Shear Strength Tests in a Lightweight Floor System with Cement and Polyurethane Adhesives

The article presents static shear tests of adhesives connecting various types of lightweight floor system (LFS) composites with a heating coil. This method is approved for use, and its advantage is the reproduction of the real working conditions of a light floor. Preliminary tests were carried out to determine the point shear strength of selected adhesives and the target ones, using a strain gauge method of deformation measurements. Three types of adhesives used in the lightweight heated floor system were researched-cement deformable, type C2S1, C2S2, and polyurethane BondT8. In addition to determining the maximum shear stress, deformation and approximate displacement of adhesives in this type of composites, the Kirchoff’s and Young’s modulus were also calculated. The method used did not allow determining Poisson’s ratio. Thanks to the three-sided model of sticking the tiles outside its outline, it was possible to determine what the maximum shear stress occurs on the edges of the floor. This is important, due to the highest LFS stresses occurring in these places, confirmed by the analyses contained in the literature cited in this article.


Introduction
The purpose of testing the samples made in the form of layered In addition to checking the shear strength of the adhesives, it was planned to determine the sizes of their Kirchoff G, Young E modulus and Poisson's ratio. A static shear test was used to define these.
In the case of a sample connected with polyurethane glue, aluminum foil which was 45μm thick was glued on the XPS insulation, which is one of the layers of the LFS model in some shear strength tests were carried out without the installation of strain gauges. After that, it was decided to reduce the surface of the floor tiles, with the same size of the insulating substrate. High sensitivity strain gauges were used in the target measurements.

Description of the Measurement Stand and Research Methodology
In the shear strength test, floor tiles glued to the thermal insulation on a surface of 10 x 20cm were initially used, using after one sample of each from the three types of adhesives to check the initial shear stress σ s1 . The tile was glued pointwise only with one-sided distribution of the adhesive outside its boundary, as shown in Figure 1

c)
BondT8 polyurethane glued to aluminum foil.
In the shear strength test, Tenmex type TFs-15 strain gauges were used, with a measuring base length of 15mm, and strain gauge constant k = 2.19 ± 0.5%, R = 120 ± 0.2%. Strain gauges were carried out in a half-bridge system with 12mV/V sensitivity and sampling. Two active strain gauges were glued to each forceexposed sample, one longitudinally to the shear force and the other transversely. Additionally, two passive strain gauges fulfilling the role of compensation were connected to the research system.
The whole was connected to a measurement recorder, 4-channel,

Analysis of Results
As the results of the tests show, the maximum shear stress of adhesives fixed in three directions compared to spot-fixing is higher and is 130% (C2S1), 150% (C2S2), and 15% (BondT8). The average    It seems that the right direction during testing of cast samples is to reduce them to the dimensions found in real joints, which will also reduce the defects occurring inside them during preparation.
In this case, for a flat state of stress, we assume the following dependencies of normal to tangents stress -σ 1 = τ i σ 2 = -τ. What deformations and stresses look like is shown in Figure 14. A B There was a flattening of the square occurred here by the angular deformation γ, hence in Figure 14, the following equation can be adopted: if we assume that γ has a small value, then tg(γ/2) ≈ γ/2), i.e.
After comparing the formula (1) and (2)     While using the method reflecting the real working conditions of the glued composite, it was assumed that the glued elements are subject to deformation only due to shear stress τ, and their value is the same along the entire length of the glue connection as described by Maćkowiak and others [4]. In fact, the glued elements are subject to both tangential and normal stress, as shown in Figure 18. Such an assumption according to [4] in glued steel materials leads to an error of approx. 6.5% when calculating the displacement of d s , and the maximum stress values are higher than the assumed averages of 1.6-3.2%. In addition, deformation and stress disproportions are greater than average values when the stiffness of the materials to be glued decreases and the connection length increases. Figure 18: Section of stress and deformation while cutting the adhesive layer, from [4].
According to the standard [8], the Kirchhoff's modulus of overlapped and stretched elements can be determined, as in Figure 19, where the measurement of force and deformation is carried out until the destruction of the prepared samples. Figure   19 shows the stress and strain directions of the adhesive layer.
Before determining the necessary strength parameters, the R H proportionality limit should be determined from the diagrams of tangential stress versus strain. R H is the boundary value of the elastic phase stress below which we determine Kirchoff's modulus, Young's modulus and Poisson's ratio. The ratio of stress to elastic deformation has a constant value and decreases with plastic deformation. Performing shear or torsion tests determine the modulus of transverse elasticity, called the Kirchhoff G module.
Its value described by Kłysza [9] depends on the type of material, temperature and pressure, and does not depend on the speed of deformation. It is about 2-3 times smaller than Young's modulus.
Using isotropic materials, it can be calculated from the formula (4):  (5) and (6) are given in (Table 1). In addition, the table provides information on the maximum stress σ s max and the related longitudinal deformation Ɛ m and the maximum approximate displacements of the selected two samples during the experiment.  It is recommended to perform experimental tests on cast samples, followed by numerical calculations using e.g. the finite element method (FEM). This will confirm whether the data obtained are consistent with the results of lightweight floor system strength experiments using the same type of adhesive.