling in this investigation was
undertaken in the two dimensional “one metre” wave
flume, at the Water Research Laboratory (WRL) of
the University of New South Wales (UNSW). This
flume measures approximately 35 m in length, 0.9 m
in width, and 1.4 m in depth. Monochromatic waves
were generated during the testing with a hydraulically
driven piston type wave paddle. Wave heights were
measured in both the deepwater section of the wave
flume, and in the near shore zone using three 600 mm
long capacitance type wave probes.
4.1.2 Model Construction
A false floor was constructed from plywood in the
base of the wave flume at a slope of 1V:50H, to
simulate a typical near shore bathymetric slope for the
NSW coast. This slope allowed waves to undergo the
normal process of shoaling and breaking as they
moved from the deep water section of the flume into
the near shore zone.
The test structure consisted of a primary armour layer
of Hanbar units, a secondary armour layer of rocks
with a median mass of approximately 1/5th that of the
Hanbar units, and a core layer of 1 - 2 mm sand. The
structure was set up on a continuous slope of 1V:1.5H
(typical placement slope for Hanbar armoured
structures in the past), and extended to an elevation
that exceeded the maximum wave runup level, so as
to prevent overtopping. The model Hanbar armour
units were injection moulded to have a density of 2.34
g/cm3, each with a mass of 155 g.
The proportioning of the physical model used in this
study was based on 10 tonne Hanbar units being
exposed to wave heights ranging from 1.5 m to 8.0 m
with a wave period of 12 seconds. Using these
proportions the water depth at the toe of the structure
was 9.4 m. The results of the modelling however, are
presented as dimensionless numbers, and any
reasonable scale can therefore be applied to them.
Two different placement methods for the Hanbar
units were tested in the modelling program. The first
placement method, referred to as the “upright
placement method” from herein, was constructed in
the same manner as Hanbar armoured structures have
been in the past. Each unit was lifted and lowered into
place by holding its top leg, allowing the flat base of
the unit to contact the secondary armour or another
unit first (see Figure 3, Lifting Method 1). In placing
the units in this manner, random orientation of the
units in the horizontal plane was allowed. Although
some units do topple once released, the majority stay
placed on their flat base. The units were placed so that
60% of the total number used were in the bottom
layer, and the remaining 40% in the top layer.
The second placement method tested, referred to as
the “interlocking placement method” from herein,
was proposed to maximise the interlocking of the
units, and hence increase their effectiveness at
resisting displacement from the armour layer. The
interlocking placement method was found to have a
lower placement density, and a similar double layer
thickness compared to the upright placement method,
meaning that it required less units, and had a higher
porosity. The interlocking placement method was
constructed by lifting the Hanbar units in the bottom
layer into place with their back down and legs up (see
Figure 3, Lifting Method 3). The Hanbar units in the
top layer were then lowered into place with their back
up and legs down, to form an interlocking placement
with the bottom layer (see Figure 3, Lifting Method
2). The units in both the top and bottom layers were
allowed to rotate randomly in the horizontal plane. As
with the upright placement method, 60% of the total
number of units used were placed in the bottom layer,
with the remaining 40% placed in the top layer.
Figure 3 Lifting Methods for Hanbar Units
4.1.3 Test Procedure
Each model structure was tested by running pulses of
10 monochromatic waves at a time, which started
with small unbroken waves, and increased
incrementally until depth limited breaking waves
were observed. Typically the structure was exposed to
50 waves at each wave height, and if no continuing
damage was observed, the wave height was increased.
If damage was still progressing after 50 waves,
further waves were run, until the structure was
observed to stabilise. In total each test structure was
exposed to approximately 1000 waves. Damage to the
structure was defined as units that were displaced
from their original position by more than one
equivalent diameter, expressed as a percentage of the
total number of units below the maximum runup
level. The wave height, number of units displaced,
and the wave runup level, were determined for each
pulse of waves from overhead video footage recorded
throughout the duration of the testing. As the structure
was not rebuilt before testing each different wave
height, damage to the structure was cumulative, and
has been treated as such in the analysis of the results.
5 Results
Damage levels for the range of wave heights tested
varied from 0% to 12% for the upright placement
method, while the interlocking placement method
sustained a lower damage of only 0% to 9%. Damage
to both placement methods was found to occur
slowly, with no catastrophic failure observed; each
incremental increase in wave height would typically
initiate the displacement of a further 2 to 3 Hanbar
units. Most damage was observed to occur within one
wave height above and below still water level (SWL)
for both placement methods, with the majority of the
units that were displaced, originally coming from the
upper layer of Hanbars, just above SWL.
Due to the shape of the Hanbar unit, a large degree of
self healing was observed within the modelled
structures. Approximately half of the units that were
displaced during the testing were able to move down
the structure slope, and restabilise at a different
location, often plugging the hole of a previously
displaced unit. It was observed during testing of the
upright placement method, that the orientation of
restabilised units was similar to that used in
construction of the interlocking placement method.
Throughout the testing program completed for the
upright placement method, armour unit sliding was
observed on several occasions. As waves would run
up and run down the structure face, armour units in
the bottom layer were observed to slowly slide down
the slope. This resulted in regions of tightly packed
units that had closed together, and other regions
where the underlayer had become exposed. This
mode of damage was only observed in the upright
placement method, and is likely to be a result of the
Hanbar units sitting on their smooth base. While the
units were not physically displaced from the armour
layer, this kind of damage and exposure of large
regions of underlayer material, could result in
detrimental effects for structures with armour placed
in this manner.
Comparison of observed damage between the two
placement methods shows a clear trend that the
interlocking placement method is more stable under
all wave conditions tested. However, the results
obtained for both placement methods fall within the
scatter of results from the previous studies, shown in
Figure 2. As damage to the structure was recorded in
a cumulative manner, it cannot be directly implied
from these results, that a certain wave height will
result in a certain percentage damage. For this reason,
use of the damage coefficient (Kd) in this case is not
strictly correct, but does allow reasonable comparison
between the results of this, and previous studies.
Figure 4 shows a plot of the damage coefficient (Kd)
versus the percentage damage observed for the results
obtained in this physical modelling study.
The results for the upright placement method confirm
that for Hanbars placed in this manner, a conservative
value of Kd for the 5% damage level is approximately
7. A conservative value of Kd for Hanbars used in the
interlocking placement method however, could be
almost as high as 12, and at least 10 for the 5%
damage level. It is advised in SPM (1984) that a
suitable Kd for Dolosse is 16 and Tribars is 12. The
results from this study indicate that if Hanbars are
placed using the interlocking placement method, they
can be almost as effective as the Tribar and Dolos
units, yet simpler to produce and structurally stronger.
Wave runup levels were determined for all wave
heights tested, from the overhead video footage taken
during the modelling. Dimensionless wave runup
values (ratio of wave runup, R to wave height, H) for
the upright placement method ranged from
approximately 1.1 to 1.8, and for the interlocking
placement method from approximately 1.4 to 1.9.
Dimensionless wave runup (R/H) is plotted against
the surf similarity parameter (ξ) in Figure 5, where it
can be seen that the upright placement method
generally results in slightly lower wave runup levels
compared to the interlocking placement method.
Although the interlocking placement method has a
slightly higher porosity which would reduce wave
runup, it has increased planar surfaces aligned with
wave uprush that are thought to increase wave runup.
6 Conclusions and Recommendations
The Hanbar armour unit has been developed to be
simpler and more economical to produce than other
concrete armour units. For this reason the Hanbar has
been, and is likely to continue to be, widely used
along the NSW coast. In maintaining the high
production rates achieved through the slip moulding
of the Hanbar units, the method used to lift and lower
the units into place has been restricted to lifting the
units from their top leg, meaning that most of the
units have ended up placed on their flat base. This
method of placement, (the upright placement method)
0
5
10
15
20
25
30
35
02468101214
Percent Damage (%)
Damage Coefficient, K
d
Upright Placement Method
Interlocking Placement Method
0
0.5
1
1.5
2
2.5
3
3.25 3.75 4.25 4.75 5.25
Surf Similarity Parameter, ξ
Dimensionless Runup, R/H
Upright Pla cement Method
Interlocking Placement Method
Figure 4 Observed Damage Results Figure 5 Observed Runup Results
has lowered possible interlocking between armour
units, which has inturn, lowered the damage
coefficients adopted for design of structures with
Hanbar units. Previous modelling studies indicate that
the typical damage coefficient associated with a 5%
damage criterion, is approximately 7. Values in this
order were confirmed by the recent physical modelling
undertaken at WRL. It was also identified that when
the Hanbar units are placed in this manner, they are
prone to sliding down the structure slope on their flat
base, creating holes in the primary armour layer, and
exposing the underlayer material.
A second method of placing the units was proposed
(the interlocking placement method), which aimed at
improving the interlocking between units. This
placement method resulted in a lower placement
density, a higher porosity, and was observed to be
significantly more stable than the upright placement
method, as used in the past. With use of the
interlocking placement method, damage coefficients
for the 5% damage level were observed to be
approxiametly 12. It is recommended that further
physical model testing be completed under irregular
wave conditions, to obtain more accurate estimates of
damage coefficients, reflection coefficients, and runup
levels that can be used in future breakwater designs.
Large variations in previously recommended
placement densities could imply that in some cases,
the placement density used was higher than required. It
was recommended in PWD (1979a), that 15 tonne
Hanbars be placed on the Port Kembla coal loader
seawall, with a density of 235 units per 1000 m2 of
face. In contrast, PWD (1981) and PWD (1984)
recommended that 15 tonne Hanbars be placed on
Eden breakwater with a density of 290 units per 1000
m2 of face. Field visits to both sites identify that in
many places, the units have been crowded so close
together, that interlocking and porosity have been
minimised. It is also apparent in field visits, and
indicated in PWD (1994), that Hanbars have been
placed on the crest of some breakwaters in a single
layer, single row, pattern placed method, to reduce
overtopping of the structure. It appears from the results
of this study that the Hanbar unit performs most
successfully when placed in a double layer, with
maximum interlocking and porosity between the units,
which has not always been achieved in the past.
7 References
Coastal Engineering Research Centre (SPM 1984)
Shore Protection Manual, Department of the Army, 4th
Edition, USA
Lawson & Treloar and Public Works Department
NSW (1984) Flume Testing of Eden Breakwater Stage
2, Sydney
Manly Hydraulics Laboratory (MHL 2004) Forster
South Breakwater Physical Model, Report No.
MHL1209, NSW Department of Commerce, Sydney
Manly Hydraulics Laboratory (MHL 1997) Ballina
South Breakwater Artificial Armour Unit Comparative
Performance Basin Testing, Report No. MHL897,
NSW Department of Public Works and Services,
Sydney
Manly Hydraulics Laboratory (MHL 1999) Coffs
Harbour Eastern Breakwater Physical Model Study,
Report No. MHL941, NSW Department of Public
Works and Services, Sydney
Public Works Department NSW (PWD 1979a) Port
Kembla Revetment Model, Manly Hydraulics
Laboratory Report No. MHL262, Sydney
Public Works Department NSW (PWD 1979b) Port
Kembla Coal Loader Seawall Model, Manly
Hydraulics Laboratory Report No. MHL272, Sydney
Public Works Department NSW (PWD 1981) Flume
Testing of Eden Breakwater Stage 1 – Monochromatic
Waves, Manly Hydraulics Laboratory Report No.
MHL313, Sydney
Public Works Department NSW (PWD 1994) NSW
Breakwaters – Asset Appraisal Part 4 – South Coast
Region, Manly Hydraulics Laboratory Report No.
MHL648, Sydney
