APPENDIX 1
CASE STUDIES

1. University of Technology, Haymarket Campus, Ultimo, Sydney
2. Group Housing, Department of Housing, Glebe, Sydney
3. Office Building, Castle Hill, Sydney
4. Office Building, Surry Hills, Sydney
5. Caroline Chisolm High School, Chisolm, ACT

CASE STUDY 1

UNIVERSITY OF TECHNOLOGY HAYMARKET CAMPUS, ULTIMO

Client	University of Technology, Sydney
Architect	Philip Cox; Richardson, Taylor & Partners Pty. Ltd.
Builder	Concrete Constructions (N.S.W.) Pty. Ltd.
Bricklayer	Reno Pty. Ltd.
Brick Supplier	Statebrick Pty. Ltd: (a) Brickworks Red (b) Cream Wirecut (c) Salmon Wirecut
Contract Value	$20,000,000
Brickwork Subcontract Value	$1,400,000
		University of Technology, Sydney Architect: Cox, Richardson, Taylor and Partners Photo: Tom Balfour

The building is a 3 and 4 level reinforced concrete frame with a non-load bearing cavity brick facade. External brickwork throughout is banded in cream and salmon coloured 8 course bands with a variety of large circular openings at windows and louvres. The building was constructed from 1981 to 1983.

Relevant isues:

1. Project Size
   The 17,000 m² building required a huge order of 1.2 million bricks. This created an opportunity for the contractor to place a single brick order. The opportunity was taken and large quantities of bricks were stored at the brickmaker's yard (Statebricks, Blacktown - now closed). Towards the end of the job, however, additional bricks had to be ordered and produced afresh due to losses by wastage and some underestimating in the initial order resulting in some brick colour batch problems.
2. Characateristic of Expansion ('e' Factor)
   Fortunately, as both the cream and the salmon brick were obtained from the same clay materials fired to different temperatures and levels of oxidation, the 'e' factors were nearly identical. This minimised the problem of differential brick expansion along the band joint. Advice on brick growth data was provided by the Melbourne office of the Clay Brick and Paver Institute. The similar 'e' factors between face bricks and the relatively weak mortar mix resulted in the decision to not have a slip-joint in horizontal bed courses between two bands.
3. Brick Size
   The longer firing period of the salmon brick resulted in a brick some 4mm shorter than the cream. The consequence on site was to target 9mm perpends for cream bricks and 13mm perpends for salmon bricks. This was further complicated by considerable variation in brick length within each colour. The result, in parts, is some unsightly fat perpends and smaller areas of excessively tight (i.e. 4mm) perpends to maintain horizontal gauge.
   Coping successfully with the basic size and gauge differences between cream and salmon bricks, and developing a productive momentum in bricklaying speed and accuracy, came soon after the first level was laid.
4. Brick Shape
   Cupped, or "Banana" bricks plagued the project, particularly in the cream brick. In retrospect, this probably could have been corrected by forcing the manufacturer to improve production processes. In practice, however, the very large order of bricks was made in one production run and the programming constraints of the job made it difficult to stop progress to wait for a better brick. Wherever possible, placement of artificial lighting was amended to avoid glancing light being cast over the brickwork which would have accentuated the effect of warped brick.
5. Colour
   Problems of brick colour batch variations were common in the cream brick, but rare in the salmon brick. The quality control procedure adopted on site was for the architect to sign samples of the lightest and darkest acceptable brick colour of each type. Incoming brick pallets were then checked by the Clerk of Works for compliance with the approved samples. Non-conforming bricks were rejected for return to the manufacturer, or to be used in below-ground or rendered applications. The greatest factor working against detecting brick colour variations for bricks which got passed this process was the method of finishing joints. As joints were bagged flush, considerable amounts of mortar smear made it difficult to judge whether bricks were well laid, whether they had chipped edges, or were likely to have colour batch problems. By the time cleaning down occurred (i.e. following completion of each wall to parapet height), it was too late to remove large areas of colour mismatched bricks.
   Fortunately, however, the blood-and-bandage pattern of banded colours masked a lot of brick colour batch problems. Further, the subcontractor was co-operative (because of the Clerk of Works eternal vigil!) and laid off several brick pallets in each work area.
6. Brick Joints
   The joint design was to be struck flush. The bricklayers initially protested against this, largely on the grounds that:

   (a) The mix design was very lean, to achieve the lightest possible mortar colour using off-white cement, and to avoid a true slip-joint between colour bands. The lean mix was, in turn, slow to cure to a stiff enough consistency to cut neatly with the trowel.
   (b) Coarse sand (which was used in initial sample panels) resulted in the trowel cutting action dragging out the mortar in an unsightly manner.

   The problems were basically overcome by experimentation in methods of "bagging" the finished joint. The solution was to change the sand to a finer Sydney sand mix and to flush joints some 20 minutes (approximately 3 courses higher) after laying to allow the mortar to stiffen. Flushing was achieved by a rag-covered sanding block.
   The last (unresolved) difficulty was that perpends could never be flush struck and bagged to a good line due to a slightly rounded arris on the bricks' vertical edge.
7. Scored Bricks
   Given that three different types of face bricks were used, it was important to avoid the additional cost of purpose-made sill bricks. Thus for both sill bricks and half-brick batts, the manufacturer agreed to score the face during extrusion of the green brick clay column. This enabled sills and half-bricks to be bolster cut which is considerably cheaper than sawcutting. In retrospect, purpose-made sill squints would have cost little more.
   Conduit-bricks were produced as specials with voids to suit electric wiring within double-face 110mm walls. This was not a success, as the conduit-bricks were fired in their own batch, and invariably showed up as a colour mismatch in the finished building.
8. 110mm Double-Face Walls
   For economy, a design decision was made to lay internal 110mm partition walls to fair face each side. It quickly became evident that this was a mistake, as the extruded bricks in use had a "first" and a "second" grade face. The second face and markings originated in the conveyor rollers which transport bricks through the extrusion, cutting and setting stages pre-kiln. This was exacerbated by brick width tolerances, which in themselves mean that the "second" face of a 110mm wall was out-of-plane, but combined with multi-coloured bricks, the variation out-of-plane was even worse. One should only attempt to design 110mm work double face if the brick face appearance and width tolerances permit.
9. Brick Impurity Control
   The brick manufacturer's chosen method of controlling vanadium staining (which was believed to be a problem relating to the particular clay being used) was to silicon-dip all bricks just prior to palletisation. The vanadium was certainly contained as a result (because its migration to the surface was limited by the silicon "skin"), but the cure was worse than the complaint, as the bricks were laid off wet and dry pallets, depending upon the weather at the time of delivery. Locked-in moisture in some bricks, but not others was very slow to dry out. The result was quite unsightly and appeared, to all intents, like brick colour batch problems.
10. Gauge
    The design intent included creating a bold gauge (8 courses = 700mm) of 12mm bed joints, to keep polychromatic brick bands monolithic in appearance. However, in practice, the fat bed joint in a relatively weak mortar led to brickwork sagging if more than 20 courses were laid in a day. This is not unusual in internal partition work, and slowed the job down.
11. Internal Polychromatic Walls
    The use of banded and panelled polychromatic brickwork on internal walls worked well.

CASE STUDY 2

GROUP HOUSING, DEPARTMENT OF HOUSING, GLEBE, SYDNEY

The project built in 1990, consists of 108 medium-density residential units for the Department of Housing in an inner-city suburb. Construction is in loadbearing cavity brickwork, rising 2 to 3 stories above reinforced concrete basement carparks. Brick colours chiefly vary between adjoining units and architectural elements; the design intent being to establish individual identity and liveliness of composition. Top parapet coursing is invariably in darker brick for both visual strength and for better weathering.

Relevant issues:

1. Brick Selection
   For a medium-density domestic project, the large volume of brickwork, and the length of the site (one large inner-suburban block) led to difficulties in contracting and quality control. For example, the loadbearing floor-slab design coupled with the long narrow site, determined a slow overall program. That program was at all times dependent upon the bricklaying subcontractor who for a number of reasons, was a small firm. The combination of small-scale subcontractors, program demands, inadequate supervision, and the eruption of industrial unrest brought the project to a halt 60% through. At that time the subcontractor went into liquidation. It is worth mentioning this in the context of polychromatic brickwork - as the greater skill level and supervision demands of polychromatic projects must confront the sometimes volatile nature of bricklaying as an industry. It is not uncommon, as in this case, that the technical challenges presented by polychromatic work, are added to the strains of head contractors managing a team, firstly, driven by a failing subcontractor, and then under daywork to complete the project. There is merit in considering at tendering stage issues such as select bricklaying subcontract lists, and dividing jobs into more manageable contract sizes.
2. Brick Order
   At documentation stage, the architect contacted several major brickmakers in an attempt to establish reliability of supply, colour uniformity, expansion coefficients and the ability to stockpile a single-run production order. The final selection of brick supplier was based largely on the preference for all brick colurs (excluding cappings) to be from the one supplier and the right colours when viewed in combination. The budget determined that only extruded bricks would be used. Although the contractor was urged to place a single production order, at the later stages of the job extra bricks were supplied resulting in colour batch variation.
3. Brick Colour Batch Problems
   Over the life of the job, brick colour batch problems frustrated both architect and builder, resulting in a loss of productivity, morale and smooth progress. Colour variations were worse in creams than in deeper pinks. Despite signed brick samples at the outset defining the acceptable colour and tonal range, the manufacturer delivered brick colours outside these limits. In fairness to the supplier, most colour problems were only identified when brickwork was in place and cleaned. The practicalities of checking every incoming brick pallet against signed samples were not achievable (although there remains no better system). Further, the contractor's failure to order enough bricks initially led to second and third production runs, in which the various clay seam characteristics had altered. Much correspondence and interaction between all parties concerned whether colour problems were short-term (moisture differences between pallets as laid) or long term. More brick walls were removed for relaying than solutions were forthcoming from the contractual arguments over brick colour problems. The failure of the bricklaying company made delineation of relative responsibilities difficult. The design employed bricks of too similar a colour/tonal range. As a consequence, colour batch problems in one brick type often blurred the design distinction demanded between two adjacent brick panels.
   At one stage within a single wall of cream bricks, the only acceptable means of dealing with brick colour change was the introduction of a rendered 2-course band. This argues in favour of some polychromatic banding.
4. Joint Design
   A contributing factor to the lateness in brick colour problem detection was the use of a flush struck joint. The amount of mortar smear on the brick face (which would have been much less with a better bricklayer) was considerable, making inspection difficult until the time of final clean.
5. Brick Size Irregularities
   "Banana" bricks and length variation of up to 8 mm, led to on-going difficulty in maintaining flush joints and regular gauge.
6. 'e' Factor Compatibility
   One advantage of a single-sourced range of the main face bricks was that coefficients of expansion were relatively close. This obviated slip joint complications.
7. Weathering
   Although the red tile-faced bricks work well as a dirt-shedding parapet brick, the cream brick areas suffer from dark mortar joints as weathering (dirt runoff and fungal growth) occurs. In the context of polychromatic work, where light brick panels attract more visual scrutiny, greater care in weathering the finished brickwork is needed.

CASE STUDY 3

OFFICE BUILDING, CASTLE HILL

This is a 2,200m² office building in an industrial park setting built in 1991. Two storey in scale, the construction is an outer brick cladding skin wwith steel wall stiffeners fixed to a reinforced concrete frame. Internal walling throughout is lightweight. Face brickwork is banded with a very simple dark brick plinth and some panels of the same brick surrounding major windows and entries.

Relevant issues:

1. Brick Selection
   The initial design intent was to select a light cream with a medium salmon brick at the plinth. Every available dry-pressed and extruded cream brick was surveyed. Two problems, endemic to New South Wales creams, recurred. Firstly, the majority of creams are pasty, parchment and buff-coloured, lacking freshness and vigour. A second group, including the dry-pressed creams, consists of many overly yellow creams. Yellow can easily dominate en masse, making a cream/salmon combination difficult. The solution was to change focus to a flesh/cream which avoided the dullness of buff-creams and the danger of yellow domination. The flesh/cream works well with a range of salmon and deep pinks. The red (salmon) brick finally selected had to be strong to contrast with the flesh/cream. The first red brick nominated was chosen on the basis of several sample packs delivered by the manufacturer. When the first pallet arrived on site, a combination of marble frit on the brick face (removable, but not easily) and colour variations suggesting an oxide additive made the brick unacceptable. An alternative red brick was quickly found.
2. Brick Gauge
   The design anticipated brick length variations between cream and red bricks. A two-course cut soldier broke gauge between colour bands. On site, the cream and red face bricks as first delivered were measured and were of average length:
   Cream - 229mm
   Red - 224mm
   Detailed discussions between architect, contractor and bricklayer were held. In this case the builder took a stand that the brick supplier could do better. This was rewarded by the brick supplier then delivering red bricks of an acceptable average length of 227mm. Although in the final analysis the two-course soldier was deleted, the perpend gauge variation between red and cream bricks was acceptable. The general problem that simple colour banding raises is this difference in brick size and the difficulty in both keeping perpends vertically aligned and maintaining a similar perpend dimension.
3. Colour Batching
   Colour variations in the cream brick, whether by different moisture levels of brick or by brick batch differences, were problematic in the latter stages of the project. The chief problem is the lateness at which such colour variations come to light.
4. Joint Design
   A client preference for mortar colour matching the cream brick led to the use of a strong coloured sand (in the bush sand component of the mix). In turn, the mortar sand grain size was coarser than normal - inconsequential with raked joints, but more difficult with flush struck joints as was the case here. Cutting a coarse sand mortar to a clean and full joint is harder, and any brick joint voids in the initial bricklaying work are difficult to fill or rectify after the mortar has set.

CASE STUDY 4

OFFICE BUILDING, SURRY HILLS

This is a 5 storey office building in an inner-city suburb built in 1988. Construction is an outer brick cladding skin fixed to a concrete frame. Internal walling is lightweight gypsum plaster blockwork. The design intent was to create a commercial building compatible with a streetscape of terrace houses, large warehouses and a church; all of loadbearing brick construction. Face brickwork is banded in varying widths, colours and textures.

Relevant issues:

1. Brick Selection
   Several factors dictated the colour and shape of the bricks. The initial design intent was for a building banded from a textured, dark coloured base to a smooth, lighter top. The lower section (from a varying ground line to Level 3) is generally carparking with few windows but the need for large free area ventilation openings. To achieve a solid base, many small ventilation openings were incorporated into two dark coloured alternating bands. Texture in the base was achieved by using squint bricks laid with points out in stack bond alternating with ventilation openings. The upper section is banded with three colours, one of which also occurs in the base. Altogether four colours are used with one colour requiring a squint brick as well as a standard brick shape.
   In consideration of the above factors the initial preference was for all bricks to be supplied by one manufacturer. In the end this was not possible with three colours (red, pink and cream) from one manufacturer and the other colour (dark brown) from a different supplier. All bricks are dry pressed.
2. Brick Size
   Brick sizes varied greatly between the four colours with average stretcher lengths as follows:
   Brown 225mm
   Cream 232mm
   Pink 232mm
   Red 228mm
   As a consequence, the initial dimensional setout was abandoned in favour of a setout of "best fit". This resulted in some unsightly fat perpends in the short lengths of walling in the red bands to maintain horizontal gauge with the cream and pink bands.
   The 3mm difference in length between the brown and red bricks was not discernible in the alternating bands of Flemish bond red brickwork with bands of brown squints laid in stack bond. This "textured" base was separated from the upper brickwork by an exposed concrete slab toe.
3. Bricklaying
   The textured base, while not particularly complicated, was slow to lay and the narrow bands resulted in the need to frequently supply different bricks (colour and shape) to the scaffolds. Some care was required to ensure the headers laid in the red band lined up with the header width ventilation opening in the stack bond brown band. The bricklayers would only do this work on an hourly rate basis and a small team was used to ensure consistency for the whole base. Overall this base was very costly in time and money to the builder.
   In an attempt to minimise his losses on the brickwork subcontract, the builder used a different bricklaying subcontractor after the Level 3 concrete toe. This very large team proved difficult to control and, combined with the brick size problem, resulted in much brickwork being demolished and rebuilt before it was accepted by the architect.

CASE STUDY 5

CAROLINE CHISHOLM HIGH SCHOOL, CHISHOLM, A.C.T.

Client	National Capital Development Commission (now ACT Public Works)
Architect	Lawrence Nield & Partners Australia Pty. Ltd.
Builder	Civil & Civic Pty. Ltd.
Bricklayer	Gougard (original contractor)
Brick Suppliers	Clifton Bricks Extruded 'run-of-kiln' red Cowra Brickworks Dry-pressed red Dry-pressed silver Bowral Brickworks Simmental silver Goulburn Brickworks Dry-pressed red
Contract Value	$7,300,000
Brickwork Subcontract Value	Not known
		Caroline Chisholm High School, ACT Architect: Lawrence Nield and Partners

The high school consists of twelve separate buildings, of single storey loadbearing cavity wall construction and diaphragm wall construction to the Gymnasium Hall. The external brickwork is banded red and silver, three courses each to twelve courses, then 'run-of-kiln' red above. Lintels are in the dry-pressed red constructed as a flat arch. A toothed course runs for the perimeter of each building at eaves level, topped by a brick-on-edge. Parapet coursing was documented as solids 'run-of-kiln' red.

Because of the scale of the Gymnasium, the base course of red and silver dry-pressed bricks was increased in height with additional banding at the eaves and to the gable ends.

In Stage 3 the bricklayer fell behind the construction programme, delaying the erection of steel roof framing. Where previously the steel was supported by the brickwork, 'carport' columns now supported the main roof frames, encased by brickwork and the brick cavity grout filled.

Construction started in October 1983, Stage 1 was completed in January 1985, Stage 2 in May 1985 and Stage 3 in December 1985.

Relevant issues:

Project Size
The 7200m2 building required 920,000 bricks. The project was constructed in three stages to allow a staged occupation because of the local boom in residential development. The 9 hectare site allowed large quantities of bricks to be stored on site. Whilst Clifton Bricks quick turnover resulted in large quantities of rejected face brick, Cowra Brickworks was not able to supply at the required rate. Cowra Brickworks eventually fell behind programme in the production of dry-pressed silver bricks and Bowral supplied the alternative - Simmental Silver. When Cowra Brickworks could not supply the red bullnose for door sills, Goulburn Brickworks fortunately had a similar brick.

'e' Factor
The base course bricks were all dry-pressed, obtained from similar clay material and fired by the same process. The 'e' factors were, however, different (0.6mm/m red; 0.282mm/m silver). The 'run-of-kiln' red surprisingly had a low 'e' factor (0.25 mm/m). Advice on brick growth was provided by the BDRI and a relatively weak mortar mix (1:1:6) was specified removing the need for a slip joint between the brick bands.

Brick Size
The different clay material and manufacturing process resulted in the 'run-of-kiln' red bricks being some 4-5mm shorter than the dry-pressed bricks. To maintain the horizontal brick rod it was necessary to increase and/or reduce the perpends. The articulation of external brick skin at the windows (30mm indent for window) required the bricklayer to accurately set out the width of the horizontal dimensions at the footing level.

Brick Quality
Fissures and cracks to the face of the 'run-of-kiln' red bricks plagued the project. The elimination of rejected bricks was important because of the incidence of freezing temperatures where water trapped in the crack would freeze and cause spalling to the outside face. Constant supervision resulted in reject bricks being cut out of facework. By the end of Stage 1 the bricklayers were more selective when choosing facebricks.

Brick Colour
Problems in brick colour variations were overcome by screening the pallets as they arrived on site and mixing pallets when selecting bricks. Non-conforming facebricks were returned or used in common work.

Brick Joints
Brick joints were ironed with a 10mm raked joint between the brick colours. The raked joint accentuated the banding and concealed any movement at the bed joint between differing bricks.

Special Bricks
Sill bricks were to be manufactured as a 'cant' brick. Window sills were laid brick-on-edge with a 5mm fall; door sills were laid with a bullnose brick in a Goulburn red, similar to the Cowra red.
Parapet copings were specified to be a 'run-of-kiln' red solid. At the time of detailing the parapets BDRI recommended the use of solids to increase the impermeability of the top courses and increase the mass. Clifton Bricks produced a solid in the same colour, however, when the pallets arrived on site they were a peach colour. Several firings later, the colour approached the red of the extruded bricks. The increase in the mass of the brick and additional water content were given as the reasons for the colour difference.

Internal Polychromatic Walls
Corridor walls were laid as face, red to twelve courses (dado height to disguise the grubby marks and bag scrapes) with silver above. Walls were laid as face one side only, the other rendered and painted. Particular care was required when chasing for services. Steel 'T' wall stiffeners were co-ordinated with the brickwork to work with the horizontal brick rods.

Quality Control
The original bricklayer left the project at the end of Stage 2 (50% of the bricklaying contract). At the time the residential market was booming and bricklayers were being paid more per thousand for simple residential than the rate tendered on the high school. The contract was terminated and a new 'gang' was engaged to complete the works. This involved a review of quality standards in both brick selection and laying techniques with the new bricklayers.

'Chisholm' Lintel
For openings wider than could be accommodated by a galvanised steel arch bar or angle a 'Chisholm' Lintel was designed. Proprietary alternatives were investigated, e.g. Hesbia lintel, but were rejected for two reasons:
(a) The galvanised strip built into the bed course must be tensioned and there is no guarantee it will not corrode.
(b) Stretcher Bond Straight brick coursing spanning a three metre opening with no visual 'architectural' device is not convincing.
The 'Chisholm' lintel consists of a reinforced concrete beam formed in a wide brick cavity with the bricks propped and used as permanent formwork. Brick ties are built into all the vertical and horizontal joints.
The flat arch used for all window and door openings was again expressed in the brickwork for the larger openings. This detail was also employed in the 760mm diaphragm walls of the Gymnasium where an expressed concrete lintel was not appropriate.