Fibre Reinforced Facade Encyclopedia: Properties
Creep is the slow, time dependent deformation of a material held under a constant load. It is not failure and it is not damage; it is the material continuing to move long after the load stopped changing. In facade design it matters because the loads that persist are the ones that cause it. Wind comes and goes in gusts, but a panel carries its own weight, and any prestress or permanent restraint, for the entire design life. Polymer composites creep significantly under sustained load. Cementitious composites creep too, though the mechanism and the consequences are different.
A wind gust lasts three seconds. Dead load lasts fifty years. It is the second one that decides how much a panel will have sagged by the time somebody notices.
At a glance
| Property | Typical value | Note |
|---|---|---|
| Mechanism, polymer composites | Viscoelastic flow of the resin matrix | Fibres do not creep; the resin does |
| Mechanism, cementitious composites | Moisture migration in the cement gel plus matrix densification | Coupled with shrinkage and hard to separate |
| Creep modulus, thermoset GFRP | Effective modulus reduced to roughly 50 to 70% of short term at 50 years | Typical; resin and temperature dependent |
| Creep coefficient, concrete and cementitious | Typically 1.5 to 3.0 long term | Ratio of creep strain to initial elastic strain; typical |
| Governing variable | Sustained stress as a fraction of capacity | Keep sustained stress well below short term capacity |
| Temperature sensitivity, polymers | Creep accelerates sharply approaching glass transition | Dark facades reach 60 to 80 °C surface; typical |
| Moisture sensitivity | Wet cementitious material creeps more | Drying creep exceeds basic creep |
| Design response | Use long term modulus, not 28 day or short term values | Deflection limits checked against the reduced modulus |
| Test methods | ASTM D2990, ISO 899, BS EN 1170-7 | Sustained load creep and dimensional change |
Why polymer composites creep and glass does not
In a fibre reinforced polymer laminate the glass fibre is elastic and effectively does not creep at service temperatures. The resin does. A thermoset polymer is viscoelastic: under sustained stress the cross-linked network slowly rearranges and the laminate extends, and the extension does not fully recover when the load is removed.
That means creep in a laminate depends on how much of the load the resin is carrying. Stress applied along continuous fibres is carried almost entirely by the fibre, so creep is small. Stress carried in shear through the matrix, across the fibres, or in a chopped strand laminate where fibre orientation is random, is carried substantially by the resin, so creep is much larger. A laminate schedule designed for the load path is therefore also a creep control measure.
Temperature is the multiplier. Creep rate in a thermoset rises steeply as the material approaches its glass transition temperature, and a dark facade panel in Australian sun reaches surface temperatures of 60 to 80 °C. A creep check performed at 20 °C on a panel that will spend summer afternoons at 70 °C is not a check. ShapeShell-RT is published at 221 to 241 MPa in flexure; that is a short term strength figure and, like all short term figures, it is not the number to use for a sustained load case.
Where it actually bites in a facade
- Self weight deflection of a panel spanning between fixings, particularly a large or slender FRP panel where the initial deflection is already close to the limit
- Cantilevered elements: fins, blades, sunshades, copings and signage brackets, where dead load acts at a lever arm permanently
- Bolted and bonded connections: a bolt through a laminate relaxes as the material creeps under the clamping force, and preload is lost
- Bonded joints and adhesive fixings, where the adhesive itself creeps and the joint slowly moves
- Panels restrained against thermal or moisture movement, where the restraint force is sustained rather than transient
- Permanent formwork shells that carry load until the structural concrete takes over
Cementitious composites
Cementitious materials creep as well, through a different mechanism: moisture migrating within the cement gel under stress, plus continuing densification of the matrix. Creep coefficients of 1.5 to 3.0 are normal for concrete, meaning the long term creep strain is one and a half to three times the initial elastic strain.
In a thin facade panel this is usually a second order effect, because the sustained stresses are low. A 15 mm glass fibre reinforced concrete (GRC, GFRC) panel hanging from flexible anchors carries its self weight through the anchors, not through sustained bending across a long span, so creep deflection is small. Where it matters is in restrained conditions: a panel fixed rigidly at multiple points and prevented from moving will relax some of the restraint stress through creep, which is helpful, and will also accumulate permanent deformation, which is not.
Creep and shrinkage are hard to separate in cementitious materials, because drying causes both simultaneously. Standard test methods measure the combination and separate them by running a companion specimen under no load. For facade design the practical response is the same either way: allow the panel to move on flexible anchors, set joint widths to accommodate the total movement, and do not detail restraint that the panel has to fight for fifty years.
The design check
The approach is uniform across materials even though the numbers differ. Establish the sustained load case, which is self weight plus any permanent restraint or prestress, excluding wind. Apply a reduced long term modulus rather than the short term value: for a thermoset laminate this is commonly the short term modulus reduced to somewhere in the range of half to seventy percent at fifty years, taken from the resin supplier’s creep data or from sustained load testing. Check deflection against the serviceability limit using that reduced modulus, and check any connection that relies on preload for relaxation.
Then apply the design temperature. A creep modulus derived at 20 °C should be adjusted for the service temperature the panel will actually reach, using the resin data or a time temperature superposition approach. Two further habits are worth keeping. Keep sustained stress to a modest fraction of short term capacity, which is the single most effective creep control available, and prefer fibre dominated load paths over matrix dominated ones in the laminate schedule.
Standards and references
- ASTM D2990 Standard Test Methods for Tensile, Compressive, and Flexural Creep and Creep-Rupture of Plastics
- ISO 899 parts 1 and 2, plastics, determination of creep behaviour
- BS EN 1170-7 Test method for glass-fibre reinforced cement, Measurement of extremes of dimensional variations due to wetting and drying
- AS 3600 Concrete structures, creep and shrinkage provisions for cementitious materials
- ASTM C512, creep of concrete in compression
- ISO 6721 and ASTM E1640, dynamic mechanical analysis for glass transition temperature of thermosets
- AS/NZS 1170.0, structural design actions, general principles including serviceability limit states
Related terms
Questions specifiers ask
Does FRP creep under load?
Yes. The glass fibre does not creep meaningfully at service temperatures but the resin matrix does, so a laminate held under sustained load continues to deform. Design uses a reduced long term modulus rather than the short term value.
Is creep a strength problem or a deflection problem?
Almost always deflection. Panels rarely fail from creep; they sag, joints go out of line and preloaded connections relax, which shows up as a serviceability and appearance problem years after handover.
Does temperature make creep worse?
Substantially, in polymers. Creep rate rises steeply as a thermoset approaches its glass transition temperature, and dark facade panels reach surface temperatures of 60 to 80 °C, so the check has to be done at the design service temperature.
Do cementitious panels creep?
Yes, through moisture migration in the cement gel and continuing matrix densification, with long term creep coefficients typically 1.5 to 3.0. In a thin panel hung on flexible anchors the sustained stresses are low, so the effect is usually second order.
How is creep controlled in design?
Keep sustained stress well below short term capacity, design the laminate so load runs along fibres rather than through the matrix, use a reduced long term modulus for deflection checks, and avoid detailing permanent restraint into the panel.
Long term deflection under sustained load is a design case, not a footnote. Ask for creep data before sizing a slender or cantilevered panel. See ShapeShell-RT specifications
Reviewed by Richard Halvorsen, General Manager, ShapeShift Technologies. Last updated September 2026.