The reduced cross-section method (RCSM) is included in Eurocode 5 (EN-1995-1-2) for the design of timber members in fire conditions. The method considers the strength and stiffness reduction beneath the charred layer by adding an additional depth (known as the 'zero-strength' layer) to the charring depth. The zerostrength layer is one of the key parameters for the fire design of timber members. Recently, some concerns have been raised that the zero-strength layer might be nonconservative in some applications. This paper presents the background to the RCSM, followed by a short discussion on the mechanical assumptions, simplifications and possible limitations of the method itself. Further, it discusses determination of the zero-strength layer thickness for members in bending, tension and compression, and provides guidelines on the use of standard experimental tests to determine this quantity. For demonstration of the determination procedure, the results of fire tests in bending, tension and compression were analysed following the described procedure. Results show that the zero-strength layer exceeds the value used in practice, indicate that the method of Eurocode 5 may be non-conservative and should be revised.

No abstract

No abstract

The effects of surface pretreatment (water and alkali) and modification with silane on moisture sorption, water resistance, and reaction to fire of hemp fiber reinforced polylactic acid (PLA) composites at two fiber loading contents (30 and 50 wt.%) are investigated in this work. Moisture adsorption was evaluated at 30, 50, 75 and 95% relative humidity, and water resistance was determined after a 28-day immersion period. The cone calorimetry technique was used to investigate response to fire. The fiber surface treatment resulted in the removal of cell wall components, which increased fiber individualization and homogeneity as shown in scanning microscopic pictures of the composite cross-section. Although the improved fiber/matrix bonding increased the composite’s water resistance, the different fiber treatments generated equal moisture adsorption results for the 30 wt.% reinforced composites. Overall, increasing the fiber amount from 30 to 50 wt.% increased the composite sensitivity to moisture/water, mainly due to the availability of more hydroxyl groups and to the development of a higher pore volume, but fire protection improved due to a reduction in the rate of thermal degradation induced by the reduced PLA content. The new Oswin’s model predicted the composite adsorption isotherm well. The 30 wt.% alkali and silane treated hemp fiber composite had the lowest overall adsorption (9%) while the 50 wt.% variant produced the highest ignition temperature (181 ± 18 °C).

No abstract

The load resistance of timber members exposed to fire is determined from the uncharred residual cross section. Owing to elevated temperatures in parts of the residual cross section, the strength and stiffness properties are lower than under normal conditions. The effective residual cross-section model provides a simplified user -friendly design concept to account for the reduced properties of timber exposed to fire. A fictitious zero-strength layer is removed from the residual cross section obtain ed after removal of the char layer, and the remaining cross section is assumed to have normal strength and stiffness properties. The method is implemented in Eurocode 5 as the reduced cross-section method. This paper deals with the background of this method, originally developed for rectangular cross sections of glued laminated tim ber, and shows extensions to other types of cross sections such as solid timber frame members and I-joists. While the thickness of the zero-st rength layer was originally given as 7,6 mm, the results of simulations presented here show that the thickness of the zero-strength layer depends on a number of parameters, such as the dimensions and geometry of the cross section, the stress conditions (compression or tension) of the fire-expose d side(s), the load ratio and the duration of fire exposure. It is concluded that the assumption of a fixed value of 7,6 mm is often non-conservative.

Cross-laminated timber, typical abbreviations CLT or XLAM, is currently one of the most innovative product in building with wood. This solid engineered timber product provides advantages compared to other solid timber slabs as the dimension stability, i.e. swelling and shrinkage, is controlled by the crosswise laminations. As for other components, the fire resistance has to be verified for this type of product. While fire testing is time consuming and costly, simulations provide flexibility to optimize the product or to develop simplified design models for structural engineers. In this paper, a simulation technique is presented which can be used to determine the fire resistance of CLT. The technique was then used to develop simplified design equations to be used by engineers to predict the behavior of CLT in fire resistance tests and verify its fire resistance. Following existing models, the simplified design model aims for a two-step process whereby in a (i) first step the residual cross section and in (ii) a second step the load bearing capacity of the partly heated residual cross section is determined. The presented simulations consider the effective thermal-mechanical characteristics of wood exposed to standard fire and perform an advanced section analysis using a temperature profile corresponding to the actual protection and the location of the centroid together with the possibility of plasticity on the side of compression. It was shown that simulation results agree well with test results and that they can be used to determine layup specific modification factors used by the reduced properties method or zero-strength layers used by the effective cross section method. It was shown that the use of the zero-strength layers is favorable compared to the modification factors to calculate the resistance of the residual cross section. This is due to the large range of modification factors answering the typical layup of CLT comprising layers with their fiber direction cross the span direction. Subsequently, the methodology was used to determine design equations for initially unprotected and protected three-, five-and seven-layer CLT in bending and buckling. While the zero-strength layer for glulam beams in bending is assumed to be 7 mm (0.3 in), for CLT the corresponding value is in most of the cases between 5 mm and 12 mm but is different for other loading modes such as buckling (wall elements) and depending on the applied protection.

The fire resistance of glued-laminated timber (glulam) beams is usually determined by calculation models or by tests. The evaluation of the fire resistance tests is accompanied by an uncertainty with regard to the timber material properties, which is, however, important to know when comparing the test results to the effective crosssection method according to EN 1995-1-2. The present paper presents therefore for the first time the results of six fire resistance tests with glulam beams with specific local mechanical material properties, such as exact position and dimension of knots, weak sections and finger joints, densities and strength properties of small cells. The tested glulam beams had an approximate height of 250 mm and reached a fire resistance of up to 68 minutes. The performed fire tests allowed the determination of the zerostrength layer with a higher certainty than investigations performed in the past. Additionally, reference tests at ambient temperature were considered in the evaluation. The determined zero-strength layer was in the range of 7 mm, as given in the current version of EN 1995-1-2 for beams in bending.