Pull-Out Tests and Numerical Simulations …
47
To conduct these tests, a test stand was made, which consisted of 1 cm thick
resistance slab with 80 × 40 cm cross-sectional dimensions with a 10 × 10 cm cutout the hole in the center. This allowed free pulling of the bolt with any shape of a
broken concrete cone. Other parts of the stand were 4 or 6 pieces of an M20 screws
fixing the resistance slab with to bottom beam. The rope with a diameter of Ø13 mm
was ended on one side with a hook with a load capacity of 3 tons. On the other site
it was ended with an M12 eye, which was predicted to screw the anchor. In order
to attach the hook to the hydraulic cylinder, a double-sided thread was used: M22
mounted in the hydraulic cylinder and M24 ended with a threaded eye. Additionally,
the rollers between the resistance slab plate and the upper surface of a specimen has
been added to eliminate the secondary frictional forces caused by horizontal (shear)
force. In order to block a sample movement in a bottom area, the steel angle was
screwed to the stand beam, with a roller placed between it and the sample.
In the first stage of tests, anchors fixed at an angle of 90° in relation to the concrete
surface were tested (Fig. 1). They were both bonded anchors (Fig. 1a) and fixed in
complete solid concrete elements (Fig. 1b).
Additional tool plates with an opening were attached to the anchors and screwed
to help in displacements measurements. The first stage of the research proceeded
as follows: for the previously prepared auxiliary equipment, a vertical hydraulic
cylinder was installed in HYSDOZOK to obtain a pulling force, programmed in an
experimental study to achieve the final limit of 80 kN strength. The results reading
has been done in steps of 5 kN with a steady increase in force of 0.5 kN/s. The shear
force in the case of three-layer samples, has been established using a unit load of 1.4
kN corresponding to a 1 m
2 of the surface of the texture layer. For testing oblique
anchors in the 2nd stage of testing, the test stand (Fig. 2) was additionally equipped
with racks, made at the expected angles on which subsequent samples were placed.
Diagrams of test stands are shown in Figs. 3 and 4.
Samples with dimensions: 20 cm × 20 cm × 17 cm were tested, in the case of
three-layer samples, the cross section thickness of 17 cm is: 6 cm construction layer,
6 cm insulation layer, 5 cm textured layer and 30 cm × 20 cm × 6 cm for anchoring
at 30°. The conditions for supporting solid elements are shown in Figs. 5 and 6. The
material constants contained in Table 1 were adopted for testing and calculations.
3 Numerical Modeling of the Block Elements
Numerical modeling of block elements was performed in ANSYS Workbench. One
of the most complex processes was the proper design of contact zones in each model.
Modeling of block elements using the Finite Element Method was carried out with
all parameters and material constants (Table 1) taken into account during the experimental research. The assumption was to build the simplified, practical model suitable
for the experimental validation and proposition for a future general assessment of the
load capacity of multilayered walls connections. The concrete, steel and resin materials were described using bilinear relationships. Nonlinearity caused by cracking
47
To conduct these tests, a test stand was made, which consisted of 1 cm thick
resistance slab with 80 × 40 cm cross-sectional dimensions with a 10 × 10 cm cutout the hole in the center. This allowed free pulling of the bolt with any shape of a
broken concrete cone. Other parts of the stand were 4 or 6 pieces of an M20 screws
fixing the resistance slab with to bottom beam. The rope with a diameter of Ø13 mm
was ended on one side with a hook with a load capacity of 3 tons. On the other site
it was ended with an M12 eye, which was predicted to screw the anchor. In order
to attach the hook to the hydraulic cylinder, a double-sided thread was used: M22
mounted in the hydraulic cylinder and M24 ended with a threaded eye. Additionally,
the rollers between the resistance slab plate and the upper surface of a specimen has
been added to eliminate the secondary frictional forces caused by horizontal (shear)
force. In order to block a sample movement in a bottom area, the steel angle was
screwed to the stand beam, with a roller placed between it and the sample.
In the first stage of tests, anchors fixed at an angle of 90° in relation to the concrete
surface were tested (Fig. 1). They were both bonded anchors (Fig. 1a) and fixed in
complete solid concrete elements (Fig. 1b).
Additional tool plates with an opening were attached to the anchors and screwed
to help in displacements measurements. The first stage of the research proceeded
as follows: for the previously prepared auxiliary equipment, a vertical hydraulic
cylinder was installed in HYSDOZOK to obtain a pulling force, programmed in an
experimental study to achieve the final limit of 80 kN strength. The results reading
has been done in steps of 5 kN with a steady increase in force of 0.5 kN/s. The shear
force in the case of three-layer samples, has been established using a unit load of 1.4
kN corresponding to a 1 m
2 of the surface of the texture layer. For testing oblique
anchors in the 2nd stage of testing, the test stand (Fig. 2) was additionally equipped
with racks, made at the expected angles on which subsequent samples were placed.
Diagrams of test stands are shown in Figs. 3 and 4.
Samples with dimensions: 20 cm × 20 cm × 17 cm were tested, in the case of
three-layer samples, the cross section thickness of 17 cm is: 6 cm construction layer,
6 cm insulation layer, 5 cm textured layer and 30 cm × 20 cm × 6 cm for anchoring
at 30°. The conditions for supporting solid elements are shown in Figs. 5 and 6. The
material constants contained in Table 1 were adopted for testing and calculations.
3 Numerical Modeling of the Block Elements
Numerical modeling of block elements was performed in ANSYS Workbench. One
of the most complex processes was the proper design of contact zones in each model.
Modeling of block elements using the Finite Element Method was carried out with
all parameters and material constants (Table 1) taken into account during the experimental research. The assumption was to build the simplified, practical model suitable
for the experimental validation and proposition for a future general assessment of the
load capacity of multilayered walls connections. The concrete, steel and resin materials were described using bilinear relationships. Nonlinearity caused by cracking