Fibre Reinforced Facade Encyclopedia: Manufacturing

Prototyping in facade manufacture is the staged production of physical samples that prove finish, geometry, fixing and tolerance before the production moulds are committed. It is a sequence, not a single event: a small control sample settles colour and texture, a larger sample settles the relationship between the finish and the real panel surface, and a full prototype panel proves that the designed geometry can be made, demoulded, handled, fixed and sealed. Each stage is cheaper than the stage that follows, which is the whole point of running them in order.

A 300 mm square control sample costs very little and settles an argument about colour that would otherwise be had in front of four hundred installed panels.

At a glance

PropertyTypical valueNote
Control sample200 to 400 mm squareColour, texture, finish method; the reference for all later work
Finish range sampleTwo or more samples showing the acceptable spreadDefines the tolerance band, not a single target
Sample panel1 to 2 m² section of a real panelProves finish at scale, edge returns, joint detail
Prototype panelOne full size production panelProves geometry, weight, anchors, handling and fit
Visual mock-upTwo or more prototype panels on a frameAssessed at the correct viewing distance and orientation
Performance mock-upMulti-panel wall tested to AS/NZS 4284Separate exercise; water, air and structural performance
Typical time to first control sample2 to 4 weeks from award
Typical time to prototype panelAfter mould manufacture
RetentionApproved samples retained in factory and on siteThe reference for production acceptance

Why the sequence exists

Every stage of prototyping answers a different question, and answering them out of order wastes money. Colour and texture are properties of the mix and the finishing method, so they can be settled on a small sample before anyone knows what the panel looks like. Surface quality at scale is a property of the mould face, the spray technique and the panel size, so it needs a piece large enough to show the variation a small sample cannot. Geometry, weight, anchor position and joint tolerance are properties of the whole panel, so they need a whole panel.

The cost gradient is steep. A control sample is a bench item. A sample panel needs a partial mould. A full prototype needs a production mould, which is the tooling the project has just committed to. Running the cheap stages first means that when the expensive tooling is cut, only the questions that genuinely require a full panel are still open.

Stage one: the control sample

A control sample is a small piece, commonly 200 to 400 mm square, made in the actual mix with the actual finishing method. It settles the colour, the texture, the aggregate exposure and the surface treatment, whether that is off-form, acid etched, sandblasted, polished or coated. It is the reference against which every later panel is judged, and it is retained.

The single most useful thing a specifier can do at this stage is to insist on a range rather than a single sample. Cementitious materials with integral colour and exposed aggregate do not produce identical panels; they produce a distribution. Approving one sample and then rejecting production panels for not matching it exactly is the most common cause of dispute on a moulded facade package. Approving a pair or a set that brackets the acceptable spread converts colour consistency from an argument into a measurable acceptance criterion.

Assess samples wet and dry, and in the orientation and light the facade will actually see. A cementitious panel looks different wet, and a textured face reads differently on a vertical surface in raking light than it does flat on a table under fluorescent light.

Stage two: the sample panel

The next stage is a section of real panel, typically one to two square metres, made in a partial mould. It shows what the finish does across a larger area: whether the texture is uniform, whether spray or casting marks show, how an edge return reads, how a corner resolves, and what the joint between two panels looks like when they are set side by side at the design gap.

Stage three: the prototype panel

  • Can the designed geometry be demoulded without damage, and with the draft angles provided
  • Does the panel weigh what the structural engineer assumed, and is the dead load per square metre within the substructure design
  • Are the anchors, flex anchors and gravity anchors where the shop drawings say they are, and can they be reached during installation
  • Does the panel hang on the substructure, and does the joint close to the designed width once site tolerances are packed out
  • Can the panel be lifted, transported and handled without the lifting points cracking it
  • Is the panel within the dimensional tolerance the installation method assumes

A prototype that reveals a problem is a successful prototype. Finding an unreachable anchor or a joint that cannot be sealed on panel one costs a mould modification. Finding it on panel two hundred costs a facade.

Prototypes, mock-ups and performance testing

The terms are used loosely and the distinction matters commercially. A prototype panel is a manufacturing exercise: it proves that the panel can be made. A visual mock-up is an approval exercise: two or more panels erected on a frame, assessed at the correct viewing distance, and formally signed off as the standard for the project. A performance mock-up is a testing exercise: a multi-panel section of wall built as designed and tested for air infiltration, water penetration and structural performance under wind pressure, in Australia to AS/NZS 4284.

The three are often conflated in a tender and then argued about later. A specification should say which of them is required, how many panels each involves, where they will be erected, who attends, what the acceptance criteria are, and whether the mock-up may be incorporated into the works or must be demolished. Those clauses are worth more than another paragraph describing the finish.

What prototyping does to programme

The way to protect the programme is to start the small stages early. Control samples can be made from the specification before shop drawings are complete, because they test the mix and the finish rather than the geometry. Prototype panels cannot start until the mould exists, so they follow tooling. Structuring approvals so that finish approval and geometry approval run in parallel rather than in series is usually worth several weeks of lead time.

Standards and references

  • AS/NZS 4284 Testing of building facades, for performance mock-up testing of air, water and structural performance
  • AS 3610 Formwork for concrete, surface finish classes, used as a reference for off-form finish acceptance
  • GRCA Specification for the Manufacture, Curing and Testing of GRC Products, including sample and batch testing requirements
  • NATSPEC and NBS facade worksections, which set out sample, prototype and mock-up requirements in the specification
  • BS EN 1169 factory production control, which requires reference samples to be retained
  • AS/NZS 1170.2 wind actions, as the basis for the pressures applied in a performance mock-up

Related terms

Questions specifiers ask

What is the difference between a sample, a prototype and a mock-up?

A sample proves finish. A prototype panel proves that a full size panel can be manufactured, handled and fixed. A mock-up is erected for formal approval, and a performance mock-up is additionally tested to AS/NZS 4284 for air, water and structural performance.

How many samples should a specification ask for?

At least a control sample and a range of samples bracketing the acceptable spread of colour and texture, then a sample panel, then a full prototype. Approving a single sample as an exact target is the most common cause of later dispute.

When should sampling start?

As early as possible. Control samples test the mix and the finish, not the geometry, so they can start before shop drawings are complete and run in parallel with tooling design.

Can the prototype panel be used in the building?

Sometimes, if the specification allows it and the panel passes inspection. Say so explicitly in the specification, because the alternative is a panel that has to be scrapped and a cost that nobody priced.

What does a prototype panel actually catch?

Anchor positions that cannot be reached, joints that will not close within site tolerance, lifting points that crack the panel, weight above the substructure design, and demoulding damage on tight geometry. None of these show on a sample.

Early samples cost days. Late finish disputes cost months. ShapeShift produces control samples and full prototype panels for every project. Talk to the ShapeShift team

Reviewed by Richard Halvorsen, General Manager, ShapeShift Technologies. Last updated September 2026.