Forward and Reverse shear transfer in beech LVL-concrete composites with singly inclined coach screw connectors
Introduction
The slip modulus, longitudinal shear strength and ductility of timber-concrete connections often strongly influence the deflections, vibration characteristics, load capacity and stress redistribution capabilities of timber-concrete composite (TCC) floors, an example of which is shown in Fig. 1. It is thus important to quantify these connection properties. To that end multiple studies [1], [2], [3], [4], [5] show that, by comparison with test data, existing standards predict these properties to quite variable degrees of accuracy. Thus, into the foreseeable future, testing will be the primary means of characterising connection behaviour.
TCC connections based on mechanical connectors have been extensively characterised in experimental studies to date. These include nails or nail-plates [1], [5], [6], [7], vertical or inclined screws [6], [8], [9], [10], [11], [12], dowels [4], [13], [14], perforated steel plates or steel meshes bonded into the timber [3], [15], [16], [17], and notches [3], [6], [18], [19], [20], [21], [22]. The tests have considered both normal and lightweight concretes, short-term and long-term behaviours, and the presence or absence of a formwork interlayer. In one study [21], rectangular notched connections reinforced with coach screws were found to outperform toothed metal plate connections under fatigue loading for application to road bridges. Other forms of mechanical connector include a “flowering” steel cylinder installed using a cartridge-operated gun [23], others which performed as well with prefabricated slabs as with cast in-situ slabs [24], and a hybrid cylindrical connector comprising an ultra-high performance fibre reinforced concrete (UHPFRC) annulus surrounding a solid circular steel core [25].
Adhesive (hence chemical) connections have also been investigated [26], [27], [28], [29], [30] as a means to avoid drilling into the timber, and to improve TCC connection (and so member) stiffness by reducing interface slip. Both dry slab-on-wet adhesive and wet concrete-on-wet adhesive options have been studied. Key fabrication variables (e.g. the pouring height for the concrete in the wet-on-wet option), connection performance under combined hygrothermal and mechanical loads, and long-term behaviour (up to 4.5 years) have all been experimentally verified in these studies.
A review [31] of relevant experimental research concluded that adhesive connections and connections comprising long notches with dowels are of superior slip stiffness and strength but are also brittle, while metal plate connectors and connections comprising short notches with dowels give excellent stiffness, excellent strength and moderate ductility. Finally, the review highlighted that dowel type connectors can combine good stiffness and strength with excellent ductility.
Each of the key structural properties of TCC connections has come under scrutiny in targeted studies. For example, longitudinal shear strength was found [14] to be palpably higher for the concrete-timber-concrete double shear test configuration than for the timber-concrete-timber alternative. Moreover, high ductility of the connections was shown [32], [33] to lead to user-friendly analysis approaches for TCC members at ultimate, and also in one case [34] to enable up to 87% and 144% increases in ultimate load and midspan deflection, respectively. Finally, the slip modulus deduced from testing of TCC connections was found [35] to need a 60% reduction in cracked concrete zones of a FE model, to enable reliable prediction of deflection for the associated TCC member.
The aim of the study reported in this paper was to characterise the quasi-static behaviours of TCC connections based on beech LVL, a recent advance in engineered timber. The objectives were to:
- •
Understand the fabrication issues arising from driving large diameter coach screw connectors parallel to each other into beech (a hardwood) LVL, so that all screws are at a single inclination.
- •
Experimentally determine, separately under forward and reverse longitudinal shear, the constitutive behaviours of the TCC connections based on these singly inclined coach screws.
This study builds significantly on the above discussed and other experimental investigations into TCC connections, for the following key reasons, namely:
- •
Hardwood LVL is a recent innovation which represents engineered timber with the best possible structural timber properties, best consistency of those properties and highest efficiency (double that of glulam) of use of original wood. Indeed in all three senses LVL is superior to glulam, owing to the circumferential peeling of the very thin LVL layers off the original log, giving little waste and minimising the influence of defects on overall section properties, as opposed to the longitudinal machining of the same log to give the much thicker, more defect-sensitive layers and more waste for glulam. Given this recency of hardwood LVL, it is not surprising that only very few studies, including two [36], [37] on beech LVL-concrete composite slabs and one [38] on beech LVL-concrete composite beams, have been reported using this advance in engineered timber. Hence the study presented in this paper advances previous work because that earlier work by necessity could only employ hardwood or softwood original logs or glulam, or softwood LVL, but not hardwood LVL. Given the high and consistent strength of beech LVL, the use of large section coach screws is highly advisable, in order to draw out as much capacity as possible from this connection.
- •
Coach screws are installation-friendly fasteners which lead to connections with good structural properties. Now if the dominant design load for any given TCC floor is a UDL, it is prudent to use these screws in a singly inclined layout symmetrically about midspan, with the screw-heads pointing towards the supports, because that layout endows the resulting connections with high slip stiffness and strength. It must then be recognised that in service, the connections may well experience longitudinal shear reversal due to shrinkage of the concrete (if it is cast wet around the connectors), also due to moisture-induced expansion of the timber, possibly due to oscillatory seismic loading or to lorries moving along a TCC bridge deck, etc. Given the asymmetry of this connection about the vertical, such reversal induces a different constitutive behaviour of the connection. For that reason, it is strongly advisable to test singly inclined screw TCC connections alternately under longitudinal shear in opposite directions. This has influenced the testing regime adopted in the present study.
In what follows the hardwood LVL-concrete connection specimens and the reverse shear tests to which they were subjected are described, the resulting constitutive behaviours are quantified using the recorded test data, and the connections are compared. The paper then closes with some suggestions for further work.
Section snippets
Local details
Twelve double-shear beech LVL-concrete connection specimens, each with a central LVL stub flanked by concrete slabs, were fabricated and tested in this study. All connection specimens included the following features, namely:
- •
400 mm deep × 200 mm wide sections of GL70 Baubuche beech LVL joists.
- •
M12 coach screws of 180 mm length.
- •
Concrete of a specified C32/40 nominal grade although, as seen later, the actual concrete was stronger.
Within the twelve specimens three screw inclination angles to the
Material properties
Manufacturer-supplied properties for the beech LVL material are provided in Table 2. From crushing and split tests performed on the cubes and cylinders, the individual, average and CoV compressive and tensile strengths of Table 3 were obtained. Note the high consistency of results (low CoV) and the high-strength concrete (compressive 51.9 N/mm2) achieved.
Testing strategy
Double shear testing of specimens was performed in a 500 kN capacity DARTEC machine. Fig. 7 shows a specimen under test. Roller and pin supports were used under the slabs. Load was applied onto the top section (horizontal) face of the LVL stub halfway between the slabs, using a steel plate and thin rubber pad to spread the load onto that top section. At mid-height of each set of four screws (hidden in Fig. 7), slip was measured at the slab-joist interfaces using calibrated potentiometers. In
Slip moduli and longitudinal shear force capacities
For each specimen, Table 4 summarises the longitudinal shear force capacities, failure modes and slip moduli at 40% failure load per four screw connector group. Two alternative definitions of slip modulus have been used, both based on advice in EN 26891: 1991 [40]. Taking ν0.4 and ν0.1 as the slips at 40% and 10% respectively of the peak force, Fmax, these slips are [40]:These definitions render ki the true secant modulus and ks a modified secant
Comparisons with previously tested screw-based TCC connections
Blass et al. [6] reported the results of longitudinal shear tests on different TCC connections, including several based on self-tapping screws laid in X formation at ± 45°. Fig. 15(a) shows the general layout of these specimens. Key features of the tests are as follows, namely:
- •
The partially threaded screw shown in Fig. 15(b) and spruce stub joists were used.
- •
Some specimens contained two X pairs of screws as shown, while others contained only one.
- •
Three interlayer thicknesses, namely 0 mm, 19 mm
Conclusions
From the study presented in this report the following key points emerge, namely:
- •
Three sets each of four double-shear Baubuche LVL-concrete specimens were tested to failure. Four M12 coach screws along a line were used as the connectors at each concrete-timber interface. Each set of four connection specimens was defined by the inclination angle of the screws to the interface. The screw inclinations used were 90° (C90 specimens), 45° with the screws in compression (C45C specimens) and 45° with
References (42)
- et al.
Experimental analysis of laterally loaded nailed timber-to-concrete connections
Constr Build Mater
(2009) - et al.
Tangential behaviour of nailed composite timber-concrete floor structures
Constr Build Mater
(2013) - et al.
Environmentally friendly high performance timber-concrete panel
Constr Build Mater
(2016) - et al.
Early-age performance of lag screw shear connections for glulam-lightweight concrete composite beams
Constr Build Mater
(2017) - et al.
Influence of test specimen on experimental characterisation of timber-concrete composite joints
Constr Build Mater
(2010) - et al.
Evaluation of perforated steel plates as connection in glulam-concrete composite structures
Constr Build Mater
(2012) Short-term performance of the HSB shear plate type connector for timber-concrete composite beams
Constr Build Mater
(2012)- et al.
Fatigue behaviour of timber-concrete composite connections and floor beams
Eng Struct
(2013) - et al.
Mechanical behaviour of connections for glubam-concrete composite beams
Constr Build Mater
(2017) - et al.
Experimental study of a composite wood-concrete beam with the INSA-Hilti new flexible shear connector
Constr Build Mater
(1999)
A new composite connector for timber-concrete composite structures
Constr Build Mater
Analysis of the behaviour of a bonded joint between laminated wood and ultra high performance fibre reinforced concrete using push-out test
Constr Build Mater
Adhesive-bonded timber-concrete composites – experimental and numerical investigation of hygrothermal effects
Eng Struct
Long-term performance of adhesively bonded timber-concrete composites
Int J Adhes Adhes
The effect of ductile connectors on the behaviour of timber-concrete composite beams
Eng Struct
Experimental study on the influence of concrete cracking on timber concrete composite beams
Eng Struct
Bending tests on timber-concrete composite members made of beech laminated veneer lumber with notched connection
Eng Struct
Insight into mechanics of externally indeterminate hardwood-concrete composite beams
Constr Build Mater
Indicative comparisons between bonded and dowelled hardwood studs for limecrete-spruce connections under load-unload-reload double-shear testing
Eng Struct
Long-term and collapse tests on a timber-concrete-composite beam with glued-in connection
Mater Struct
Experimental tests of notched and plate connectors for LVL-concrete composite beams
ASCE JSE
Cited by (30)
Timber-concrete composites (TCC) floors subjected to hogging moment
2024, Engineering StructuresHardwood and softwood timber - natural stone composite connections
2024, Construction and Building MaterialsExperimental and theoretical investigation on shear performances of glued-in perforated steel plate connections for prefabricated timber–concrete composite beams
2023, Case Studies in Construction MaterialsEffect of test configurations and loading protocols on performance of timber-concrete connectors
2023, Construction and Building Materials