Thermal expansion coefficient testing of extruded polystyrene (XPS) insulation boards is a critical quality control step for underfloor (radiant) heating systems, where insulation layers undergo repeated thermal cycling between room temperature and 40–70 °C. This article explains the testing principles, specimen requirements, test items, methods based on ISO 11359-2, ASTM E831, ASTM D696, GB/T 10801.2 and EN 13164, typical acceptance criteria, and why accredited third-party laboratories perform these evaluations.
Technical Overview
Extruded polystyrene (XPS) foam boards are the dominant thermal insulation layer in modern underfloor heating systems, owing to their closed-cell structure, low thermal conductivity (typically 0.028–0.035 W/(m·K)), high compressive strength, and excellent moisture resistance. In service, the boards are sandwiched between the heated screed above and the structural slab below, and they are repeatedly cycled between room temperature and the heating medium temperature, which commonly reaches 40–60 °C and may peak above 70 °C in high-temperature systems.
The coefficient of linear thermal expansion (CLTE, symbol α) describes the fractional change in length of a material per unit temperature change, expressed in 1/K or µm/(m·K). It is calculated as α = ΔL / (L₀ × ΔT), where L₀ is the initial length, ΔL is the length change, and ΔT is the temperature change. For rigid polystyrene foams, the in-plane CLTE typically falls between 5 × 10⁻⁵ /K and 8 × 10⁻⁵ /K (50–80 µm/(m·K)). If the thermal expansion of the insulation boards exceeds the tolerances assumed in the floor build-up design, it can force joints open, transfer stress into the screed and floor coverings, displace heating pipes, and ultimately cause cracking, lippage, or hollow tiles. Systematic thermal expansion coefficient testing therefore forms an essential part of the quality evaluation of XPS boards for underfloor heating, complemented by the dimensional stability requirements of standards such as GB/T 10801.2, EN 13164, and EN 1604.
Testing Samples
Samples are cut from finished boards at least 50 mm away from edges, free of visible defects, and clearly marked with the machine direction (MD) and transverse direction (TD) because extrusion imparts directional cell orientation that affects expansion behavior. Specimens are conditioned in a standard atmosphere of 23 ± 2 °C and 50 ± 5 % relative humidity for at least 24 hours (48 hours preferred for thick boards) in accordance with ISO 291 or GB/T 2918 before measurement.
Specimens for Thermomechanical Analysis (TMA)
- Length typically 5–10 mm with an approximately 3–5 mm square cross-section, as suitable for ISO 11359-2 and ASTM E831 geometries
- Flat, parallel end faces to ensure uniform load transfer from the probe
- Cut from the board core, or with skins retained when whole-thickness representation is required by the specification
- Minimum of three specimens per direction (MD and TD)
Specimens for Quartz Dilatometer Testing
- Length between 50 mm and 125 mm with a uniform rectangular cross-section, per ASTM D696
- Ends machined flat and perpendicular to the measurement axis
- Minimum of three specimens per direction
Specimens for Dimensional Stability Verification
- 200 mm × 200 mm × full original thickness, following GB/T 8811 and EN 1604 practice
- Minimum of three specimens with marked measurement points for length, width, and thickness
Testing Items
Depending on the project specification and certification scope, the following items are typically evaluated:
- Mean linear coefficient of thermal expansion (CLTE) over the underfloor heating service range, for example 23–60 °C
- Extended-range CLTE, for example −30 °C to +80 °C, for complete material characterization
- Directional CLTE in the machine direction (MD) and transverse direction (TD), and through-thickness expansion where required
- Dimensional stability at elevated temperature, commonly 70 ± 2 °C for 48 hours, reported as percentage change in length, width, and thickness
- Optional thermal cycling endurance, such as repeated 20–70 °C cycles with dimensional measurement after each cycle
Testing Methods
Thermomechanical Analysis (TMA)
The specimen is placed on a quartz platform and contacted by a low-mass quartz probe under a negligible load, with dry nitrogen purging to prevent oxidation and moisture interference. The specimen is equilibrated at the starting temperature, then heated at a controlled rate of 3–5 K/min to the endpoint temperature while the dimensional signal is recorded continuously. The mean CLTE is derived from the slope of the length-versus-temperature curve between the defined temperature limits. This approach, conforming to ISO 11359-2 and ASTM E831, offers high sensitivity and allows expansion behavior to be mapped across multiple temperature intervals within a single run.
Quartz Dilatometry
The specimen is installed in a vitreous silica dilatometer, and its length is measured after thermal equilibrium is reached at two controlled temperature points, commonly −30 °C and +30 °C as specified in ASTM D696. The mean CLTE is calculated as ΔL / (L₀ × ΔT). Because the specimens are larger than TMA coupons, dilatometry is often considered more representative of bulk board behavior.
Dimensional Stability Test at Elevated Temperature
Initial dimensions are measured with calibrated instruments (typically 0.02–0.05 mm resolution). The specimens are then placed in a ventilated oven at 70 ± 2 °C for 48 hours, cooled to the standard atmosphere, and remeasured. The percentage changes in length, width, and thickness are compared against the limits of the applicable product standard, which for XPS boards are commonly in the range of 1.0–2.0 % linear change. This test captures irreversible shrinkage caused by residual expansion stresses and blowing-agent diffusion, which superimposes on the reversible thermal expansion measured by TMA or dilatometry.
Data Evaluation and Reporting
The test report presents individual and mean CLTE values for each direction and temperature interval, together with the test method, specimen orientation, conditioning history, and heating rate. Results are compared with the design assumptions: for example, a 2400 mm board with an in-plane CLTE of 7 × 10⁻⁵ /K exposed to a 40 K temperature swing will expand approximately 6.7 mm, which must be accommodated by perimeter gaps and expansion joints. Accredited third-party laboratories operating under an ISO/IEC 17025 management system issue reports suitable for engineering acceptance and product certification.
Applications
- Factory quality control and batch release of XPS boards destined for radiant floor heating systems
- Type testing and product certification against GB/T 10801.2, EN 13164, and related CE marking requirements
- Engineering acceptance and supervision of underfloor heating installations
- Design input for expansion joint spacing, fixing and bonding strategies, and perimeter insulation detailing
- Research and development on the influence of foam density, cell orientation, skin layers, and blowing agents on expansion behavior
- Failure diagnosis in cases of tile cracking, joint opening, floor heave, or screed stress
FAQ
Q: What is a typical coefficient of thermal expansion for XPS underfloor heating boards?
A: The in-plane CLTE of XPS boards typically lies between 5 × 10⁻⁵ /K and 8 × 10⁻⁵ /K (50–80 µm/(m·K)), with some variation between the machine direction and transverse direction due to cell orientation during extrusion.
Q: Which temperature range should the test cover?
A: For underfloor heating applications, the service range of approximately 20–60 °C is the primary interval. An extended range such as −30 °C to +80 °C is recommended for full characterization, particularly where floors are exposed to unheated conditions during construction or seasonal shutdown.
Q: Why is dimensional stability testing needed in addition to CLTE measurement?
A: CLTE quantifies reversible thermal expansion, while dimensional stability testing at elevated temperature reveals irreversible shrinkage from residual stresses and blowing-agent loss. Both mechanisms contribute to in-service dimensional change and must be controlled jointly.
Q: How many specimens are required?
A: A minimum of three specimens per direction is standard; five specimens per direction improve statistical confidence and are recommended for certification-level testing.
Q: What is the typical turnaround time?
A: Including conditioning, testing, and reporting, most accredited third-party laboratories complete CLTE and dimensional stability evaluations within 5–10 working days.
