Fibre Reinforced Facade Encyclopedia: Properties

Moisture movement is the reversible swelling and shrinkage of a cementitious material as its moisture content changes with the weather. A panel that is wet is larger than the same panel dry, and it cycles between the two for the life of the building. It is measured as a strain, conventionally quoted in millimetres per metre, and for glass fibre reinforced concrete (GRC, GFRC) it is of the order of one millimetre per metre between saturated and oven dry. That is a larger movement than thermal expansion over the same panel, and it is the reason movement joints and flexible anchors exist.

A 3 metre panel that moves 1 mm per metre grows and shrinks by 3 mm with the seasons. If it is bolted rigidly at both ends, that movement becomes a crack.

At a glance

PropertyTypical valueNote
Glass fibre reinforced concrete, reversible movement0.05 to 0.15%, or 0.5 to 1.5 mm per metreTypical range; commonly taken as up to 1.0 mm/m for design
Design value commonly used1.0 mm per metre (0.1%)Saturated to dry; confirm against the supplier’s data
Irreversible drying shrinkage, first cycleAdditional, occurs onceLargely complete before installation if cured correctly
Thermal expansion, cementitious composites10 to 20 × 10⁻⁶ per °CFor comparison; from the same panel
Thermal expansion, GFRP20 to 35 × 10⁻⁶ per °CTypical; polymer composites move more thermally
Moisture movement, FRPNegligibleNo capillary pore network to fill
Moisture movement, gypsum compositesSignificant; interior use onlyNot an external material
Typical facade joint width10 to 20 mmMovement plus manufacturing and site tolerance; typical
Anchor type requiredFlexible anchors allowing in-plane movementRigid restraint at more than one point cracks panels
Test methodBS EN 1170-7Extremes of dimensional variation, wet to dry

What is moving and why

Water sits in the capillary pores and in the cement gel of a hardened cementitious panel. As the panel dries, water leaves the gel and surface tension in the remaining pore water pulls the solid structure closer together, and the panel shrinks. Wet it again and it expands. The cycle is reversible and it repeats indefinitely.

This is distinct from the irreversible drying shrinkage that happens once, early, as the material dries for the first time after curing. Irreversible shrinkage is largely finished before panels reach site if they have been cured and conditioned properly. Reversible moisture movement is not finished, ever; it is a permanent characteristic of the material and a permanent design case.

The magnitude depends on the same things that govern water absorption: pore structure, water to cement ratio, compaction and curing. A dense, well compacted, properly cured panel with low absorption moves less than a porous one. Polymer modification reduces it. So does a lower water to cement ratio and proper roller compaction of each sprayed layer.

Fibre reinforced polymer is different in kind. There is no capillary network for water to fill, so moisture movement is negligible. FRP panels move because of temperature, not humidity, and they move more thermally than cementitious panels do, at roughly 20 to 35 × 10⁻⁶ per °C against 10 to 20 for a cementitious composite.

Converting movement into a joint width

The joint has to swallow everything that moves, plus everything that was not built exactly where it was drawn. The arithmetic is straightforward and it should be done explicitly rather than by rule of thumb.

  • Moisture movement: panel length times the moisture strain. A 3 m cementitious panel at 1.0 mm per metre gives 3 mm
  • Thermal movement: panel length times the coefficient times the design temperature range. A 3 m panel at 15 × 10⁻⁶ per °C over a 60 °C swing gives about 2.7 mm
  • Structural movement: frame shortening, creep and deflection of the supporting structure, and inter-storey drift where relevant
  • Manufacturing tolerance on the panel, and site tolerance on the structure and the substructure
  • Sealant capability: a joint sealant has a movement accommodation factor, commonly 25 percent for a good silicone, so a joint that must accommodate 6 mm of movement needs to be at least 24 mm wide if the sealant is the only mechanism

That last point is the one most often missed. A joint is not sized by movement alone; it is sized by movement divided by what the sealant can tolerate. This is a strong argument for open jointed rainscreen construction with moulded panels, where the joint is a designed gap with a baffle or a returned edge rather than a line of sealant working at the edge of its capacity.

Joint width is set by the longest dimension of the panel, not the average. Halving a panel halves the movement across each joint, which is one reason a facade of many smaller panels is easier to detail than one of few very large ones.

Why anchors have to be flexible

A panel that cannot move will crack. That is the whole design principle behind flexible anchors. A typical glass fibre reinforced concrete panel is supported by gravity anchors at two points, which carry its weight, and restrained by flex anchors elsewhere, which carry wind load perpendicular to the panel but deliberately allow in-plane movement. The panel expands and contracts around its gravity points and the flex anchors bend to follow it.

Get this wrong and the failure modes are consistent and well documented: cracking radiating from anchor positions, cracking at re-entrant corners and around openings, spalling at panel edges where two rigidly fixed panels have closed on each other, and sealant failure in joints that have been asked to move more than they can.

Specifying and testing it

BS EN 1170-7 measures the extremes of dimensional variation due to wetting and drying: a coupon is saturated, measured, dried to constant mass and measured again, and the difference is the reversible moisture movement. It is a slow test and it is not usually run on every batch, so most projects use the manufacturer’s published value with the batch test record for density and absorption as corroboration, on the reasonable basis that the panel that absorbs less also moves less.

A good specification asks for three things: the reversible moisture movement value with the test method and condition, the coefficient of thermal expansion, and confirmation that the joint widths shown on the shop drawings have been calculated from those values plus the tolerance stack rather than assumed. It is worth asking for the calculation, because a facade with 10 mm joints and 6 m panels has a problem that is far cheaper to find on a drawing than on a building.

Standards and references

  • BS EN 1170-7 Precast concrete products, Test method for glass-fibre reinforced cement, Measurement of extremes of dimensional variations due to wetting and drying
  • BS EN 1170-6, water absorption and dry density, which correlate with moisture movement
  • GRCA Specification for the Manufacture, Curing and Testing of GRC Products, and GRCA guidance on movement and fixings
  • AS 3600 Concrete structures, shrinkage and creep provisions
  • ISO 11600 and ASTM C920, classification of building sealants including movement accommodation factor
  • AS/NZS 4284 Testing of building facades, where joint performance is proven on a mock-up
  • AS/NZS 1170.2 wind actions, for the out of plane loads the flexible anchors must still carry

Related terms

Questions specifiers ask

How much does a GRC panel move with moisture?

Typically 0.05 to 0.15 percent, or 0.5 to 1.5 mm per metre, between saturated and dry. A design value of about 1.0 mm per metre is commonly used, but confirm it against the supplier’s tested figure.

Is moisture movement bigger than thermal movement?

For cementitious panels, usually yes. A 3 metre panel moves about 3 mm with moisture and about 2.7 mm over a 60 °C temperature swing, so the two are comparable and both have to be added into the joint.

How wide should facade joints be?

Wide enough for moisture movement, thermal movement, structural movement and the tolerance stack, divided by the movement accommodation factor of the sealant. Ten to twenty millimetres is typical, but the number should come from the calculation, not from a rule of thumb.

Why can a panel not be bolted rigidly at every fixing?

Because it changes size. Gravity anchors carry the weight at two points and flexible anchors carry wind load while allowing in-plane movement. Rigid restraint at multiple points produces cracking from the anchor positions outwards.

Does FRP have moisture movement?

Negligibly. There is no capillary pore network to fill, so an FRP panel moves with temperature rather than humidity, and it moves more thermally than a cementitious panel does.

Joint widths should be calculated from tested movement values and the tolerance stack. ShapeShift issues panel movement and fixing data with every shop drawing package. Talk to the ShapeShift team

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