Methods for Determining the Thermal Conductivity of Insulating Glass Units: Standards, Procedures, a

The thermal conductivity of insulating glass units (IGUs) is the key parameter governing the energy performance of windows, curtain walls, and glazed façades. This article presents the professional laboratory and calculation methods used to determine the thermal conductivity and U-value of insulating glass, including the guarded hot plate method, heat flow meter method, hot box method, and standardized numerical calculation. Architects, glazing manufacturers, and testing engineers can use this guide to select the appropriate procedure, prepare valid specimens, and interpret results in accordance with international standards.

Technical Overview

An insulating glass unit (IGU) consists of two or more glass panes separated by a hermetically sealed cavity formed by a spacer bar and dual sealants. The cavity is filled with dry air or an inert gas such as argon, krypton, or xenon. Because the thermal conductivity of static gas is roughly thirty to sixty times lower than that of soda-lime glass (approximately 1.0 W/(m·K)), the sealed gas gap provides the dominant thermal resistance of the assembly, and its accurate characterization is essential for predicting building energy performance.

Two related but distinct quantities must be understood when measuring insulating glass. Thermal conductivity, expressed in W/(m·K), is a material property describing heat flow through a homogeneous layer per unit thickness. Thermal transmittance, or the U-value, expressed in W/(m²·K), describes the heat flow through the complete glazing system including all glass panes, gas gaps, coatings, and surface film resistances. For glazing products, standards typically define measurement of the U-value, from which an effective conductivity of the assembly can be derived.

The principal international and regional standards governing this work include:

• ISO 10291 and EN 674: Determination of the steady-state U-value of multiple glazing by the guarded hot plate method
• ISO 8301 and EN 675: Determination of the U-value by the heat flow meter method
• EN 673 and ISO 10292: Determination of the U-value by numerical calculation
• ISO 12567-1, ASTM C1199, and ASTM E1423: Hot box determination of thermal transmittance for complete window and door assemblies
• ISO 8302 and ASTM C177: Guarded hot plate apparatus fundamentals
• ASTM C518: Heat flow meter apparatus fundamentals
• ISO 15099 and NFRC 100: Detailed calculation procedures for fenestration thermal performance
• GB/T 22476 and GB/T 8484: Chinese national standards for calculating and testing the thermal transmittance of building glass and windows

Testing Samples

Double-Glazed Units

Standard double-glazed specimens consist of two panes, commonly 4 mm to 6 mm thick, separated by a 6 mm to 20 mm gas cavity. Specimens are typically square panels of at least 500 mm × 500 mm for plate-type apparatus, and larger sizes are preferred because a wider specimen reduces the relative influence of edge heat transfer on the measured result. The edge seal must be intact, since a leaking cavity will change the gas composition and invalidate the measurement.

Triple-Glazed and Multi-Cavity Units

Triple-glazed units contain two sealed cavities and often employ asymmetric pane thicknesses, for example 5 mm–12 mm airgap–4 mm–16 mm argon gap–5 mm constructions. Multi-cavity units take longer to reach thermal steady state due to their greater thermal mass, and the test report should clearly identify the cavity sequence, spacer type, and cavity widths so the result can be compared with calculated values.

Coated and Gas-Filled Units

Low-emissivity coated units must be documented with the coating position, for example surface 3 of a double-glazed unit, because the radiant heat exchange across the cavity depends directly on coating emissivity. Gas-filled units should be accompanied by gas fill analysis, typically by gas chromatography, since the actual argon concentration can deviate from the nominal fill and significantly affect the measured U-value.

Specimen Preparation and Conditioning

Before testing, specimens should be visually inspected for seal defects, checked for dimensional tolerance, and conditioned in a laboratory environment for at least 24 hours to reach moisture and temperature equilibrium. Thermocouples or other surface temperature sensors are attached to both glass faces within the metering region, and the specimen edges are often masked with low-conductivity insulation to suppress lateral heat exchange between the cavity and the laboratory ambient air.

Testing Items

A complete thermal characterization program for insulating glass typically includes the following measured or derived parameters:

• Steady-state thermal transmittance (U-value) of the glazing, in W/(m²·K)
• Effective thermal conductivity of the gas cavity, in W/(m·K)
• Thermal resistance (R-value) of the glazing assembly, in m²·K/W
• Surface temperatures of the hot-side and cold-side glass faces under steady-state conditions
• Heat flux density through the metering area, in W/m²
• Normal emissivity of coated surfaces, which controls radiative transfer across the cavity
• Cavity gas concentration, determined by gas chromatography for argon, krypton, or mixed fills
• Auxiliary durability indicators such as dew point, which confirm the cavity is dry and sealed

For product certification, the measured U-value is normally compared against the value calculated by EN 673 or ISO 10292, and agreement within the tolerance specified by the certification scheme demonstrates that the product design and manufacturing process are under control.

Testing Methods

Guarded Hot Plate Method

The guarded hot plate method, specified by ISO 10291, EN 674, ISO 8302, and ASTM C177, is the primary reference technique for glazing. The specimen is sandwiched between an electrically heated hot plate and a temperature-controlled cold plate. A central metering section is surrounded by a guard section maintained at the same temperature, which cancels lateral heat flow and forces one-dimensional heat transfer through the metering area. At steady state, the heat flow Q, the metering area A, the specimen thickness d, and the surface temperature difference ΔT are used to compute the thermal properties, with U = Q / (A · ΔT) after correcting for the standardized surface film resistances. The method offers the highest accuracy, with typical uncertainties of ±2 to 4 percent.

Heat Flow Meter Method

The heat flow meter method, covered by ISO 8301, EN 675, and ASTM C518, replaces the direct electrical power measurement with calibrated heat flux transducers mounted on one or both sides of the specimen. The apparatus must first be calibrated using certified reference materials traceable to national metrology institutes. The technique is faster and well suited to routine quality control, with typical uncertainties of ±3 to 5 percent, but its accuracy depends directly on the stability of the calibration.

Hot Box Method for Complete Fenestration

When the performance of an entire window, curtain wall section, or façade element is required, the calibrated or guarded hot box method per ISO 12567-1, ASTM C1199, and ASTM E1423 is used. The specimen is mounted between a climatic chamber maintained at a cold temperature, for example −18 °C, and a metering chamber at approximately 20 °C. Measured metering power, surface temperatures, and area allow determination of the assembly U-value, including frame, spacer, and edge-of-glass effects that plate methods exclude. Standard specimen sizes such as 1230 mm × 1480 mm are commonly used for operable windows.

Numerical Calculation Method

EN 673 and ISO 10292 define a widely used calculation procedure in which the total thermal resistance is assembled from the internal and external surface resistances (conventionally 0.13 and 0.04 m²·K/W), the resistances of the glass panes, and the resistances of the gas gaps. Within each gap, the effective conductivity equals the product of the gas conductivity and the Nusselt number, which accounts for convection and is derived from the Rayleigh and Prandtl numbers and the gap aspect ratio. Radiative exchange is governed by the emissivities of the gap surfaces. Typical gas conductivities at 10 °C are approximately 0.025 W/(m·K) for air, 0.017 W/(m·K) for argon, 0.0094 W/(m·K) for krypton, and 0.0054 W/(m·K) for xenon, which explains the strong performance benefit of heavy noble gas fills.

Test Procedure and Conditions

Regardless of the apparatus, a reliable determination follows the same logical sequence:

• Verify specimen identity, dimensions, cavity width, coating position, and gas fill against the test request
• Condition the specimen and instrument both surfaces with calibrated temperature sensors
• Apply a controlled temperature difference across the specimen, typically 10 to 20 K with a mean temperature near 10 °C as specified by the governing standard
• Monitor heat flow and temperatures until steady state is achieved, which may require several hours for triple glazing due to thermal mass
• Compute the U-value, correct for standardized boundary conditions, and compare with the EN 673 calculated value where required
• Report the method, apparatus, boundary temperatures, gas concentration, and measurement uncertainty

Applications

Thermal conductivity and U-value determinations for insulating glass support a wide range of engineering and commercial activities:

• Compliance demonstration for building energy codes in China, the European Union, North America, and other jurisdictions that impose maximum glazing U-values
• Input data for whole-building energy simulation and HVAC load calculation in curtain wall and façade projects
• Green building certification programs such as LEED, BREEAM, and China's green building evaluation standard, where envelope thermal performance earns credit
• Type testing and CE marking, CCC certification, and third-party product certification of IGU product families
• Factory quality control to verify spacer, coating, and gas-fill consistency between production lots
• Research and development of low-emissivity coatings, warm-edge spacers, vacuum glazing, quadruple glazing, and thin triple glazing
• Forensic evaluation of field failures, where a measured U-value higher than the design value can reveal seal leakage and gas loss

Typical measured center-of-glass U-values include approximately 2.8 W/(m²·K) for a clear double-glazed unit with an air cavity, 1.1 W/(m²·K) for a double-glazed low-emissivity unit with argon, and 0.6 to 0.8 W/(m²·K) for a triple-glazed unit with two low-emissivity coatings and argon fill.

FAQ

What is the difference between thermal conductivity and the U-value of insulating glass?

Thermal conductivity is a material property of a single layer, expressed in W/(m·K), while the U-value describes the complete glazing assembly, including all panes, cavities, coatings, and surface films, expressed in W/(m²·K). Standards for insulating glass primarily determine the U-value, and an equivalent effective conductivity can be back-calculated from it when required.

Which measurement method is the most accurate?

The guarded hot plate method is generally regarded as the reference technique, achieving uncertainties of ±2 to 4 percent because heat flow is measured directly by electrical power. Heat flow meter and hot box methods are practical for routine and assembly-level testing but depend on calibration transfer.

How long does a typical test take?

Reaching thermal steady state typically requires three to eight hours for double glazing and can exceed eight hours for triple-glazed or multi-cavity units. Total turnaround, including conditioning, instrumentation, and repeated steady-state verification points, is usually one to two working days.

Why must gas fill concentration be checked alongside thermal testing?

The cavity gas provides most of the thermal resistance, and leakage or incomplete filling reduces the gas conductivity benefit. A unit nominally filled with 90 percent argon but actually containing mostly air can show a U-value 10 to 20 percent higher than designed, so gas chromatography verification accompanies credible measurements.

How do measurement results compare with calculated values?

When specimen construction, coating emissivity, and gas fill are accurately documented, measured U-values normally agree with EN 673 or ISO 10292 calculations within approximately 0.1 W/(m²·K). Larger deviations usually indicate construction errors, coating misplacement, or gas loss, and should trigger investigation of the production process.