Fire Spread Through External Walls of Wooden Materials in Multi-Story Buildings—Part I

Abstract

The increasing use of wooden cladding in multi-storey buildings raises critical fire safety concerns, especially in ventilated façade systems where the chimney effect can accelerate vertical flame spread. This study combines theoretical analysis with three full-scale fire tests to investigate key factors influencing fire propagation, including the influence of façade design details. Results show that poorly constructed lintels and jambs significantly accelerate flame entry into ventilated cavities, while wooden fire barriers—despite being combustible—can delay flame spread if properly installed. These findings inform design recommendations and underscore the need for more robust fire safety strategies in modern timber construction.

1. Introduction

1.1. Research Significance

In recent decades, the construction industry has increasingly focused on the use of renewable materials. Wood, as a renewable material, significantly contributes to sustainable construction and environmental protection. A significant advantage of wood as a construction material lies in its sustainability and ability to reduce the carbon footprint of buildings. Additionally, wood offers excellent thermal insulation properties, leading to more energy-efficient buildings. Wooden structures have been traditionally used in construction for centuries, particularly in North America, Scandinavia, and other regions with abundant forest resources. Thanks to modern technologies, wood can now be processed into various forms, making it suitable not only for the construction of single-family houses but also for multi-storey buildings and complex architectural elements.

Recently, this trend has also spread to the Czech Republic, where the proportion of timber buildings is increasing, both in the construction of single-family houses and in the construction of multi-storey timber building [1,2]. In the Czech Republic, buildings are classified as multi-storey when their fire height exceeds 12 m (approximately five storeys) according to [3]. Timber construction is permitted only up to this threshold. Their use above 12 m remains limited, primarily due to stringent fire-safety requirements.

The use of wood in façade systems presents a natural choice for sustainable and architecturally attractive buildings but requires careful consideration of fire safety, as façades are among the most vulnerable parts of a building. Façades may contain various types of combustible materials, including combinations thereof. Each product must be tested for fire resistance and subsequently classified based on their contribution to fire spread and intensity [3,4,5].

While timber construction offers numerous benefits, fire safety remains a significant challenge, particularly in the context of external walls and ventilated façade systems. The chimney effect, which occurs within ventilated cavities, has been identified as a critical driver of vertical fire propagation [6]. Research indicates that fires in ventilated cavities can spread up to ten times faster than open fires, with fourteen times greater heat flux and significantly increased temperatures [6]. Fire incidents such as the Grenfell Tower disaster in 2017, which resulted in 72 fatalities, highlight the urgent need for improved fire prevention strategies in ventilated façade systems [7].

Despite advancements in fire-resistant treatments and engineered timber materials, ventilated façade cavities introduce additional risks that current regulations struggle to address. The Czech national fire-safety regulations currently set only general requirements for ventilated façade systems and do not provide detailed guidance on cavity compartmentation, barrier spacing, or ventilation control. This stands in contrast to the more prescriptive regulatory frameworks implemented in Switzerland, Austria, and Germany, where standards such as DIN [8] and VKF guidelines explicitly define cavity barrier intervals, opening limitations, and fire-resistance requirements for façade components. Highlighting this difference underlines why additional experimental evidence is necessary in the Czech context to inform future updates to national codes.

Studies show that inadequate fire barriers, combustible insulation, and poorly designed ventilation systems exacerbate fire spread [9]. Furthermore, cavity leaks and unintended ventilation gaps can compromise fire protection measures, allowing flames and hot gases to bypass barriers [9]. Research on buoyant turbulent jet flames confined by parallel walls [10] confirms that air entrainment plays a crucial role in fire spread. Narrow cavities with restricted airflow increase the flame height and intensity, further accelerating the fire due to increased turbulence and heat flux. These findings reinforce the idea that cavity design and ventilation control are essential in mitigating fire hazards [6].

Hietaniemi et al. [11] emphasize that ventilated cavities in timber façades present a significant fire hazard due to their potential to act as vertical channels for flame propagation, driven by the chimney effect. Their findings, based on large-scale fire testing and case studies of Nordic multi-storey timber buildings, demonstrate that poorly segmented or inadequately sealed cavities can significantly accelerate vertical fire spread, particularly around façade openings. The study highlights the limited reliability of fire barriers when improperly installed or exposed to high thermal loads, and stresses the necessity of using non-combustible, thermally stable materials in cavity compartmentation. Moreover, it underscores the importance of controlling airflow within cavities during fire exposure through optimized geometry and the integration of reactive elements such as intumescent seals. Crucially, Hietaniemi et al. [11] advocate for performance-based fire design and standardized large-scale testing, asserting that small-scale methods fail to capture the complex interactions of materials and ventilation effects in real façades. Their work supports the need for harmonized European regulations and validates the present study’s focus on cavity design, barrier effectiveness, and the role of ventilation in fire dynamics.

Engel and Werther [12] present a detailed technical investigation into how structural design elements can enhance the fire safety of rear-ventilated wooden façade systems. Their central finding is that fire resistance in such façades is highly dependent not only on the combustibility of materials but on the geometric detailing and integration of fire-resistant components within the system. A particular emphasis is placed on window openings, identified as the primary entry point for flames into the ventilated cavity. Once fire breaches the façade through a window, the rear ventilation layer can act as a chimney, rapidly accelerating vertical flame spread if not adequately compartmentalized.

To mitigate this risk, the authors recommend horizontal fire barriers installed at regular vertical intervals (such as floor levels) and especially at vulnerable points like window lintels and sills. These barriers must be constructed from non-combustible materials, and their effectiveness relies heavily on airtight installation and integration with the surrounding substructure. The paper also underlines the significance of non-combustible insulation, which, when used in tandem with cavity segmentation, substantially reduces the risk of uncontrolled flame propagation.

The study supports the use of timber cladding in multi-storey buildings but stresses that it must be approached through performance-based design, emphasizing that a façade’s fire behaviour is governed by the interaction of all layers and interfaces—cladding, cavity, substructure, and insulation—not by any single component alone. Engel and Werther [12] offer real-world design examples, including tested fire-safe configurations with protected window frames, integrated fire stops, and compartmented ventilated cavities. These demonstrate that fire-resilient design is achievable through a systems-based approach and careful detailing.

In conclusion, Engel and Werther [12] argue that wooden ventilated façades can meet modern fire safety requirements when equipped with thoughtfully designed and tested structural mitigation measures. Their research contributes significantly to bridging the gap between traditional material behaviour concerns and contemporary façade fire safety demands, offering a technically grounded framework for safer, sustainable timber façade applications.

In this study, we explicitly focus on ventilated timber façades with untreated spruce cladding of 19 mm thickness, investigating how ventilated cavity geometry and the installation of horizontal and vertical fire barriers affect fire spread. The experimental program consists of three large-scale façade-level tests conducted under controlled conditions, designed to replicate realistic fire exposure from a fully developed compartment fire. By referencing the relevant Czech standards [3,4,5] and the ISO [10] methodology for large-scale façade fire testing, the research situates the results within the broader regulatory context and ensures their practical relevance for building design and code development.

1.2. Aim and Objectives

This study aims to investigate fire spread mechanisms in ventilated timber façade systems by evaluating the role of cavity geometry, workmanship, and fire barriers. Through a combination of large-scale experimental fire testing and systematic review of existing regulatory standards, this research seeks to provide insights into optimizing façade design and strengthening fire-safety regulations for modern multi-storey timber construction. The findings are intended to contribute to enhancing the fire resilience of timber buildings and ensuring compliance with both national (ČSN) and international fire-safety standards (Eurocode, NFPA).

To achieve this aim, the study addresses the following research questions:

• How does cavity geometry affect the rate, height, and intensity of vertical flame spread?
• To what extent does the quality of workmanship in critical details (e.g., lintels, jambs) influence cavity ignition timing?
• Can horizontal and vertical fire barriers effectively delay or limit flame propagation within ventilated cavities?

The research design integrates three large-scale façade fire tests conducted under controlled conditions with a theoretical review of façade fire dynamics. These tests were designed to replicate realistic compartment fire exposure, record temperature profiles and flame heights, and compare different cavity and barrier configurations.

This paper represents Part I of a two-part study, presenting the experimental results and their interpretation. Part II will focus on numerical modelling of the experimental setup described here, enabling further parametric exploration and validation of the observed phenomena.

1.3. Behaviour of Façade Systems During Fire

Ventilated façade systems are among the most widely used cladding solutions in contemporary construction, combining aesthetic appeal with functional benefits [12]. Their multi-layered assembly—comprising the primary building envelope, insulation, vapor-permeable membranes, ventilated cavities, and external cladding—provides weather protection and enhances thermal performance. However, when combustible materials such as timber are employed, fire safety becomes a critical concern due to their susceptibility to rapid ignition and flame spread.

A key parameter governing fire behaviour in ventilated façades is cavity geometry and ventilation, as these directly influence the potential for flame acceleration through the chimney effect [6]. The chimney effect occurs when buoyancy-driven hot gases are channelled upward through the cavity, producing a strong vertical draft that significantly accelerates flame spread [6,10]. In combustible cladding systems, this mechanism is particularly hazardous. The intensified convective flow not only preheats the cladding above but also sustains timber pyrolysis, creating a continuous fuel source that amplifies fire growth [11,13].

In contrast, in non-combustible façades the cavity walls do not contribute additional fuel and may even act as thermal sinks, redistributing heat away from the fire [12]. Consequently, flame spread in non-combustible systems is driven primarily by the external fire source and airflow, whereas in combustible systems the chimney effect couples with ongoing material degradation to produce higher flame heights and faster vertical flame spread [14].

Experimental evidence indicates that flame heights inside ventilated cavities can significantly exceed those observed on the external surface [15,16]. This intensified behaviour is particularly hazardous in multi-storey buildings, where it can facilitate rapid vertical fire propagation to upper storeys.

Façade fires are generally driven by three mechanisms: radiant heat transfer, convective heat transfer, and direct flame impingement from external sources (e.g., burning vehicles or containers) or from internal compartment fires breaching openings such as windows or balconies [17,18]. Convective heat transfer, in particular, plays a dominant role, as rising hot gases significantly heat adjacent façade surfaces, as demonstrated in full-scale tests with both solid and gaseous fuels [19]. This mechanism has been experimentally shown to correlate with elevated wall surface temperatures during compartment fires [20,21]. Among these mechanisms, internal fire spread through windows is considered the most critical, since it can produce flame intensities reaching several megawatts, drastically increasing the likelihood of façade ignition.

Window opening geometry also strongly affects fire behaviour: larger opening angles (0–60°) increase ventilation and internal mixing, elevating compartment temperatures while simultaneously reducing external flame height [22]. These interdependencies highlight that both internal fire dynamics and façade design parameters jointly determine fire exposure conditions.

Fires in ventilated cavities can be up to ten times faster than free-burning fires, with fourteenfold higher heat flux and significantly elevated temperatures [6]. These phenomena are particularly dangerous in multi-storey timber construction, where they enable rapid fire spread to upper levels [6]. Additionally, cavity fires produce higher smoke velocity and toxicity, posing severe risks to occupants, as smoke inhalation is a major contributor to fire fatalities [9].

The significance of cavity geometry is further supported by large-scale tests reported by McNamee et al. [23], which demonstrated that adding a 20 mm ventilated cavity behind plywood cladding nearly doubled the total heat release—from 1870 MJ to 2140 MJ—relative to an inert façade baseline of 1550 MJ. These results confirm that even relatively narrow cavities can dramatically intensify fire dynamics by providing a vertical channel that enhances oxygen supply and sustains combustion.

This behaviour can be explained by fundamental fluid mechanics and heat transfer principles: narrow cavities accelerate upward gas flow, increase turbulence, and raise convective heat flux to the cladding, driving faster ignition and flame spread. Combustible fire barriers initially act as thermal shields, delaying ignition by obstructing airflow, but once pyrolysis is initiated, they become additional fuel sources, eventually losing effectiveness.

Collectively, these findings emphasize that façade fire performance is governed by the complex interaction of geometry, airflow, and material behaviour. Careful design of cavity dimensions, barrier configuration, and opening geometry is therefore essential to limit fire spread and improve the fire resilience of ventilated timber façades.

1.4. Methods for Limiting Fire Spread in Ventilated Façade Systems

Controlling fire spread in ventilated façade systems is a critical aspect of their design. The most effective method to prevent fire spread is the installation of fire barriers. Several types of fire barriers are used, with differences in construction and material composition [6,14].

The first type consists of barriers made from non-combustible materials, such as mineral wool, which are supplemented with intumescent components. These components expand when exposed to high temperatures, thereby sealing the ventilated cavity to prevent flame propagation. However, these barriers may have a delayed with response times ranging from several tens of seconds to several minutes. For systems with wooden cladding, metal barriers penetrate the cladding and are attached perpendicularly to the structural wall have proven to be the most effective solution [11]. These barriers act as flame deflectors, redirecting flames away from the façade system, thereby reducing thermal stress. However, they disrupt the façade’s structural integrity and reduce airflow within the cavity. Other types include barriers without reactive materials, such as perforated metal sheets or labyrinth barriers, which slow down fire spread but cannot completely stop the flames [12].

The influence of barriers in ventilated façade cavities can be analyzed through their effect on airflow and heat transfer. In unobstructed cavities, buoyancy-driven chimney flows create strong upward currents that accelerate flame spread by increasing turbulence, oxygen entrainment, and convective heating of the cladding. Barriers interrupt this flow, reducing the effective chimney height, creating localized recirculation zones, and thereby lowering heat transfer to the upper cavity. They also act as thermal shields, absorbing and deflecting radiant and convective flux. While combustible barriers provide only a temporary delay before contributing fuel once pyrolysis begins, non-combustible barriers can maintain cavity segmentation and continue to limit vertical flame propagation in timber buildings [14].

In addition to installing barriers, attention must be paid to the quality and thickness of materials used for external cladding. Increasing the thickness of wooden cladding or applying special surface treatments, such as the “shou sugi ban” technique (charred wood finish), can reduce fire spread rate [6]. Another option is using fire-retardant coatings based on silicate or intumescent paints that create a protective layer when exposed to high temperatures. Ensuring high-quality construction details, such as proper sealing around window openings and joints between materials, is also crucial for minimizing the risk of flames penetrating the ventilated cavity [9].

1.5. Fire Testing of Façade Systems

Large-scale façade fire testing plays a critical role in evaluating the performance of ventilated timber façades under realistic fire exposure conditions. While small-scale tests provide valuable material-scale data, they cannot capture key phenomena such as flame spread within cavities, the influence of openings, or the interaction between construction details and fire dynamics [18,24]. European approaches, such as DIN [8] and the BS 8414 series [25], have therefore emphasized large-scale methods that reproduce compartment fire scenarios and enable assessment of cladding systems as installed.

In this study, large-scale tests were used specifically to examine how cavity geometry, workmanship quality, and the presence of fire barriers influence fire spread in ventilated timber façades. This approach ensures that the findings are representative of real façade configurations and can inform practical design and regulatory decisions.

2. Methodology

The fire testing methodology employed in this study consists of full-scale fire tests designed to evaluate fire spread in ventilated façade systems with wooden cladding. Large-scale fire experiments were conducted under controlled conditions to replicate real-world fire scenarios, focusing on façade design details, fire barrier placement, and ventilation strategies affecting flame propagation. These experiments assessed the influence of window lintel construction, cavity geometry, and the effectiveness of horizontal and vertical fire barriers in mitigating fire spread. Temperature profiles, flame heights, and heat flux measurements were recorded at various points along the façade to analyze fire dynamics in the ventilated cavity.

A key aspect of the study was the investigation of unintended air leaks in the cavity, which could significantly influence fire behaviour by altering airflow patterns and oxygen supply. The experimental setup included variations in cavity width and ventilation opening configurations to determine how these factors impact the rate and intensity of fire spread.

Phase One served as a preliminary setup to verify the functionality of the combustion chamber and cladding attachment. No cavity tests were conducted during this phase, and its main purpose was to refine the apparatus used for subsequent tests.

The test stand consisted of a combustion chamber made of refractory concrete panels (Betonika s.r.o., Brno, Czech Republic) equipped with a steel frame support (Hilti AG, Schaan, Liechtenstein).

Cladding panels of spruce and larch timber were prepared and installed in cooperation with VSB—Technical University of Ostrava, Faculty of Safety Engineering (Ostrava, Czech Republic).

Surface and gas-phase temperatures were measured using Type K thermocouples (TT-K-36-SLE, Omega Engineering Inc., Norwalk, CT, USA) with a measurement uncertainty of ±1.5 °C.

Temperature signals were recorded through a National Instruments NI-9213 data acquisition module (National Instruments Corp., Austin, TX, USA) connected to a CompactDAQ system (Model cDAQ-9174).

Insulation materials used in the ventilated cavity included stone wool boards (ROCKWOOL A/S, Hedehusene, Denmark) with a nominal density of 40 kg·m⁻³ and thermal conductivity of 0.035 W·m⁻¹·K⁻¹.

Cavity barriers were fabricated from 15 mm thick gypsum-fiber boards (Knauf Gips KG, Iphofen, Germany).

All timber cladding boards were kiln-dried to 12 ± 2% moisture content prior to installation.

3. Large Scale Fire Tests

3.1. Description of Tests and Experimental Setup

This section outlines the large-scale fire tests focused on flame spread in ventilated façades without thermal insulation. We conducted the tests using larch and spruce cladding panels.

The experiments were carried out in an unused industrial facility where spatial and budgetary constraints necessitated simplifications in the installation of the test apparatus. Existing spaces and structural elements of the building were used. As a result, the dimensions of the setup did not conform to those used in certified fire tests. The experimental apparatus consisted of a combustion chamber and a supporting structure for mounting the test samples (wooden cladding panels). The installation proceeded in two phases:

Phase One—The ceiling of the combustion chamber was modified with low-biopersistent alkaline earth silicate (AES) fibre insulation, and steel support profiles were installed to secure the cladding. The opening of the combustion chamber had a rectangular shape with a longer vertical side, allowing for the study of exterior cladding behaviour with various surface treatments. In this phase, the cavity of the ventilated façade was not tested.

Phase Two—The experimental setup was further modified. The walls of the combustion chamber were lined with refractory bricks, and a new L-shaped supporting structure made of ceramic blocks was constructed. We modified the opening of the combustion chamber to a rectangle with a longer horizontal side. The newly constructed wall-mounted support structure enabled testing of ventilated façade systems. Images depicting the original state, modifications, and final arrangement of the test setup are shown in the attached Figure 1 and Figure 2.

Three full-scale façade specimens were constructed to represent typical ventilated timber cladding assemblies used in Central European practice. Each specimen consisted of a 5 m high, 3 m wide wall with a ventilated cavity behind 19 mm untreated spruce cladding mounted on vertical battens. The assemblies were installed on a steel frame and incorporated a central window opening (1.2 × 1.2 m) to reproduce a compartment fire plume impinging on the façade. Cavity widths of 20 mm and 40 mm were tested, with and without horizontal cavity barriers installed at mid-height. Lintel and jamb details were intentionally varied to represent both well-sealed and poorly sealed workmanship. The fire source consisted of wooden pallets with a total mass of approximately 335 kg (12% moisture content) arranged in the combustion chamber to generate a sustained fire exposure. Temperature measurements were taken using thermocouples placed along the cavity height and façade surface to capture ignition times, flame spread rates, and heat flux trends.

A comprehensive summary of all tests conducted is presented in

Table 1. Description of Tests.

	Timber Species and Cladding Thickness	Opening Geometry	Cavity, Width	Barriers	Lintel/Jamb	Fire Load (kg, Moisture%)
Experiment 1	Spruce, 19 mm	Vertical window (1.2 × 1.2 m)	20 mm	None	Poorly sealed	335 kg pallets, ~12% moisture content
Experiment 2	Spruce, 19 mm	Vertical window (1.2 × 1.2 m)	20 mm	Horizontal barrier at mid-height	Precisely sealed	335 kg pallets, ~12% moisture content
Experiment 3	Spruce, 19 mm	Vertical window (1.2 × 1.2 m)	40 mm	Horizontal barrier at mid-height	Precisely sealed	335 kg pallets, ~12% moisture content

, detailing the experimental configurations, material specifications, cavity dimensions, opening factors, and workmanship variations considered in this study.

In all tests, the fire load consisted of wooden pallets with a total mass of approximately 335 kg and an average moisture content of 12%. While the stacking arrangement in the combustion chamber was adjusted slightly between tests to fit within the available space and ensure reliable ignition, the overall fuel mass and distribution per unit floor area were kept constant. Such minor variations are inherent to large-scale fire testing and may influence the exact timing of pallet collapse or localized flame impingement; however, the heat release potential was comparable across all tests, allowing for meaningful comparison of cavity ignition times and flame spread behaviour.

3.2. Evaluation of Ventilated Façade Tests with Wooden Cladding

Fire tests conducted confirmed several theoretical assumptions, with the quality of window jambs and lintels influencing the rate of flame transfer into the ventilated cavity. Wooden fire barriers limit flame spread within the ventilated cavity. The shape of the combustion chamber opening plays a significant role in determining flame position—when the opening is rectangular with a longer vertical side, the flames are deflected away from the test sample, reducing thermal stress. However, if the opening is rectangular with a longer horizontal side or square, the flames rise directly upwards, substantially increasing the thermal exposure of the sample. This effect is illustrated in Figure 3. The left image shows flames emerging from a combustion chamber with a vertically oriented rectangular opening. In this configuration, the flames are narrow, tall, and closely attached to the wall, producing a highly concentrated plume. This vertical orientation promotes a strong buoyancy-driven jet that entrains air efficiently and results in a longer external flame height, increasing the thermal exposure to the façade above the opening.

By contrast, the right image illustrates flames from a horizontally oriented opening of similar area. Here, the flame plume spreads laterally across the opening, resulting in a broader but shorter flame that detaches earlier from the façade surface. This geometry encourages lateral dispersion of hot gases, reducing vertical momentum and leading to lower external flame height and a smaller heat flux to the cladding directly above the opening.

Together, these images highlight the critical influence of opening geometry on flame impingement and façade heat exposure. They demonstrate why vertical openings pose a greater risk of vertical fire spread along façades, as they intensify upward convective flow and produce higher flame attachment zones compared to horizontal openings of equal area.

This phenomenon was first systematically investigated by Yokoi [26], whose foundational experiments explored the influence of opening geometry on flame behaviour. Through scaled model tests, Yokoi [26] demonstrated that vertical openings—those with a longer vertical edge—promote the formation of taller and more concentrated flame plumes, as the buoyant hot gases are more effectively channelled upward. In contrast, horizontal openings tend to disperse the flow laterally, producing shorter and less intense vertical flames. This research laid the groundwork for understanding how façade opening configuration influences the development of hot upward currents and the likelihood of vertical flame attachment, which are critical factors in fire spread to upper stories. Yokoi’s findings [26] remain a cornerstone in the field of façade fire dynamics and continue to be referenced in both contemporary experimental research and fire safety engineering textbooks for their enduring relevance to real-world fire scenarios.

A total of three large-scale fire tests were conducted using the final configuration of the test apparatus to evaluate the fire behaviour of ventilated façade systems. The experiments specifically examined the influence of construction workmanship at window jambs and lintels on the onset of cavity ignition and subsequent flame spread. The results demonstrated that precise detailing and sealing in these locations can delay fire penetration into the ventilated cavity; however, once the cladding was consumed by sustained fire exposure, cavity ignition and upward flame spread occurred in all cases, regardless of workmanship quality.

The effect of horizontal barriers placed at various heights above the combustion chamber lintel was also examined. The barriers consisted of wooden battens attached to the supporting structure of the test apparatus using metal brackets and screws, with gaps filled with mineral insulation. Although these barriers were made of combustible wood, the tests yielded interesting results, demonstrating how such barriers might behave during a fire. The designed wooden barrier represented the behaviour of a combustible horizontal barrier that, in the event of a fire, would ideally seal the ventilated cavity of a façade system clad with wooden panels in a horizontal direction. Simply put, if such a “perfect” combustible barrier existed, its impact on fire spread was evaluated.

The test samples consisted of horizontally installed spruce panels of B/C quality (untreated), with a thickness of 19 mm and a height of 121 mm, interconnected via a tongue-and-groove system. The supporting structure was secured using wooden battens and metal brackets. The façade system did not include thermal insulation, focusing solely on the behaviour of the wooden cladding. The wall had a total length of 3.5 m and a height of 5 m, with a section above the chamber lintel measuring 3.5 m in height.

At the lintel and jamb areas (around the combustion chamber opening), wooden battens clad with panels were used. All gaps between façade elements were packed with stone wool insulation (λ = 0.035 W/m·K, density ≈ 35 kg/m³) to maintain cavity compartmentation and minimize airflow bypass. In the third experiment, the lintel was constructed solely from wooden cladding (without battens or sealing), with a 10 mm gap between the cladding and the wall to simulate poor workmanship. This setup was intended to facilitate faster flame penetration into the cavity, thereby better demonstrating the impact of horizontal barriers on fire spread.

The experimental measurements varied only in the presence of horizontal barriers and the quality of the lintel (see

Table 2. Differences between experiments.

	Horizontal Barriers	Lintel with Wooden Battens Including Sealing
Experiment 1	YES	YES
Experiment 2	NO	YES
Experiment 3	½ YES, ½ NO	NO

). During the third experiment, horizontal barriers were installed on one half of the sample, while the other half remained without them. The construction of the wooden cladding and the amount of fuel (335 kg of wooden pallets in the combustion chamber) remained consistent across all tests. Each experiment commenced by igniting the wooden pallets in the combustion chamber. Temperatures were recorded using thermocouples, and video footage was captured throughout the tests. The schematic layout of the thermocouple placement for experiment 3 is shown in Figure 4.

In the experimental part of the study, commercially available wooden pallets commonly used in the Czech Republic were used as fuel. These pallets were made of softwood with a bulk density of around 550 kg/m³. The moisture content, depending on storage and climatic conditions, was approximately 15%.

The design and dimensions of the pallets conformed to [27], measuring 1200 × 800 mm and a weight of approximately 22–25 kg per unit. During testing, the pallets were arranged in two stacked layers with an air gap between them, allowing for improved airflow and thereby producing more intense pyrolysis. This setup promotes flame development and simulates realistic outdoor storage or transport conditions.

K-type bare-wire thermocouples were used to measure surface and gas-phase temperatures. The thermocouples had a wire diameter of 0.5 mm, which offers a suitable compromise between mechanical durability and temporal response, with a typical time constant on the order of 1–2 s in fire conditions. This size ensures that the thermocouples can respond adequately to both radiative and convective heat transfer, although it is acknowledged that bare-wire sensors may show slight differences in response depending on the dominant heat transfer mode.

For surface temperature measurements, thermocouple junctions were attached directly to the substrate using metal quick-release clamps or tape. All attachments were made carefully to ensure that the thermocouple tip remained flush with the surface or slightly recessed, in accordance with standard procedures for accurate surface temperature measurement.

The tests were concluded once the section above the combustion chamber was completely burned through. Differences between the individual experiments are illustrated in Figure 5, Figure 6, Figure 7 and Figure 8 and summarized in the subsequent table.

The findings in this section are based on an analysis of video footage and temperatures recorded by thermocouples.

The moment any section of the wooden cladding loses integrity, flames penetrate into the ventilated cavity. The most thermally stressed areas are the lintels and jambs, emphasizing the importance of precise construction and high-quality materials in these sections. Wood typically chars at a rate of approximately 0.7 mm/min [28], which leads to a slower degradation and loss of integrity compared to other materials like plastic cladding. Nevertheless, the wooden cladding eventually burns through, and once flames penetrate the ventilated cavity, thermal stress and fire spread occur on both sides of the cladding.

The installation of horizontal barriers within the cavity proved insufficient to prevent flame spread along the outer surface of the cladding. Due to the chimney effect, the height of flames inside the cavity was significantly greater. At peak fire intensity, flames reached heights exceeding 6 m (measured from the lintel of the combustion chamber), surpassing the height of the testing apparatus. Therefore, it should not be expected that the first barrier above the façade opening would function effectively. Containment of fire spread could only be achieved starting from the second or third level of barriers.

The results of individual experiments are not easily comparable, as the wooden pallets were arranged differently in the combustion chamber for each test. This resulted in varying temperature profiles both inside the chamber and on the external surface of the cladding. In experiment 3, at approximately 7 min 45 s, several pallets fell out of the combustion chamber. This affected the combustion process as it restricted the air supply, leading to soot deposits on the refractory walls. Despite these circumstances, it cannot be concluded that the experiment was invalid. Maximum temperatures were indeed lower, but the duration of heat exposure was extended.

A comparison between experiments 1 (with horizontal barriers) and 2 (without barriers) does not conclusively demonstrate the positive effect of horizontal barriers due to the precise workmanship of the lintels and jambs and the limited sample size. However, it is clear that horizontal fire barriers positioned above the level of flames emerging from the combustion chamber positively impact limiting flame spread. This is particularly true for barriers (flame deflectors) that penetrate through the cladding, as they also prevent spread along the exterior side of the façade system.

In none of the experiments was horizontal fire spread observed. Vertical battens used to attach the cladding to the supporting structure of the test setup effectively prevented horizontal spread. These vertical battens acted similarly to horizontal barriers, although they were not sealed with mineral insulation. Despite the lack of sealing, no significant gaps were detected between the vertical battens and the test sample or between the battens and the supporting structure. The question remains as to what gap size would allow horizontal fire spread.

The following section focuses in detail on experiment 3. The test sample was divided into a section with horizontal barriers and one without (see Figure 7). In both sections, the lintel area was constructed solely of cladding, without horizontal battens or sealing. Additionally, a gap of approximately 10 mm was left between the cladding and the supporting wall (see Figure 8).

Scheme 1 shows that just above the combustion chamber, the temperature inside the cavity increased slowly compared to the temperature on the surface of the cladding at the same location. We observed this phenomenon in both the section without barriers and the section with barriers, even with the gap at the lintel. In the section without barriers, the temperature began to rise sharply after reaching approximately 250 °C, at approximately 15 min 45 s. In the section with barriers, the temperature increased rapidly only after reaching about 370 °C, at approximately 21 min 55 s. Video footage indicates that at approximately 15 min, the section without barriers was more thermally exposed to flames, eventually leading to sections of cladding detaching. Thus, it cannot be definitively stated that flames penetrated the cavity earlier due to the chimney effect in the section without barriers. It is crucial to note that even in the section without barriers, fire penetrated the cavity at approximately 16 min, despite the gaps in the lintel area. At approximately 19 min, flames began to emerge from the cavity at the end of the test sample (approximately 3.5 m above the combustion chamber lintel).

Scheme 2 illustrates that in the section with barriers, the “perfect” installation of barriers effectively halted the fire spread within the cavity. The section with barriers eventually failed after approximately 30 min, due to the burning through of the cladding and subsequent fire transfer from the outer surface of the façade system.

Although large-scale fire tests assess samples up to approximately 9 m in height, this may not always be sufficient. It should be acknowledged that even the most meticulously conducted large-scale fire test cannot perfectly simulate the behaviour of a façade system under real fire conditions. It would be beneficial to further investigate the performance of fire barriers positioned above the flame level, particularly the second and third barriers above the façade opening. The influence of the chimney effect with increasing height of the façade system, horizontal fire spread, and the impact of fire on lower façade openings could significantly affect the results.

Figure 9 illustrates the behaviour of the ventilated façade minute-by-minute during Experiment 3. Figure 9 indicates the moment at 7 min 45 s, when several pallets fell out of the combustion chamber.

3.3. Consideration of Experimental Uncertainty

As with all large-scale fire experiments, the results of this study are subject to several sources of uncertainty. Variability in fuel arrangement and combustion behaviour influenced the repeatability of the tests, as illustrated in Experiment 3, where the partial collapse of wooden pallets altered the burning rate and heat release profile. Instrumentation also introduces uncertainty, with bare-wire thermocouples exhibiting typical response times of 1–2 s and calibration errors on the order of ±15 °C. Placement accuracy of the sensors further contributes to variability in recorded temperatures. Environmental factors, such as minor airflows within the test hall, may have affected flame behaviour and smoke movement. In addition, the limited number of tests (three replicates) limits the statistical robustness of the findings.

To reflect these uncertainties, the timing values reported in this paper should be considered approximate, with an uncertainty of ±1 min due to observational and measurement limitations. While additional replicates would be required to achieve stronger statistical confidence, the observed trends remain consistent across all tests: workmanship at lintels and jambs critically influences flame penetration, combustible fire barriers can provide a temporary delay, and the chimney effect within cavities is the dominant driver of vertical fire spread.

4. Results and Discussion

4.1. Influence of Design Parameters on Fire Spread in Ventilated Timber Façades

The experimental results emphasize three interconnected aspects of fire behaviour in ventilated wooden façade systems: the quality and precision of construction details, the effect of cavity geometry and the chimney phenomenon, and the role of fire barriers. Each theme is discussed in relation to prior research.

A comparative analysis of the three experimental groups revealed pronounced differences in both ignition timing and fire dynamics. In Experiment 1, high-quality workmanship and the presence of horizontal barriers delayed cavity ignition, with the barriers locally disturbing the upward plume and slowing vertical flame spread. In Experiment 2, cavity ignition occurred after approximately 16 min despite proper workmanship, confirming that good detailing alone cannot prevent cavity involvement once flames reach the façade. Experiment 3 demonstrated the cumulative impact of deficiencies: poor lintel workmanship enabled flame penetration up to 4 min earlier, while the barrier-protected section still delayed vertical flame spread by 5–7 min compared to the unprotected side. Collectively, these results show that workmanship primarily governs the onset of cavity ignition, fire barriers regulate the rate of vertical flame spread, and the chimney effect determines flame height and intensity once established. This integrated interpretation underscores the necessity of considering these parameters collectively when designing fire-safe ventilated façade systems.

The quality and precision of construction details, particularly around openings such as windows and doors, are of paramount importance. Poorly executed details can accelerate the penetration of flames into the ventilated cavity, which in turn increases the risk of fire spread. In our tests, inadequately sealed lintels allowed flames to enter the cavity up to 4 min earlier than in precisely executed assemblies. This observation is consistent with Engel and Werther [12], who emphasize that openings are the most vulnerable points of ventilated façades, and that sealing details govern the reliability of compartmentation. Our findings reinforce their conclusion that workmanship is not a secondary factor but a primary determinant of façade fire safety.

Combustible horizontal barriers delayed cavity ignition by approximately 5–7 min under conditions of precise installation. However, once exposed to sustained flames, these barriers eventually failed, contributing to the fuel load. The experimental results confirmed that wooden fire barriers can effectively slow down flame spread within the cavity, but their effectiveness is influenced by proper placement and the quality of materials used. In the case of wooden façade systems, it is essential to also prevent fire spread along the exterior surface of the cladding. Therefore, barriers that function solely within the cavity are insufficient. A more effective solution involves barriers (flame deflectors) that penetrate the full depth of the cladding. Flames emerging from façade openings typically reach heights of several meters. Effective fire containment in ventilated façade systems requires that cavity barriers be installed at or above the level of anticipated flame impingement from window or door openings. Barriers located below this level may be bypassed by flames and thus provide limited benefit. Our results confirm that properly positioned barriers remain essential to delaying vertical fire spread and cannot be omitted from design practice.

The chimney effect is the primary driver of fire spread in ventilated cavities. The results confirmed that cavity geometry governs the intensity of vertical flame spread. In experiments without barriers, ignition within the cavity occurred within 16 min, with flame heights exceeding 6 m. Our findings indicate that flame spread rates and heat flux increase significantly in narrow cavities, an observation also supported by Mendez et al. [13]. Their experimental work demonstrated that cavity widths below 50 mm create a strong upward draft, leading to sustained fire growth. Similarly, our research suggests that minimizing uncontrolled air movement within cavities through strategically placed fire barriers can mitigate risks associated with rapid vertical fire spread.

In addition to vertical flame propagation, the potential for lateral flame spread across the façade was examined. No horizontal flame spread was observed in any of the experimental configurations, regardless of workmanship quality or barrier installation. This finding is consistent with previous studies [14,16], which have demonstrated that sustained lateral flame travel requires continuous combustible pathways, such as uninterrupted horizontal battens or connected soffit cavities—features that were absent in the present assemblies.

Although increased airflow through panel joints locally intensified surface charring and flame luminosity, it did not result in lateral flame migration across the façade. These observations reinforce that vertical cavity flow and the buoyancy-driven chimney effect remain the dominant mechanisms driving façade fire propagation. Consequently, mitigation strategies should prioritize cavity segmentation and vertical flame arrest measures rather than focusing on horizontal flame spread control for ventilated façades of this type.

Another point of agreement is the effect of ventilation openings on fire dynamics. Mendez et al. [13] highlight that open base configurations intensify fire spread due to unrestricted oxygen supply, mirroring our conclusions regarding the impact of air leaks. Both studies advocate for controlling air intake through design modifications to reduce flame extension and delay fire penetration into the building’s structural elements.

The comparison between studies underscores the need for comprehensive fire safety strategies in ventilated façade systems. Both studies suggest that cavity width and ventilation control must be carefully designed to minimize fire risks. However, our research highlights additional regulatory considerations for wooden façade systems, including the necessity of extending fire barriers beyond the cavity to limit external flame propagation.

An important aspect of façade design that complements fire safety is moisture management. Fully closed cavities, as used in the present experimental setup, are not recommended for long-term durability because they can trap moisture and promote decay of wooden components. In practice, design solutions such as compartmentalized but still ventilated cavities or the use of moisture-permeable fire barriers should be considered. These approaches allow limited airflow to maintain drying potential while providing adequate fire compartmentation. Several studies on building physics and timber durability support this trade-off between fire protection and moisture control, emphasizing the need for balanced design to ensure both fire safety and long-term performance of ventilated timber façades [29].

4.2. Comparative Analysis of Experiments

To better illustrate the differences in fire spread behaviour, the three experiments were compared at equivalent time intervals. This synchronized analysis highlights how workmanship, cavity geometry, and the presence of fire barriers influenced the combustion process and vertical flame propagation.

• Early stage (0–10 min).

In all experiments, flames emerging from the combustion chamber impinged directly on the façade surface within the first minutes. At this stage, no penetration into the ventilated cavity was observed. Surface charring developed uniformly across all samples, although minor differences were visible in the lintel regions. In Experiment 3, where the lintel was poorly sealed, visible leakage of flames and smoke occurred several minutes earlier than in Experiments 1 and 2, indicating a greater vulnerability of deficient construction details.

• Intermediate stage (10–20 min).

Significant divergence was observed during this period. In Experiment 2 (no barriers, well-sealed workmanship), cavity ignition occurred at approximately 16 min, even though detailing was precise. In Experiment 1 (with barriers, well-sealed workmanship), ignition was delayed to about 18–19 min, showing that barriers can postpone vertical fire spread by several minutes. In Experiment 3 (poor workmanship, half with barriers), flame penetration occurred as early as 15–16 min on the unprotected side of the façade. On the side with barriers, ignition was delayed until ~22 min, confirming the combined effect of workmanship and cavity segmentation on delaying fire development.

• Fully developed stage (20–30 min).

Once the cavity was ignited, vertical flame spread dominated in all experiments due to the chimney effect. In Experiment 1, barriers locally disturbed the plume and reduced flame attachment, delaying upward progression by 5–7 min. In Experiment 2, flames rose rapidly through the cavity, exceeding 6 m in height within 25 min. In Experiment 3, the unprotected section exhibited the fastest progression, with flames emerging from the cavity end at 19 min, while the barrier-protected section maintained integrity until nearly 30 min.

• Comparison across experiments.

The synchronized comparison confirms that (i) workmanship governs the timing of initial cavity ignition, (ii) barriers influence the rate of vertical flame spread once ignition occurs, and (iii) cavity geometry and the chimney effect ultimately control flame height and intensity. Poor workmanship accelerated cavity ignition by up to 4 min, while properly installed barriers delayed vertical flame spread by approximately 5–7 min. Nonetheless, in all configurations, cavity involvement eventually occurred once the external cladding was consumed.

5. Conclusions

5.1. Comprehensive Considerations and Recommendations

The spread of fire along façades is significantly influenced by the geometry and orientation of openings, the configuration of ventilation cavities, and the presence of fire-stopping elements. Experimental results confirmed that the shape and orientation of window openings affect both flame height and façade heat exposure. For example, vertically oriented openings produced flame extensions exceeding 2.5 m externally, while horizontal openings limited flame height to approximately 1.5 m, demonstrating the influence of opening geometry on vertical fire propagation.

The presence of a ventilated cavity was found to significantly increase the risk of vertical fire spread due to the chimney effect. In configurations without horizontal fire barriers, cavity ignition and flame emergence on upper levels occurred within 16 min from the start of ignition. In contrast, installing horizontal and vertical fire barriers delayed flame propagation through the cavity by approximately 5–7 min, providing a critical time window for occupant evacuation and fire service intervention.

The comparative tests further demonstrate that horizontal fire barriers, when installed above the level of flames emerging from the window opening, significantly delay vertical fire spread within ventilated cavities. This effect was most pronounced for full-depth barriers (flame deflectors), which also limited flame spread along the external cladding surface. Nonetheless, cavity involvement occurred in all tests once the cladding was consumed, confirming that barriers must be regarded as a mitigation measure rather than a complete safeguard. The results underline that precise workmanship at lintels and jambs can delay cavity ignition by several minutes but cannot, by itself, prevent vertical fire propagation once flames enter the cavity.

The overall quality of workmanship and detailing also had a measurable impact on fire dynamics. Inadequate sealing and unprotected penetrations allowed fire to breach compartments up to 4 min earlier than in properly executed assemblies, demonstrating the importance of execution quality in maintaining fire safety integrity.

These findings underline the need for testing methods that reflect the specific characteristics of ventilated façade systems, particularly the influence of cavities, open joints, and realistic installation details. Current standardized test methods often overlook these parameters, which can significantly influence fire spread mechanisms in real applications.

Finally, the use of combustible materials in façade systems on taller buildings remains problematic without additional protective measures. The results suggest that wooden cladding is considered impractical above a fire height of 12 m, due to elevated heat release rates and the complex requirements for fire-stopping design in ventilated assemblies.

Fire safety in multi-storey buildings with wooden façade systems represents a complex challenge that requires careful design, selection of high-quality materials, and the implementation of effective safety measures. The results of research and practical tests indicate that only properly designed and executed ventilated façade systems can achieve the required level of fire safety. However, this requires a thorough assessment of all risk factors, including construction details and the influence of the chimney effect.

Key measures include increasing the thickness of wooden cladding, installing fire barriers, using suitable surface treatments, and ensuring precise execution of construction details, particularly around façade openings. These measures can significantly slow fire spread in ventilated façade systems with wooden cladding. Additionally, it is essential to ensure compliance with evolving legislation and implement the latest standards and regulations, thereby ensuring a uniform and high level of fire safety across Europe. It must be stressed that the numerical values reported in this study are intrinsically linked to the specific experimental configuration employed. The façade test wall had a height of 5 m and was clad with untreated spruce boards of 19 mm thickness, exposed to a fuel load of approximately 335 kg of wooden pallets with a bulk density of 550 kg/m³, and tested with cavity geometries as detailed in the Section 2. These values should therefore not be construed as universal thresholds, but rather as representative of the boundary conditions under which the experiments were conducted. Variations in material properties, cavity dimensions, fire load intensity, or overall façade height can be expected to influence the onset and progression of fire spread. Accordingly, the results should be interpreted as indicative trends providing valuable insights into the mechanisms governing fire behaviour in ventilated timber façade systems, rather than as absolute design limits.

5.2. Future Research and Ongoing Improvements

It remains essential to pursue further research and testing on various façade systems to improve their fire resistance and mitigate risks associated with wooden construction materials. Recent studies, such as the work by McNamee [30] on the development of façade fire testing in Sweden, emphasize that continuous testing and adaptation of experimental methods are necessary to address the growing use of combustible materials in modern façade systems. Their findings highlight the importance of harmonizing international testing methodologies and incorporating real-scale façade behaviour into performance assessments. In line with these observations, the primary focus in the design and execution of buildings should be to ensure the highest possible level of fire safety, regardless of building height or material type.

This paper summarizes the current findings on fire spread in wooden façade systems and suggests key directions for future research and regulatory activities. Proper design and development of fire protection measures in ventilated façade systems can significantly contribute to enhancing safety in modern construction.