Sie können Operatoren mit Ihrer Suchanfrage kombinieren, um diese noch präziser einzugrenzen. Klicken Sie auf den Suchoperator, um eine Erklärung seiner Funktionsweise anzuzeigen.
Findet Dokumente, in denen beide Begriffe in beliebiger Reihenfolge innerhalb von maximal n Worten zueinander stehen. Empfehlung: Wählen Sie zwischen 15 und 30 als maximale Wortanzahl (z.B. NEAR(hybrid, antrieb, 20)).
Findet Dokumente, in denen der Begriff in Wortvarianten vorkommt, wobei diese VOR, HINTER oder VOR und HINTER dem Suchbegriff anschließen können (z.B., leichtbau*, *leichtbau, *leichtbau*).
Der Artikel geht der kritischen Frage nach, wie Feuer die Tragfähigkeit von Schraubverbindungen beeinflusst, die in Holzdecken verwendet werden, insbesondere für mechanische und elektrische Dienstleistungen (M & E). Durch eine Reihe kontrollierter Brandtests untersucht die Studie die Leistung verschiedener Schraubverbindungen unter Umgebungs-, Brand- und Nachbrandbedingungen. Schlüsselergebnisse zeigen, dass Schrauben mit größerem Durchmesser, die zunächst größere Lasten tragen, aufgrund von Wärmeleitung und Verkohlung eine deutliche Verringerung der Tragfähigkeit erfahren. Die Studie unterstreicht auch den raschen und vollständigen Verlust der Festigkeit von Schraubverbindungen unter Brandeinwirkung und betont die Notwendigkeit verbesserter Befestigungsmethoden und -standards. Darüber hinaus diskutiert der Artikel die möglichen Folgen der Behebung von Fehlern, einschließlich Risiken für Evakuierung und Brandbekämpfung, sowie die zunehmende Ausbreitung von Bränden und Folgeschäden. Die Forschungsarbeiten schließen mit Empfehlungen für zukünftige Arbeiten ab, einschließlich der Entwicklung neuer Befestigungsarten und der Festlegung geeigneter Prüfstandards zur Gewährleistung der Gebäudesicherheit im Holzbau.
KI-Generiert
Diese Zusammenfassung des Fachinhalts wurde mit Hilfe von KI generiert.
Abstract
The load capacity of screw fixings supporting building services, such as sprinklers in Cross Laminated Timber (CLT) ceilings was tested before, during and after exposure to fire, to investigate the risk of detachment. The pull-out strength of screw fixing in ambient conditions, during and post-fire were investigated in relation to pull-out strength versus embedment depth. The relationships between pull-out strength have been reported as functions of screw fixing embedment depth, screw dimensions, and char formation. In the experimental study, samples of standard industrial CLT were tested with two distinct types of adhesives with typical construction industry screw fixings used for the suspension of Mechanical and Electrical (M&E) services. A purpose built fire-test rig was designed to expose screw fixings embedded into CLT to a fire in a ceiling mounted configuration. A series of eight fires were conducted, and the pull-out strength of each screw fixing was assessed during or after the fire. The reduction of load capacity can be conceptualised into two factors: The charring across the whole timber surface which was deeper close to the fixings leaving a fragile char which could be scraped off; and the weakening of the timber along the length of the screw thread, resulting from the higher thermal conductivity of the screw fixing. Both these effects increased as a function of the shank width of the screw. The outcome of this study is to inform guidance on the ability of screw fixings to support M&E services beneath timber ceilings in the event of fire.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
1 Introduction
Engineered timber is gaining popularity in construction due to its availability in various forms, sustainability credentials, strength for primary structure, and visual appearance of structural components and exposed internal surfaces [1‐5]. Timber is replacing materials such as steel and concrete which are non-combustible and have well understood structural fire performance. The increased use of timber in the UK has been driven by government in their vision for addressing climate change [6]. It is also seen as conferring enhanced sustainability to a building and the businesses that might operate from it, and so internal surfaces are likely to be exposed to create the aesthetic visual qualities demanded by modern construction. Where combustible surfaces are exposed, sprinklers have been shown to be effective provided the sprinkler system is appropriate for the risk and has been serviced and maintained correctly. Activation in the early stages of fire will provide control and cooling from the sprinklers to adjacent surfaces limiting fire spread [7‐10].
It is acknowledged that timber has two mechanisms for burning: flaming combustion of the volatiles released through pyrolysis; and surface oxidation of char, often resulting in glowing combustion at the surface [7]. In some fires, the build-up of char will be sufficient to suppress flaming combustion. The temporary or permanent extinction of flaming without external intervention is sometimes described as auto-extinguishment [12‐21]. Details related to heat flux conditions for auto-extinguishment differ amongst the research and variances in air flow velocities are seen to be a contributing factor.
Anzeige
Little guidance for fixing fire sprinkler and M&E services directly to the underside of timber ceilings is provided within standards. European sprinkler standards only cover fixing into concrete [22], industry guidance provides best practice for concrete, but does not cover timber [23], and British standards only consider concrete [24]. Sprinkler and M&E services are suspended using mechanical assemblies fixed to the underside of the structure. For concrete this is achieved using anchors (also known as expansion fixings) that connect to vertically threaded drop rods. Clamps are used for steel beams that also connect to drop rods where the services will be fixed using approved components, such as mild steel loop bands for pipework.
Recently, a 100 mm water pipe collapsed from a fixing failure which resulted in a fatality [25] and historically, the deaths of firefighters have been attributed to entanglement in cables and structures that fell from the ceiling during fire (Shirley Towers) [26], and requirements have been tightened to ensure this is not repeated (IEE Wiring Regulations) [27]. In addition to the obvious life-safety benefit of fire suppression, sprinklers are likely to enhance the ability of ceiling mounted systems to remain in place through the cooling effects of the sprinkler water.
Suppression systems are fitted into many different types of construction to support life-safety and for property and business protection [22, 28‐33]. However, installation guidance has been developed around building types which do not include fixing directly to combustible structures.
Structural load capacity, bracket assemblies, and individual fixings need to be considered, not just for sprinkler systems, but for all services supported from timber ceilings.
Anzeige
This research looks at common industry screw fixings and loading capacities under ambient, fire, and post-fire conditions to provide information enabling considered decisions to be made for screw fixing into timber. It is very much needed, as lack of knowledge increases the risks that substantial loads may fall in the event of fire which might impede evacuation and response actions, allow fire spread across passive boundaries through broken ductwork, and increase consequential damage as fluid and gas bearing pipes incur mechanical damage.
Much is already known about fire in timber structures. Knowledge of timber fuel contributions with increased Heat Release Rate (HRR) [11, 34, 35], auto or self-extinguishment [12‐21], surface spread and travelling fires [36], compartment fire dynamics [37‐43], sprinkler protection [44‐54] watermist suppression [55, 56] in CLT compartments, and structural capacity of timber in fire [57] have been investigated.
Most fixings used to secure M&E services to timber ceilings are of the metal screw type. The ability of the screw fixing to hold load is determined at the contact surface between the threads and wood into which it is embedded [58‐68]. Metals are good conductors of heat and in fire screw fixings can transmit heat within, heating the timber and eventually turning timber to char and weakening the contact surfaces that assure its retaining capability. When a screw fixing is used to secure other metal framings in place, such as channel, brackets, or hanger rod this can increase the rate of heat uptake by the screw fixing, by substantially increasing the area for heat collection and conduction. For mounting services below a timber ceiling, larger diameter screw fixings are currently used to support larger loads as opposed to using more numerous or longer fixings.
Delamination of timber boards and char formation (mass loss) are known contributors to timber failure and structural loss during fire [15]. Delamination can occur in the area local to direct fire exposure from glue-line integrity failure and movement of the CLT boards. Screw fixing selection should therefore consider ply layer thickness and numbers in the determination of suitable embedment depths.
The aim of the study was to investigate key relationships between screw fixing diameter and embedment depth into exposed CLT for suspension of sprinkler systems and M&E services in ambient, fire and post-fire conditions. This was accomplished by load capacity testing during and after repeatable, and controllable fire testing. Test results captured data on pull-out forces, embedment depths, ply depths, and local temperatures. Knowledge gained from this study will provide information for improved guidance on screw fixing directly to timber, including CLT surfaces.
2 Potential Contributing Mechanisms of Timber Strength Loss
Wood is known to lose strength under the action of heat which can impact its ability to both support weight (loads in compression and tension), or hold in place devices attached by frictional fit, such as screws. Depending upon the temperatures reached some of these may result in reversible or irreversible loss of strength. The key mechanisms are:
Water evaporation
Thermal degradation of the polymer components of wood
Char formation
Wood contains ‘free’ water in cell lumens (transit routes), and ‘bound’ water in the cell walls. When heated to 100 °C, free water can boil resulting in internal vapour pressure. Up to 150 °C bound water vaporises causing cell-wall shrinkage and the formation of micro-cracks; defects which act to reduce the wood’s capacity to carry load.
The incremental loss of strength due to polymer degradation is as follows:
Between 200 °C and 300 °C hemicellulose, the flexible supporting mesh between plant fibres, loses rigidity.
On further heating (300 °C to 400 °C) the main load-bearing polymer cellulose degrades
Between 250 °C and 500 °C lignin, the ‘glue’ that binds together all fibres, softens and chars.
Char formation usually occurs between 300 °C and 400 °C during which time volatile gases escape leaving a smaller brittle carbonaceous char that cannot support compressive loads effectively. The loss of material during this process also contributes to overall loss of strength.
Table 1 below shows approximate retained bending strength for a dry softwood (e.g., pine) under standard test conditions. Exact values will vary by species, moisture content, and loading mode.
Table 1
Typical temperature to strength relationship for dry softwoods
Temperature (°C)
Retained Bending Strength (%)
Dominant Polymer Change
20 (room temp)
100
none
100
90–95
moisture loss
150
75–85
hemicellulose softening
200
50–70
hemicellulose degradation starts
250
30–50
rapid cellulose weakening
300
10–20
cellulose pyrolysis
> 350
< 5
residual char, negligible strength
An additional consideration specific to cross laminated timber is the role played by the composite layers and the type of glue used to bond them together. Some glue types soften and melt under the action of heat (thermoplastic), and some decompose and char (thermoset). Both can be used for CLT assembly. A fixing into CLT could conceivably fail due to the premature loss of strength of the glue holding the lamina in place into which it is fixed, or the wood itself weakening.
3 Methods and Materials
3.1 Development and Design of Bespoke Test Rig
A medium-scale test rig was developed incorporating CLT test panels with a wall to ceiling interface which subjected the panels to a 100 kW fire from the burner as shown in Fig. 1. The scale of the test rig allowed for exposure of an area measuring 0.81 m2 of ceiling in which to insert multiple test screw fixings. The primary heat source was a 170 × 170 × 170 mm gravel bed gas burner (as specified in ISO 9705 [69]), fuelled with propane gas via a Bronkhorst mass flow controller at a constant 100 kW. This was placed in contact with a vertical backing wall comprising sheets of 18 mm plywood which were gradually consumed during testing. This, together with a sacrificial section of CLT directly above the burner, helped ensure that the test samples of CLT were fully exposed to flames throughout the tests.
Fig. 1
The fire test rig. A Pre-fire, B During testing, and C Post-fire
See Fig. 1 shows the test rig which measured 2050 mm (L) × 1200 mm (W) × 1450 mm (H) overall. To ensure flames were channelled up the rear wall to the underside of the ceiling, and minimise the influence of draughts within the laboratory, the burner was framed with vertical plasterboard baffles. Full design of the test rig is shown in Fig. 2.
Fig. 2
Design of the bespoke test rig. Plan view and elevation on arrow ‘A’ and ‘B.’
Type ‘K’ thermocouples were used to measure temperatures below and within the CLT sample. Thermocouples were connected to a DataTaker data logger which in turn was connected to a PC for data recording.
3.2 Test Samples
3.2.1 Cross Laminated Timber (CLT)
The CLT test panels were 900 mm × 900 mm × 180 mm thick, 5-ply (consisting of layers 40, 30, 40, 30, 40 mm thick), replicating a typical floor/ceiling composition. The CLT was made from European spruce having a density ρ of 385 kg / m3 and was commercially available in Industrial Visual Quality (IVI) suitable for visually exposed applications. Panels were supplied with two different adhesive types consisting of standard Polyurethane (PU) adhesive, code HB S, and a heat-resistant modified-PU adhesive, code HB X which retains surface bonding at higher temperatures.
All panels including those for the ambient (baseline) tests were conditioned for a minimum of 4 weeks prior to testing in an enclosure controlled to 23 (± 2) °C and 50 (± 5)% relative humidity. Moisture content was measured and recorded in numerous locations on each panel and on the plywood rear wall before each test to ensure a common starting point for repeatability. Measurement of moisture content will be used as a cross reference to appreciate potential differences in results.
3.2.2 Fixings
Supporting sprinkler systems and M&E services into timber is not considered within current British or European standards [22 & 24]. BS EN 1995–1-1 (Eurocode 5) [70] allows for the ambient load carrying capacities to be evaluated through calculation but does not provide guidance on individual fixing requirements. Using European Test Approvals (ETAs) will provide fire test data on individual fixings tested in fire conditions. Fixings selected reflect the requirements of BS EN 12845:2015 + A1:2019 [22] for sprinkler systems suspended from concrete. Table 40 in BS EN 12845 identifies minimum cross section area of fixing based on minimum load capacity, reflecting different pipe diameters/loads. The minimum length of fixing is also specified for concrete.
Using Table 40 [22], 10 mm (M10) and 12 mm (M12) diameter screw fixings were selected based upon a minimum load capacity of 500 kg at 20 °C. This will support pipework up to 150 mm diameter. Screw fixings chosen for testing were M10 and M12 coach screws at 100 mm and 120 mm long respectively, shown in Fig. 3 (A and B). Embedment depths were chosen to allow for maximum penetration into the CLT with 100 mm screw fixings embedded at 90 mm and 120 mm screw fixings embedded at 110 mm (into 3-ply layers) [58]. Smaller embedment depths were evaluated using M10 and M12 screw fixings at 70 mm and 50 mm (into 2-ply layers), respectively.
Fig. 3
Screw fixings tested A—M12 coach screw, B – M10 (M8) coach screw, C – M12 stud screw, D – M10 stud screw (long), E – M10 stud screw (short) and F – M10 rod hanger (M6 thread) supplied by Midfix
Standard hanger-type industrial screw fixings, known as stud screws and rod hangers (M10 and M12), shown in Fig. 3 (C, D, E and F) were evaluated at their full embedment depth of 60 mm and smaller 8 mm (M8) coach screws in the same configuration as Fig. 3 (B) at 100 mm long embedded to 90 mm.
Screw fixings A, C, D, E and F are carbon steel with zinc coating to a thickness of 5 µm. Screw fixing B is hardened carbon steel, finished in zinc plating with a yellow chromate passivation to a thickness of 5 µm. It is acknowledged that there are a multitude of different screw types for use in timber that have different shank to thread ratios, including pitch angle and tip shapes.
The ability of a screw fixing to withstand an extractive force is dependent on the contact area of the thread with the wood, the amount of compression it exerts on the surrounding wood on insertion, and insertion angle in relation to the extraction force direction. Screw fixing engagement is also affected by the pilot hole (diameter and depth), shank diameter, thread diameter and pitch, all of which are variables which were investigated. M8 screw fixings were not piloted as advised by the fixing’s supplier. Pilot details for coach screw fixings were as follows:
M10 screw fixings (Fig. 3 B); 5.5 mm pilot hole – drilled full screw depth, 6.5 mm screw shank, 9.9 mm screw thread and 6.0 mm thread pitch.
M12 screw fixings (Fig. 3 A); 7.0 mm pilot hole – drilled full screw depth, 8.8 mm screw shank, 11.7 mm screw thread and 5.0 mm thread pitch.
3.3 Measurements
3.3.1 Pull-out Force Testing in Ambient Conditions
With screw fixings inserted into the samples, pull-out forces were recorded for all CLT panels. Figure 4 shows load testing conducted using the Staht t25 digital pull tester to BS 8539 [24].
Fig. 4
Staht t25 digital pull tester device at ambient and post-fire testing with individual result print-out
Table 2 below shows the tests performed at both ambient and post-fire condition for each combination of glue type fire exposure duration and fixing types. 13 tests in total were conducted, 5 ambient tests and 8 fire tests in order to allow pull-out force testing to be evaluated.
Table 2
Experimental tests undertaken at both ambient and post-fire conditions
3.3.2 Fire Conditions
Eight panels were fire tested incorporating screw fixings, and 41 mineral insulated type ‘K’ thermocouples, 1.5 mm dia × 300 mm long, were installed in locations below and within the CLT panel as shown in Fig. 5 and Table 3. Thermocouples were installed parallel and perpendicular to the exposed heated surface to record ceiling gas, screw (tip, mid-point, head), glue line and general ply temperatures. It is acknowledged that thermocouple output can be impacted through installation positions and orientations [71‐73]. With the exception of gas measuring thermocouples, all other thermocouples were installed on the non-fire side to reduce errors associated with conduction along the metal sheathing. To ensure thermocouples were touching the fixings at the mid-point and the tip, a jig was used for drilling accuracy until the drill tip contacted the embedded fixing. The thermocouple was then fully inserted, and contact was confirmed by metal-on-metal sound. The thermocouple was then retracted and insertion depth compared through measurement against drilling depth as a secondary check. Durations for fire testing were 30 and 60 min.
Fig. 5
Locations of thermocouples, Top: Internal locations. Bottom: Exposed/below the CLT panel. The depth of the thermocouples is detailed in Table 3
Of the 13 tests conducted, five used high temperature adhesive (HB X), and four tests FT3, FT4, FT5 and FT8, incorporated weighted testing of coach screws to a value of 150 kg. Tests FT3 and FT4 used M12 × 100 mm coach screws, and tests FT5 and FT8, M10 × 100 mm coach screws as detailed in Table 2.
A weight of mass 150 kg (1.471 kN) was selected for testing. This load represents a 150 mm diameter steel pipe 3.2 m long filled with water, commonly used in the sprinkler industry utilising a single bracket, meeting the requirements of BS EN 12845 [18].
Typical M&E service loads would also include ductwork, cable trays, conduit, filter units, sewerage pipework and potentially assemblies supported from the structure with different load requirements but were not specifically investigated in the current work.
3.3.3 Post-fire Conditions
Post-fire test measurements included pull-out force and char depths. Pull-out forces were measured once the sample had been allowed to cool for 60 min and took around 30 min to complete. Following pull-out testing, char was removed, and surface char depths were measured local to the screw fixings and in areas between the screw fixings.
4 Results and Discussion
4.1 Benchmark Ambient Pull-out Test Results
4.1.1 Coach Screws
To provide baseline pull-out-force measurements, four tests were initially conducted using pristine (non-fire tested) examples of both the HB S and HB X panels. For each panel, three of each screw fixing type (M10 × 100 mm, M10 × 120 mm, M12 × 100 mm and M12 × 120 mm) were embedded and then pull-out force determined at penetration depths of 50 mm, 70 mm, 90 mm and 110 mm. All screw fixings were equidistantly spaced and positioned so as not to lie along the joint of two wood segments.
Pull-out forces measured as a function of embedment depth are shown in Figs. 6 and 7. These illustrate the positive linear relationship between pull out force and embedment depth for the two panel types and two coach screw diameters. The increased pull-out resistance of the M12 coach screw relative to M10 coach screws is also evident for embedment depths greater than 50 mm. It is apparent that there is very little difference in results between the two different adhesives shown in the graphs.
Fig. 6
Pull-out force versus embedment depth for HB S adhesive, ambient conditions
It is assumed that the observed variability in pull out force at each depth is attributed to intrinsic variations in the physical properties of the wood (density, resin content, moisture content, structural inhomogeneity) and the panel’s construction (multiple wood lengths glued into multiple layers). Sub-surface knots may also have been a factor.
Results confirm that prior to fire exposure the capacities of the coach screws are capable of supporting the specified water filled sprinkler pipework (i.e. 4.9 kN (500 kg) at 20 °C (ambient) conditions detailed in Table 40 of BS EN 12845 [22].)
The linear line of best fit has been extended to reach an embedment depth of 5 mm with zero load capacity as it is confirmed that at this embedment depth the fixing holds its own weight.
4.1.2 Stud Screws and Rod Hangars
Stud screws and rod hangers were assessed at their maximum embedment depth of 60 mm with holes pre-drilled as advised by the supplier (see Table 2, Fig. 3 (C-F) and Fig. 8). These screw fixing types are currently supplied for suspension of M&E services in timber construction. Figure 8 shows the pull-out force results for each stud screw and rod hanger at ambient temperature. References/naming convention (A1 to D3) shown in Fig. 8 relate to location on the test panel not the fixing types detailed in Fig. 3.
Fig. 8
Standard Industrial-type screw fixings for M&E services (ambient conditions) (A1-3 M12 stud screw, B1-3 M10 stud screw-long, C1-3 M10 stud screw-short, and D1-3 M10 rod hanger)
Results show load capacities at both ambient (baseline) and post-fire in Table 4. Most notable from the results are:
The loss of strength from ambient to post-fire in the stud screws, with M12 screws losing 100% load capacity and all others greatly reduced.
The variation that can exist in pull out force between screw fixing of identical diameter and embedment depth.
Table 4
Average pull-out results for standard industrial stud screw and rod hanger fixings (showing uncertainty expressed as a ± 1 Standard Deviation (SD) showing the amount of variation in the measured results from the mean value)
4.2 Fire Challenge Assessment & Charring
Ideally the fire challenge tests would mimic the conditions in a fully developed fire in a full-scale compartment. Gas temperatures alone cannot describe the heat transfer into the CLT surfaces. Full-scale testing of CLT compartments and the potential for a different fuel source (propane) would provide a different thermal contribution from the fuel and re-radiation from adjacent surfaces. However, combining gas temperature data with the resulting char rates could give some indication of how the rig approximates to reported full-scale referenced tests.
4.2.1 Gas Temperatures
Gas/flame temperatures were measured 15 mm below the wood sample around the screw fixings shown in Fig. 5, (TC01 – TC06 inclusive). Temperature results for fire tests were all very similar. Figures 9 and 10 show average temperature data consistent with a fully developed compartment fire (i.e. > 600 °C). Temperatures were approximately consistent for 30 min for tests FT1, FT2, FT3, FT6 & FT7 and 60 min for tests FT4, FT5 & FT8. Gas temperatures were observed to be around 100 °C greater closer to the burner at TC04, 05 & 06, than those further from it at TC01, 02 & 03. This resulted in lower pull-out forces for the screw fixings located closer to the burner. For example, in FT1, three M10 × 120 mm screw fixings experienced pull-out forces of 9.1, 9.5 and 12.5 kN with locations 360, 540 and 720 mm from the S edge (see Fig. 5) moving away from the burner. These results are presented as an average value of 10.4 kN ± 1.5 Standard Deviation (SD).
Fig. 9
Average gas temperatures measured 15 mm below the underside of the CLT sample for 30 min, tests (FT1, 2, 3, 6 and 7)
It can be seen from Fig. 9 that the temperature for FT3 is lower than other 30 min fire tests. This was caused by frost developing on the cylinder affecting gas flow through the mass flow controller, causing a drop in output temperature (< 100 kW) from the burner at ~ 700 s. FT7 shows a higher temperature output which resulted from warming the cylinder during the test, providing a very steady burner output of 100 kW.
Test FT4 shows a lower than average gas temperature for the 60 min duration and a drop off at ~ 500–700 s. Burner output remained constant at 100 kW for 45 min until a drop off due to frost build up on the cylinders. The drop off at ~ 500 s could potentially be due to early char formation slowing contribution from the wood, but this is not proven in these experiments.
4.2.2 Char Formation
Char depths were measured following pull-out tests. Each sample had char carefully scraped away as shown in Fig. 11, using a wallpaper scraper and hand brush. To determine embedment depth on the resulting uneven surface, pilot holes were drilled through the full depth of the sample and measurements were taken of the remaining timber depth to provide char depth measurements shown in Fig. 18, not dissimilar to the procedure used by Brandon et al. [77].
Fig. 11
CLT panel with the char removed around the screw fixings (left) and pilot drill holes for remaining slab thickness measurements (right)
Measuring consisted of using a 3 mm diameter rod pushed through the pilot hole to a small flat circular plate ~ 15 mm diameter that sat on the face of the charred surface. The depth of insertion from the opposite face from the fire was marked on the rod then measured and recorded with digital callipers. Where screw fixings were removed from the sample, the circular plate was inserted into the depth of the ‘dishing’ and the same measuring principal applied. Figure 12 shows the details for char measurements taken.
Fig. 12
Shows how the char depths were measured with a 3 mm diameter rod pushed through pilot holes to the 15 mm diameter plate with marking point
Figures 13 and 14 show loss of structural material (char) from the surface of the sample. They also show the ‘dishing’ effect at the surface and blackening internally where screw fixings were installed. The extent of charring with depth is difficult to define, but internal temperatures measured (Table 7 and Fig. 18) confirm discolouring and charring, as described by Drysdale [76]. However, reduced pull-out results indicate that less unburnt timber is present around the M12 screws than M10.
Fig. 13
Cross section of tests FT1 and FT2 showing loss of material at the surface of the CLT panel and the ‘dishing’ effect localised around the screw fixings after 30 min fire exposure
Cross section of tests FT4 and FT5 showing loss of material at the surface of the CLT panel and the ‘dishing’ effect localised around the screw fixings after 60 min fire exposure
Table 5 (general) and Table 6 (at fixings) show loss of material due to char formation on the surface, which are slightly less than measurements recorded at the ceiling by Hadden et al. [13].
Table 5
Average loss of structural material measured in locations between screw fixings for both adhesive types after the fire
Table 6
Average loss of structural material measured at screw fixing positions for both adhesive types after the fire
Figure 15 shows a close-up of the charring effect of heat conduction through the screw fixings, with greater charring at the M12 coach screws than the M10 coach screws, which was observed across all fire tests. The M10 coach screw shows partial charring through the screw embedment and rupture of the wood fibres towards the location of the tip.
Fig. 15
Enlarged view of internal charring through heat conduction at the coach screws
Post-fire test results confirm that the larger diameter of the screw fixing, the greater the loss of char local to its insertion area.
Whilst Tables 5 and 6 reference a typical charring rate between 0.6–0.7 mm/min, it is acknowledged that charring rates are higher in the initial stages of a fire and reduce as the char layer forms. It is confirmed that charring rates differ between wood species and fire exposure, where charring rate increases with heat flux [78].
4.2.3 Fire Challenge Conclusions
Mean gas temperatures of ~ 800 °C show the fire to be significant, but the average charring rate at locations remote from the screw fittings of 0.32 mm/min is less than that measured during full-scale tests conducted by Haddon [13] of 0.6 mm/min at the ceiling. This would suggest that the fire conditions in this study are less severe than for a typical well-developed fire. This may be due to the design of the test rig and the lack of oxygen (needed for char oxidation) at the timber surface. This deviation needs to be considered as a significant factor in the overall conclusions drawn from this study.
4.3 CLT and Fixing Temperatures
4.3.1 CLT Temperatures
Measurements within the sample show an increase in temperature in all 5-ply layers. After cut-off of the burner gas supply, internal temperatures continued to rise slowly which may contribute to further loss of load capacity post-fire extinguishment. This can be seen in Fig. 16 for measurements at 20 mm above the exposed CLT and is consistent with findings by Gernay et al. [50] with general locations of thermocouples shown in Fig. 17. It is also noticeable for tests FT1, 2, 3, 4, 7 and 8 that temperature remains at a constant 100 ℃ for a period of time, until the moisture is driven off, before temperature starts to increase. Tests FT5 and FT6 show inconsistent results which could be due to the inconsistencies in CLT moisture condition at the time of the test.
Fig. 16
Measurements of temperature within the CLT test sample at 20 mm above the exposed CLT surface at 30 and 60 min
In all but one fire test, no fragments of char were seen to fall off the test panels. Fire test 8 had small fragments of ~ 10 mm depth of char, fall from a very small area of the panel < 10%. In test FT4 (a 60 min test), delamination at the glue line in a sacrificial HB S specimen (directly above the burner and not part of the test sample) was observed.
4.3.2 Screw Fixing Temperatures
Tests FT1, 2, 3, 4, 5 and FT8 utilised coach screws shown in Fig. 11. A consistent temperature rise from exposed screw head shank to the embedded screw tip was observed as shown in Table 7 and Fig. 18.
Table 7
Average gas temperatures, wood temperatures at 55 mm above the exposed CLT sample and fixing temperatures in three locations at 30 and 60 min
Test Reference/Measurement location
FT1
M10 / M12
FT2
M10 / M12
FT3
M10 / M12
FT4
M10 / M12
FT5
M10 / M12
FT8
M10 / M12
Average gas temperature @ 30 min (⁰C)
725
736
658
718
773
811
Wood temperature, 55 mm depth @ 30 min (⁰C)
13
20
19
19
15
20
Fixing head/shank temperature @ 30 min (⁰C)
717 / 633
662 / 649
610 / 554
625 / 626
641 / 635
600 / 675
Fixing mid-point (55 mm) temperature @ 30 min (⁰C)
133 / 224
217 / 206
197 / 250
182 / 238
216 / 292
212 / 262
Fixing tip temperature @ 30 min (⁰C)
65 / 87
51 / 167
59 / 104
42 / 117
44 / 73
55 / 110
Average gas temperature @ 60 min (⁰C)
X
X
X
673
651
823
Wood temperature, 55 mm depth @ 60 min (⁰C)
X
X
X
52
43
55
Fixing head/shank temperature @ 60 min (⁰C)
X
X
X
602 / 596
644 / 583
616 / 690
Fixing mid-point (55 mm) temperature @ 60 min (⁰C)
X
X
X
243 / 302
282 / 335
273 / 345
Fixing tip temperature @ 60 min (⁰C)
X
X
X
57 / 125
58 / 99
83 / 127
Note: X represents no data recorded post 30 min for fire tests FT1, FT2 and FT3
Fig. 18
Average screw fixing temperatures recorded as a function of depth at the CLT (screw shank), at the mid-point (internal) the screw tip (internal) and at 55 mm above the exposed CLT at 30 min
Measured temperatures at the screw head shank, mid-point and tip for coach screws (M10 and M12) for tests FT1, 2, 3, 4, 5 and 8 are shown in Table 7 and Fig. 18 at 30- and 60-min fire durations. At 30 min, screw head shank temperatures ranged from 554—717 ℃, mid-point ranged from 133 – 292 ℃, and tip ranged from 42—167 ℃, with an average gas temperature ranging from 658—811 ℃. At 60 min, screw head shank temperatures ranged from 583—690 ℃, mid-point ranged from 243 – 345 ℃, and tip ranged from 57—127 ℃, with an average gas temperature ranging from 651—823 ℃. In stark contrast to the lower thermal conductivity of timber, much higher temperatures were reached adjacent to the screws which would account for the degradation of the load capacity of the timber as detailed by Khelifa et al. [74]. The mid-point screw temperatures at a depth of 55 mm, are observed to be much greater than for the timber at the same time and depth shown in Table 7 and Fig. 18. Measurements within the timber confirm larger diameter M12 screws consistently result in higher temperatures than M10.
4.4 Fire Pull-out Test Results
4.4.1 Pull-out Forces – Coach Screws
Figures 19 and 20 show pull-out force versus embedment depths for each CLT adhesive type at ambient, 30 min and 60 min fire exposure. Embedment depth scatter shown in the graphs for post-fire test results derive from actual embedment depths where char has been removed. For example, after the fire actual embedment depth of 73 – 77 mm was found for pre-fire embedment depth of 90 mm.
Fig. 19
Pull-out force versus embedment depth for HB S adhesive, ambient conditions, 30 and 60 min fire exposure
Comparison of the two graphs (Fig. 19 and 20) show a similar relationship in loss of load versus embedment depth. Above 50 mm embedment and at ambient conditions, M12 screw fixings having a greater load capacity than M10. However, post-fire results indicate the opposite, M10 screw fixings retain a greater load than M12. This result might seem counter intuitive but is believed to result from greater heat transfer into the CLT through the wider screws resulting in higher temperatures at the interface between the thread and the timber leading to charring and loss of strength. The steeper slope of the post-fire tests shows the increased beneficial influence of embedment depth on the load capacity.
Figure 21 shows pull-out forces versus embedment depths for coach screws, measured across all tests (benchmark ambient, and post-fire) irrespective of CLT adhesive type.
Fig. 21
Pull-out force versus embedment depth, all results, ambient conditions, 30 and 60 min fire exposure
Table 40 of BS EN 12845 requires a load capacity of 500 kg (4.9 kN) at ~ 20 °C, benchmark ambient loads suggest this is achievable with M10 screw fixings at an embedment depth of ~ 31 mm and M12 at ~ 25 mm (indicated in Fig. 21) assuming linear extrapolation. At ambient conditions with a minimum of 2-ply embedment ~ 70 mm, load capacity for M10 and M12 fixings is just over twice the requirement of BS EN 12845.
Where screw fixings were tested at ambient conditions, load capacity exceeds the minimum specification. However, during and after fire there is significant reduction in load capacity when compared to ambient load measurements. This is due to the increase in timber temperature and char formation which leads to loss of material at the surface and loss of strength of the timber [74] and around the screw fixing. Without load, the screw fixing remains in place. However, char at these contact interfaces will increase in thermal conductivity as temperature rises [75] leading to increased heat conduction through the screw.
Four fire tests were conducted with screw fixings statically loaded replicating common sprinkler service loads. Results from these tests are discussed further in Sect. 4.4.2.
4.4.2 Statically Loaded, During Fire Tests
With screw fixings loaded, char formation at the contact point through the screw thread will be under compression, pulling the screw fixing in the direction of the force and leaving a small char void above the thread and compressing the char layer below, shown in Fig. 22. Contact surfaces between the wood and screw thread are reduced. Once the char extends to the outer circumference of the screw thread, the effective embedment depth is reduced to a point where detachment occurs.
Fig. 22
Shows a coach screw under load, heat conduction through the fixing leading to char formation at the contact point between screw thread and timber
This is evident in the two statically loaded tests (FT3 & FT4) which recorded catastrophic failure of the screw fixing at ~ 20 min (full pull-out). In the other two loaded tests (FT5 & FT8) the cable failed at ~ 10 min and ~ 19 min respectively, with the screw fixing remaining in place. However, pull-out forces of the loaded screw fixings were significantly reduced. The average capacity of the same unloaded screw fixings in tests FT5 & FT8, were recorded at 5.4 kN and 4.4 kN, indicating a significant loss from ambient test results. Loaded fixings were recorded at 3.0 kN and 1.4 kN respectively, representing a 56% and 47% loss from unloaded screws due to load.
At temperatures above 200–250 °C, and through prolonged heating at lower temperatures ~ > 120 ℃, wood discolours and chars, described by Drysdale [76]. Screw fixings A2 (M10 × 120 mm) and D2 (M12 × 120 mm) shown in Fig. 5 and Fig. 11, included thermocouples located at the screw head shank, mid-point, and tip. Temperatures were recorded in fire tests FT1, FT2, FT3, FT6 and FT7 at 30 min. The results have been averaged showing uncertainty expressed as a standard deviation for each screw fixing type.
Using ambient baseline results, the remaining capacity for the coach screws once cooled is shown in Table 8. This highlights the severity of the load capacity reduction for M12 screw fixings. This has significant implications for the M&E industries fixing into timber.
Table 8
Capacity of M10 and M12 (coach screws) post-fire at various embedment depths
Fixing Type
Embedment depth (mm)
Ambient mean pull-out force (kN) ± 1 SD
30 min fire exposure mean pull-out force (kN) ± 1 SD
60 min fire exposure mean pull-out force (kN) ± 1 SD
M10 coach screw (A2)
110
17.6 ± 2.3
11.9 ± 1.8
10.1 ± 1.7
90
15.5 ± 2.1
6.9 ± 1.2
4.7 ± 1.8
70
13.2 ± 2.1
–
–
50
8.4 ± 1.1
0#
–
M12 coach screw (D2)
110
24.4 ± 2.6
11.0 ± 3.1
5.9 ± 2.8
90
17.8 ± 2.3
3.3 ± 2.0
1.3 ± 0.9
70
15.1 ± 1.1
–
–
50
9.0 ± 0.6
0#
–
Note: 70 mm embedment depth and 50 mm embedment for 60 min was not fire tested
0# Based on recorded pull-out force of 0
M8 × 100 mm screw fixings were assessed after 30 and 60 min fire exposure. Screw fixings were embedded to a depth of 90 mm. After fire testing, pull-out forces after 30 and 60 min were measured as 7.6 kN and 7.7 kN, respectively. For comparison, both M10 and M12 screw fixings embedded at 90 mm, M10 measured (averaged) at 6.9 kN and 4.7 kN respectively, with M12 screw fixings measuring (averaged) 3.3 kN and 1.3 kN, respectively. The smaller diameter of the M8 screw fixings will conduct less heat which has been shown to reduce the loss of load capacity.
5 Conclusions
This study aimed to characterise the load bearing capacity of screw fixings of the type used to secure M&E to timber ceilings under fire conditions. The potential consequence of early failure of these fittings include entanglement, crush, and electrocution risks to evacuees and attending fire services; enhanced fire spread through fire compartment boundaries via broken and dislodged vent ducts; and increased consequential damage and impairment of systems (including fire sprinklers) from mechanical damage to fallen fluid filled pipes (liquid and gas).
Under ambient conditions screw fittings into timber were shown to comfortably exceed the load bearing requirements for typical M&E suspension applications and larger diameter screws were also shown to support greater loads.
When tested under fire conditions two factors can be considered to contribute to the loss, the complete loss of material near the surface due to char oxidation, and the weakening of the internal timber around the screw thread by charring. This results from the metal of the screw fixing conducting heat from the attached brackets into the timber and around the thread promoting char formation, coupled with the propensity of timber to weaken under the action of heat.
This study also shows that the timber is weaker during the fire than after it has cooled, although the data to support this observation is limited.
The results show that screws with the same overall width and embedment depth, but narrower shanks and wider threads have greater retention capacity in timber exposed to fire. In addition the wider shank of the M12 screws conducted more heat to the ‘vulnerable volume’ of wood between the shank and the outside of the thread than the narrower M8 or M10 screws causing more charring and greater loss of load capacity. This effect was most noticeable for screws with the greatest overall width/shank ratio (for example comparing the left and right hand screws in Fig. 17).
The complete loss of strength of screw fixings under loaded fire conditions was both very rapid and structurally significant indicating that the aforementioned concerns highlighted relating to the potential consequences of failure are both justified, highly plausible, and in need of a further detailed investigation, and a solution.
In spite of the very significant load bearing changes observed under the set fire conditions, analysis of char rates indicate that more onerous conditions are likely at in fires involving mass timber construction. Compounded by the likelihood that the screw fixing will additionally be attached to greater amounts of heat-collecting metalwork that will also be in the flame zone, it should be assumed that loss of strength will occur sooner in real fires. It is also noted that post-fire, on cooling there is an improvement in load bearing capacity of the timber substrate, indicating that it is essential to have more detailed information on the performance of fixings in the loaded fire condition.
This study was limited to considering screw fixings being installed in traditional arrangements i.e. perpendicular to the surface and concentrated on screw fixing diameters specified to the installation of fire sprinkler systems as described in European sprinkler standard BS EN 12845 [18].
The inclusion of a suitably designed and fully functional sprinkler system in a building’s design, means the potential for fire spread will be reduced, gas temperatures will be lower, and the impact upon the load-bearing capacity of the screw fixings holding the sprinkler system itself, and other M&E services will be greatly reduced. Without a sprinkler system, M&E services are vulnerable to complete and rapid failure that will result in premature detachment of suspended services– it is clear that for this situation new approaches to fixings are required, properly supported by standards and test protocols, to ensure this risk is annulled.
6 Future Work
Whilst this study was originally undertaken to gain knowledge on how fixing methods for fire sprinkler systems might need adaption for the mass timber situation, it has highlighted more serious implications for M&E in general. To ensure building safety under fire conditions and to support evacuation of occupants and response by the attending fire service, better understanding of the load bearing capacity of timber in fire is needed. There are also implications for the level of consequential damage the building might incur through fire spread and damage to other ceiling supported systems.
Potential solutions may include the development of new types of fixings, or the use of current screw fixings in a different way where the load is out of plane to the fixing insertion. The development of appropriate test standards and specification requirements are also required in response to these new building methods.
Based on the loss of load capacity described, it is expected that decisive action will be taken to prevent detachment of services and the foreseeable potential implications for escapee or firefighter entanglement, similar to that which occurred at Shirley Towers. In that instance it was the fixings that had a vulnerability to fire, not the substrate into which they were embedded, so whilst the outcomes could be the same, the necessary solutions will be different. This work provides stakeholders with information necessary to develop a new standard or guidance document for fixing M&E services into timber based upon the findings within this paper, which is urgently required.
Acknowledgements
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. RISCAuthority, through The Fire Protection Association (FPA) provided the majority of equipment used for testing as the primary author is an FPA employee. The FPA had no other involvement in the study design, collection, analysis, or interpretation of data. They also were not involved in the writing of the report; and in the decision to submit the article for publication. Materials and equipment were kindly sponsored by Midfix, and Staht. This work would also not have been possible without the help of Stevan Grkinic, Nigel Rogers, Rob Hirst, Ian Abley, Richard Glover, and George Edwardes.
Declarations
Conflict of interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Structural Timber Association - Structural timber buildings - fire safety in use guidance - Volume 6 - Mass timber structures; Building Regulation compliance B3(1) - STA fire safety research and guidance project - Version v1.1 October 2020
2.
Hopkin D, Spearpoint M (2020) Compliance road-map for the structural fire safety design of mass timber buildings in England. SFPE Europe Q4:2020
3.
Hopkin D, Gorska C, Spearpoint MJ, Fu I, Krenn H, Sleik T, Stapf G (2021) Experimental characterisation of the fire behaviour of CLT ceiling elements from different leading suppliers. Appl Structl Fire Eng 2021:10–11. https://doi.org/10.14264/5010564CrossRef
4.
Hopkin D, W˛egrzy´nski W, Spearpoint M, Fu I, Krenn H, Sleik T, Gorska C, Stapf G (2022) Large-scale enclosure fire experiments adopting CLT slabs with different types of polyurethane adhesives: genesis and preliminary findings. Fire 2022(5):39. https://doi.org/10.3390/fire5020039CrossRef
Östman, B, A-L. (2013). Fire Safety in timber buildings [Paper presentation]. Seminário Casas de Maderia.
8.
Östman, B., Mikkola, E., Stein, R., Frangi, A., König, J., Dhima, D., Hakkarainen, T., & Bregulla, J. (2010). Fire safety in timber buildings: technical guideline for Europe. SP Report No. 2010:19
9.
Zelinka, S, L., Hasburgh, L., Bourne, K., Tucholski, D., & Ouellette, J. (2018). Compartment fire testing of a two-story mass timber building. https://doi.org/10.13140/RG.2.2.26223.33447
10.
Mitchell H, Kotsovinos P, Richter F, Thomson D, Barber D, Rein G (2023) Review of fire experiments in mass timber compartments: current understanding, limitations, and research gaps. Fire Mater 47(4):415–432. https://doi.org/10.1002/fam.3121CrossRef
11.
Babrauskas V (2023) Comments on “The curious case of the second/end peak in the heat release rate of wood: a cone calorimeter investigation,” by Sanned et al. Fire Mater 47(5):733–734. https://doi.org/10.1002/fam.3143CrossRef
12.
BS EN ISO 13943:2017 – Fire safety – Vocabulary
13.
Hadden R, M. Bartlett A, I. Hidalgo J, P. Santamaria S, Wiesner F, Bisby L, A. Deeny S, Lane B (2017) Effects of exposed cross laminated timber on compartment fire dynamics. Fire Saf J 91:480–489. https://doi.org/10.1016/j.firesaf.2017.03.074CrossRef
14.
Bartlett AI, Hadden RM, Hidalgo JP, Santamaria S, Wiesner F, Bisby LA, Deeny S, Lane B (2017) Auto-extinction of engineered timber: application to compartment fires with exposed timber surfaces. Fire Safety J 91:407–413. https://doi.org/10.1016/j.firesaf.2017.03.050CrossRef
Nothard S, Lange D, Hidalgo J P, Gupta V, McLaggan MS, Wiesner F, Torero JL (2022) Factors influencing the fire dynamics in open-plan compartments with an exposed timber ceiling. Fire Saf J. https://doi.org/10.1016/j.firesaf.2022.103564CrossRef
21.
Xu H, Pope I, Gupta V, Cadena J, Carrascal J, Lange D, McLaggan M S, Mendez J, Osorio A, Solarte A, Soriguer D, Torero J L, Wiesner F, Zaben A, Hidalgo J P (2022) Large-scale compartment fires to develop a self-extinction design framework for mass timber Part 1”: literature review and methodology. Fire Saf J. https://doi.org/10.1016/j.firesaf.2022.103523CrossRef
22.
BS EN 12845:2015+A1:2019 Incorporating corrigenda December 2015 and January 2016 - Fixed firefighting systems - Automatic sprinkler systems - Design, installation and maintenance.
Bartlett A, McNamee I, Robert R, Bisby F, Bisby L, Bisby A (2020) Comparative energy analysis from fire resistance tests on combustible versus non-combustible slabs. Fire Mater 44:301–310. https://doi.org/10.1002/fam.2760CrossRef
36.
Rackauskaite E, Kotsovinos P, Barber D (2021) Letter to the editor: design fires for open-plan buildings with exposed mass-timber ceiling. Fire Technol 57:487–495. https://doi.org/10.1007/s10694-020-01047-0CrossRef
37.
Shuna N, Gernay T (2022) On the effect of exposed timber on the severity of structural fires in a compartment and required firefighting resources. Fire Technol 58:2691–2725. https://doi.org/10.1007/s10694-022-01254-xCrossRef
38.
Zelinka SL, Hasburgh LE, Bourne KJ, Tucholski DR, Ouellette JP, Kochkin V (2018) Compartment fire testing of a two-story mass timber building. Gen Tech Report Forest Prod Lab. https://doi.org/10.2737/FPL-GTR-247CrossRef
39.
Madden, J., Kriel, M., Wiesner, F., Wu, W., Hilditch, R., Ervine, A., Lange, D. (2024) Burning behaviour of a timber ceiling: a bench-scale investigation [Paper presentation]. 13th International conference on structures in fire. Coimbra, Portugal.
40.
Xu H, Pope I, Gupta V, Cadena J, Carrascal J, Lange D, McLaggan M, McLaggan S, Mendez J, Osorio A, Solarte A, Soriguer D, Torero J, Torero L, Wiesner F, Zaben A, Hidalgo J, Hidalgo P (2022) Large-scale compartment fires to develop a self-extinction design framework for mass timber—Part 1: literature review and methodology. Fire Saf J 128:103523. https://doi.org/10.1016/j.firesaf.2022.103523CrossRef
41.
Emberley R, Gorska Putynska C, Bolanos A, Lucherini A, Solarte A, Soriguer D, Gutierrez Gonzalez M, Humphreys K, Hidalgo J, Hidalgo P, Maluk C, Law A, Torero J, Torero L (2017) escription of small and large-scale cross laminated timber fire tests. Fire Saf J 91:327–335. https://doi.org/10.1016/j.firesaf.2017.03.024CrossRef
42.
Bøe A, Bøe S, Friquin K, Friquin L, Brandon D, Steen-Hansen A, Ertesvåg I, Ertesvåg S (2023) Fire spread in a large compartment with exposed cross-laminated timber and open ventilation conditions: #FRIC-01 – exposed ceiling. Fire Saf J. https://doi.org/10.1016/j.firesaf.2023.103869CrossRef
Khoat H, Khoat T, Kim J, Kim T, Quoc T, Quoc D, Kwark J, Kwark H, Ryou H, Ryou S (2020) A numerical analysis of the fire characteristics after sprinkler activation in the compartment fire. Energies 13(12):3099. https://doi.org/10.3390/en13123099CrossRef
Building, Design and Management - Fire in timber frame buildings – “A review of fire statistics from the UK and the USA” - BDM14 - First published 2011Version 01 By RISCAuthority and administered by The Fire Protection Association. [Online]. Available: https://www.thefpa.co.uk/resource-download/243
53.
Hoehler, M, S., Su, J. Lafrance, P-S. Bundy, M, F. Kimball, A. Brandon, D. Östman, B. (2018) Fire safety challenges of tall wood buildings: large-scale cross-laminated timber compartment fire tests In: SiF 2018– The 10th International Conference on Structures in FireFireSERT. [Online]. Available: http://urn.kb.se/resolve?urn=urn:nbn:se:lnu:diva-77612
54.
Sæter AB, Hox K, Mikalsen RF, Friquin KL (2024) Large-scale fire experiments in a cross-laminated timber compartment with an adjacent corridor – Partly and fully protected with a water sprinkler system. Fire Safety J. https://doi.org/10.1016/j.firesaf.2024.1042122CrossRef
Kotsovinos P, Rackauskaite E, Christensen E, Glew A, O’Loughlin E, Mitchell H, Amin R, Robert F, Heidari M, Barber D, Rein G, Schulz J (2022) “Fire dynamics inside a large and open-plan compartment with exposed timber ceiling and columns”: CodeRed #01. Fire Mater. https://doi.org/10.1002/fam.3049CrossRef
57.
Gernay T, Zehfuß J, Brunkhorst S, Robert F, Bamonte P, McNamee R, Mohaine S, Franssen J (2022) Experimental investigation of structural failure during the cooling phase of a fire: timber columns. Fire Mater. https://doi.org/10.1002/fam.3110CrossRef
Saito, S., Kikuchi, T., Nakayama, Y. Totsuka, M., Hirashima, T. (2024) Embedment strength of dowelled connections of glulam timber at high temperatures. Paper presented at the conference on structures in fire. University of Coimbra, Portugal. 1359–1370. https://doi.org/10.30779/cmm_SIF24
62.
Building Engineering Services Association (BESA) – TB58 Alternative systems for building services supports – Version 1 January 2024.
Horne P, Abu A, Palermo A, Moss P (2023) Thermal response of timber connections in the cooling phase. Fire Mater 47:479–497. https://doi.org/10.1002/fam.3136CrossRef
65.
Abdoli F, Rashidi M, Rostampour-Haftkhani A, Layeghi M, Ebrahimi G (2022) Withdrawal performance of nails and screws in cross-laminated timber (CLT) made of poplar (Populus alba) and fir (Abies alba). Polymers 14:3129. https://doi.org/10.3390/polym14153129CrossRef
Hoelz K, Dörner P, Hohlweg J, Matthiesen S (2022) Influence of thread parameters on the withdrawal capacity of wood screws to optimize the thread geometry. Eur J Wood Wood Prod 80:529–540. https://doi.org/10.1007/s00107-022-01792-9CrossRef
68.
Peng L, Hadjisophocleus G, Mehaffey J, Mohammad M (2011) Predicting the fire resistance of wood–steel–wood timber connections. Fire Technol 47:1101–1119. https://doi.org/10.1007/s10694-009-0118-4CrossRef
69.
BS 476–33:1993, ISO 9705:1993 Fire tests on building materials and structures — Part 33: Full-scale room test for surface products.
70.
BS EN 1995–1–1:2004 Eurocode 5: Design of timber structures - Part 1–1: General - Common rules and rules for buildings
Pope I, Hidalgo JP, Torero JL (2021) A correction method for thermal disturbances induced by thermocouples in a low-conductivity charring material. Fire Saf J 120:103077. https://doi.org/10.1016/j.firesaf.2020.103077CrossRef
73.
Terrei L, Acem Z, Marchetti V, Lardet P, Boulet P (2021) In-depth wood temperature measurement using embedded thin wire thermocouples in cone calorimeter tests. Int J Therm Sci 162:106686. https://doi.org/10.1016/j.ijthermalsci.2020.106686CrossRef
74.
Khelifa M, Thi VD, Oudjène M, Khennane A, Ganaoui ME, Rogaume Y (2024) Modelling the response of timber beams under fire. Int J Civ Eng 22:1537–1549. https://doi.org/10.1007/s40999-024-00973-2CrossRef
75.
Shi L, Chew MYL (2021) A review of thermal properties of timber and char at elevated temperatures. Indoor Built Environ 32(I):9–24
76.
Drysdale D (1999) An introduction to fire dynamics. Wiley, Hoboken
Xu Q, Chen L, Harries K, Zhang F, Liu Q, Feng J (2015) Combustion and charring properties of five common constructional wood species from cone calorimeter tests. Constr Build Mater 96:416–427. https://doi.org/10.1016/j.conbuildmat.2015.08.062CrossRef
