contact@welldcrane.com
Analysis of the Functionality of Wall-Mounted Joints in High-Performance Jib Tower Cranes
Modern construction relies on tower crane technology for managing substantial loads. The safety of these systems depends on embedded crane tower components attached to wall joints. Understanding the functional mechanism of these high-performance jib tower crane systems requires a methodology for analyzing the force performance of the critical joint connections.
1 Methodology
1.1 Project background
In the present study, the joints of tower cranes used in engineering practice and the wall joints of high-performance jib tower cranes were taken as research objects. The parameter settings for the joint embedded parts are listed in Table 1, and the comparison between full-size pre-embedded parts and scaled embedded parts is shown in Figure 1.
1.1 Project background
In the present study, the joints of tower cranes used in engineering practice and the wall joints of high-performance jib tower cranes were taken as research objects. The parameter settings for the joint embedded parts are listed in Table 1, and the comparison between full-size pre-embedded parts and scaled embedded parts is shown in Figure 1.
Table 1. Parameter settings for embedded parts
Figure 1. Comparison between full-size pre-embedded parts and scaled embedded parts
1.2 Tower Crane Specimen Design and Fabrication
Jib tower cranes used in the construction of super high-rise buildings have heavy self-weight and large embedded parts. Their group anchor forms are often adopted, and the bearing capacities of their joints are strong. In laboratory conditions, full-size pre-embedded parts can hardly achieve monotonic static loading, which leads to structural damage. The scale model specimens with a scale ratio of 1:2 were used due to the limited research resource input and laboratory equipment conditions. Moreover, a single anchor slab was arranged separately at the four corners of embedded parts only. The embedded parts were fabricated according to the scale ratio using the actual construction site size as a reference. C30 concrete, grade 3 rebar, Q345C hot-rolled steel plate, similarity relations between model and prototype, and comparison of embedded parts are shown in Table 2.
Jib tower cranes used in the construction of super high-rise buildings have heavy self-weight and large embedded parts. Their group anchor forms are often adopted, and the bearing capacities of their joints are strong. In laboratory conditions, full-size pre-embedded parts can hardly achieve monotonic static loading, which leads to structural damage. The scale model specimens with a scale ratio of 1:2 were used due to the limited research resource input and laboratory equipment conditions. Moreover, a single anchor slab was arranged separately at the four corners of embedded parts only. The embedded parts were fabricated according to the scale ratio using the actual construction site size as a reference. C30 concrete, grade 3 rebar, Q345C hot-rolled steel plate, similarity relations between model and prototype, and comparison of embedded parts are shown in Table 2.
Table 2. Similarity relations between model and prototype
Figure 2. Specimen construction drawings
According to the Saint-Venant's Principle, the effect of local load influences the stress field distribution only within a certain range. After considering the economic efficiency and experimental rationality, the size of shear wall test surface was set three times the size of each side of the embedded part panel, whereas the thickness of the wall was set two times the minimum embedded depth of the embedded parts according to the actual embedded depth ratio. The wall reinforcement was installed according to the scale ratio with reference to the actual construction drawings and by using the same materials. The specimen fabrication was completed at the Chongqing Jiangbeizui International Finance Square (IFS) project office under the China Construction Second Engineering Bureau. The section and anchor slabs were set up, as shown in Figure 2, whereas the experimental facility is shown in Figure 3.
1.3 Tower Crane Experimental Facility and Loading Protocol
In the static pullout test, stepwise monotonic static loading was adopted. During the specimen installation, attention was paid to the geometric alignment. After alignment and setting out of the various test system components of the tower cranes, they were calibrated and installed with laser level instrument to ensure that the jack center, embedded part centerline, and hinge center turning hinge midpoint were in the same vertical plane. Afterward, preloading was performed. The role of preloading is to check whether all test instrumentations are functioning properly and whether the facility is reliable. We can determine whether the relationship between load and deformation is stable and whether the specimen and the bearing are in good contact through the data collected from the preloading. Therefore, a preloading is necessary prior to formal loading, and the size of preload should be 20% of the theoretical ultimate load. Then, various instrumentations were adjusted to normal state and unloaded and zeroed before formal loading.
In the static pullout test, stepwise monotonic static loading was adopted. During the specimen installation, attention was paid to the geometric alignment. After alignment and setting out of the various test system components of the tower cranes, they were calibrated and installed with laser level instrument to ensure that the jack center, embedded part centerline, and hinge center turning hinge midpoint were in the same vertical plane. Afterward, preloading was performed. The role of preloading is to check whether all test instrumentations are functioning properly and whether the facility is reliable. We can determine whether the relationship between load and deformation is stable and whether the specimen and the bearing are in good contact through the data collected from the preloading. Therefore, a preloading is necessary prior to formal loading, and the size of preload should be 20% of the theoretical ultimate load. Then, various instrumentations were adjusted to normal state and unloaded and zeroed before formal loading.
Figure 3. Experimental loading device
Figure 4. Schematic of loading protocol
Attention was paid to a few key joints, and the cracking point, yield point, limit point, and several drop points were captured. The differential was adjusted downward near the point locations, and each level of load was sustained for 1 min. When the load started to decrease to disable loading continuance, the specimen could be considered to reach an ultimate carrying capacity and already entered the unloading phase. The data also need to be collected during the unloading phase. The experimental setup is shown in Figure 3, whereas the monotonic stepwise loading is shown in Figure 4.
1.4 Tower Crane Finite Element Modeling
The specimen parameters were assigned, and the same sectional size, ratio of reinforcement, and boundary conditions were set up. The finite element models of various components were built in the ABAQUS pre-processing program as shown below.
The specimen parameters were assigned, and the same sectional size, ratio of reinforcement, and boundary conditions were set up. The finite element models of various components were built in the ABAQUS pre-processing program as shown below.
-
- Figure 5. Finite element models of the wall joints
-
2 Result Analysis and Discussion
2.1 Tower Crane Ultimate Pullout Capacity
In the static pullout test of the attached to the wall joints of tower crane, the ultimate capacity of embedded parts was used to compare and check the safety. The ultimate capacities of the three specimens in the test are shown in Table 3.
2.1 Tower Crane Ultimate Pullout Capacity
In the static pullout test of the attached to the wall joints of tower crane, the ultimate capacity of embedded parts was used to compare and check the safety. The ultimate capacities of the three specimens in the test are shown in Table 3.
Table 3. Ultimate capacities of specimens
The comparison of ultimate capacity between specimens clearly showed that the cracking load of structure is approximately 30%–40% of the ultimate load. Moreover, the general trend is the greater the ultimate value, the larger the cracking load. Except for the blank control specimen SJ165000N, the remaining two specimens were both controlled by the flexural capacity of shear walls due to the same failure pattern and adequate anchorage.
2.2 Tower Crane Load-displacement curves
The changes in displacement varied between the high and low planes of the embedded parts. The displacements at two measuring points of each specimen under various levels of load were used to plot the scatter curves as shown below.
The changes in displacement varied between the high and low planes of the embedded parts. The displacements at two measuring points of each specimen under various levels of load were used to plot the scatter curves as shown below.
-
- Figure 6. Load-displacement curves of specimens
-
As shown, the variation trends of increase in displacement with load were basically consistent, all of which were small in the former section and large in the latter section. The changes in the slope of curves showed that the curves were approximately steep before concrete cracking. Displacement changes were minimal at increased load, and the displacements of the upper and lower edges were close to each other. The differential of stepwise loading was not subdivided finely under the test conditions. Therefore, the stiffness degradation at the initiation of member cracking was not clearly observable in the curves. After cracking, the variation of displacement accelerated, and the slopes of curves began to drop as the load continued to increase, indicating that the stiffness of members degraded markedly after cracking. Moreover, the greater the load, the denser the fracture distribution, the larger the crack width, and the severer the stiffness degradation of members. When the ultimate load was approached, the load at a certain level of subdivision was sustained, yet the displacement remains continuously increased, and platform damage occurred. Afterward, the bearing capacity decreased, the descending sections appeared in the curves, and the structure underwent damage.
2.3 Tower Crane Finite Element Model Verification
The simulation calculations were extracted from the ABAQUS visualized post-processing module and compared with the test results of the corresponding specimens for mutual verification of the authenticity of the experimental data and the correctness of the finite element models. For the attached to the wall joints models of tower cranes that were loaded stepwise statically, the primary concern was the damage pattern and the ultimate capacity of the joints. The stress and displacement cloud diagrams of the specimens were compared, as shown below.
The simulation calculations were extracted from the ABAQUS visualized post-processing module and compared with the test results of the corresponding specimens for mutual verification of the authenticity of the experimental data and the correctness of the finite element models. For the attached to the wall joints models of tower cranes that were loaded stepwise statically, the primary concern was the damage pattern and the ultimate capacity of the joints. The stress and displacement cloud diagrams of the specimens were compared, as shown below.
-
- Figure 7. Comparative analysis diagrams for specimen SJ165025N
-
-
- Figure 8. Comparative analysis diagrams for specimen SJ204025N
-
-
- Figure 9. Comparative analysis diagrams for specimen SJ165000N
-
In ABAQUS, the solid elements were under bidirectional loading. The comparisons of the main stress cloud and main strain cloud show that the principal tensile stress at the lower part of shear wall was large during damage, which was extended to both sdes of concrete members. As shown from the stress cloud of the embedded parts themselves, the stress of anchor slabs was smaller than their yield strength. Only elastic deformation occurred under the simulated external load, whereas the ending slabs did not deform obviously. This finding is in agreement with the test measurements. The stress cloud of specimen SJ165000N shows that the embedded part stress had a small influence range, the displacement was also only concentrated in the locations of four anchor slabs, and the bearing capacity was determined by the defined interface bonded slipping constitutive relation. For the remaining two specimens, the bearing capacity was controlled by the strength of concrete due to adequate anchorage. The principal stress concentration locations shown in the stress clouds corresponded to the locations of the initial cracking and major cracks of the test specimens, which were the concentrated damage locations of members. The comparison results between the simulated and tested bearing capacity values of specimens are listed in Table 4.
Table 4. Comparison of specimen ultimate capacity
As shown in Table 4, the damage patterns from specimen simulation were the same as the experimental results. For specimen SJ165000N, the embedded parts slipped because of insufficient anchorage, whereas the rests were all shear wall damages. The ultimate capacity values from simulation and test were close, showing a maximum error of 9.38%, and the simulation value was less than the test value. According to the analysis of error causes, the relative errors between the two were caused by a certain difference between the constitutive model and bond–slip constitutive model defined by finite element method from the real constitutive model of material, as well as the discreteness of the mechanical properties of concrete itself. The negative error of specimen SJ165000N may be due to the occurrence of slippage displacement of embedded parts. The finite element-based bonded slipping constitutive relation always exists during the loading process and continued to bear the pullout load. By contrast, during the test, the chemical adsorption function of anchor slabs and concrete failed once the slippage displacement of embedded part occurred, and the pullout load was borne only by the frictional and mechanical build-in forces. Thus, the test values were less than the simulation values, showing a difference of −5.85%. From the perspective of the overall force performance evaluation of embedded parts, such an error rate was still within an acceptable range, the test data were credible, and the numerical models were correct, which can be used for the parametric analysis of factors influencing the bearing capacity of embedded parts.
4.3 Comparison of model calculations with test results
The load-displacement data were extracted from the ABAQUS numerical simulations and compared with the experimental data to mutually verify the correctness and reliability of the data. The three specimens were cross compared, as shown in Figures 10-12.
The load-displacement data were extracted from the ABAQUS numerical simulations and compared with the experimental data to mutually verify the correctness and reliability of the data. The three specimens were cross compared, as shown in Figures 10-12.
- Figure 10. Comparison of load-displacement curves for specimen SJ204025N
- Figure 11. Comparison of load–displacement curves for specimen SJ165000N
- Figure 12. Comparison of load–displacement curves for specimen SJ165000Y
As shown in Figure 10, the specimen is a reinforced embedded part with no pegs on the end slab. The test and simulation values of ultimate capacity differed by 10.6%, which is acceptable. The bearing capacity is always controlled by the strength of shear wall concrete because of the same damage pattern between the two. However, the two exhibited a large difference in the displacement curve. In the early loading stage, the elastic modulus of concrete defined in the simulation calculation can rather truly reflect its actual force state under the test loads because the concrete is still in the elastic stage without crack generation and the structural stiffness does not undergo significant change. The displacement curves of the two were well fitted and the slopes were basically identical. However, as the load continued to increase, cracks appeared, broadened, and increased, and the displacement curves shifted and enlarged gradually. In the later stage of loading, the test displacement value was markedly larger than the simulation value due to the continuously increasing crack width, which is approximately three times that of the simulation value when near the limit, showing a large error. Compared to the simulation curve, the test curve exhibited larger platform damage, stronger load sustainability, and better ductility. Nonetheless, from the perspective of the variation regularity of displacement, the two were basically identical in trends.
As shown in Figure 11, the load–displacement curves of specimen SJ165000N basically coincide in the early loading elastic stage, with only small difference in stiffness. However, with the increase of load, the concrete cracked, stiffness degraded, and test and simulation values of displacement showed a rapid difference, which ever enlarged. The cause of difference in displacement at the upper measuring point between the two was because the cracking discontinuity of concrete cannot be preferably simulated with the concrete constitutive model selected in the simulation calculation. Moreover, on the contact surface between the anchor slab and the concrete, the defined bonded slipping constitutive relation differed somewhat from the experimental one. Moreover, the discreteness of concrete material strength itself and the inhomogeneity of medium were the causes of such discrepancy. The simulation values of displacement at the upper and lower measuring points were basically the same, which was quite different from the test case. The effects of the shear deformation of concrete along the axial direction of shear wall, the cracking of concrete to the limit state, and the width of major crack on the increase in the displacement of
As shown in Figure 12, the specimen SJ165000Y is an anchoring reinforced with pegs. In the ABAQUS finite element simulation, the anchoring reinforcing effect of pegs was realized by adding spring elements and defining the stiffness. The two types of values were rather close regarding the ultimate capacity, with a difference of 11.9%, where the test value was higher than the simulation value. In the early loading stage, the two types of displacement values were relatively close, the curves were well fit, and the displacements of embedded parts at the upper and lower measuring points did not differ much. In the later loading period, the differences between the two values began to increase. The curve for simulation value rose basically along the initial slope to the damage platform of structure, the structure cracked, and the influence of stiffness degradation was small. For the test values, the curve gradually flattened out in the later loading stage, the displacement increased rapidly, and the difference between the embedded part displacements at the upper and lower edges also increased gradually, reflecting that the concrete cracking and the embedded part rotation around the geometric center were greatly influential to the displacement growth. The total displacement was nearly twice the simulation value. Clearly, the simulation value was accurate in calculating the bearing capacity, because the damage patterns of the two were basically identical, both of which are controlled by the strength of concrete. By contrast, the difference in displacement was rather large, mainly because the diffusion crack model incorporated in ABAQUS cannot well simulate the actual cracking pattern of concrete. In the later stage of loading, the cracks widened. Their intensive internal distribution greatly affected the pullout displacement of embedded parts, thus resulting in a large displacement difference. Compared with the SJ165000N without pegs, the simulation value of bearing capacity defined by the spring stiffness increased by 15.4%, which reflected the contribution of pegs to the anchorage reinforcement. In addition, the bearing capacity must be calculated by using the simulation method of defining spring elements.
As shown in Figure 11, the load–displacement curves of specimen SJ165000N basically coincide in the early loading elastic stage, with only small difference in stiffness. However, with the increase of load, the concrete cracked, stiffness degraded, and test and simulation values of displacement showed a rapid difference, which ever enlarged. The cause of difference in displacement at the upper measuring point between the two was because the cracking discontinuity of concrete cannot be preferably simulated with the concrete constitutive model selected in the simulation calculation. Moreover, on the contact surface between the anchor slab and the concrete, the defined bonded slipping constitutive relation differed somewhat from the experimental one. Moreover, the discreteness of concrete material strength itself and the inhomogeneity of medium were the causes of such discrepancy. The simulation values of displacement at the upper and lower measuring points were basically the same, which was quite different from the test case. The effects of the shear deformation of concrete along the axial direction of shear wall, the cracking of concrete to the limit state, and the width of major crack on the increase in the displacement of
As shown in Figure 12, the specimen SJ165000Y is an anchoring reinforced with pegs. In the ABAQUS finite element simulation, the anchoring reinforcing effect of pegs was realized by adding spring elements and defining the stiffness. The two types of values were rather close regarding the ultimate capacity, with a difference of 11.9%, where the test value was higher than the simulation value. In the early loading stage, the two types of displacement values were relatively close, the curves were well fit, and the displacements of embedded parts at the upper and lower measuring points did not differ much. In the later loading period, the differences between the two values began to increase. The curve for simulation value rose basically along the initial slope to the damage platform of structure, the structure cracked, and the influence of stiffness degradation was small. For the test values, the curve gradually flattened out in the later loading stage, the displacement increased rapidly, and the difference between the embedded part displacements at the upper and lower edges also increased gradually, reflecting that the concrete cracking and the embedded part rotation around the geometric center were greatly influential to the displacement growth. The total displacement was nearly twice the simulation value. Clearly, the simulation value was accurate in calculating the bearing capacity, because the damage patterns of the two were basically identical, both of which are controlled by the strength of concrete. By contrast, the difference in displacement was rather large, mainly because the diffusion crack model incorporated in ABAQUS cannot well simulate the actual cracking pattern of concrete. In the later stage of loading, the cracks widened. Their intensive internal distribution greatly affected the pullout displacement of embedded parts, thus resulting in a large displacement difference. Compared with the SJ165000N without pegs, the simulation value of bearing capacity defined by the spring stiffness increased by 15.4%, which reflected the contribution of pegs to the anchorage reinforcement. In addition, the bearing capacity must be calculated by using the simulation method of defining spring elements.
Conclusion
The following conclusions could be drawn:
1) The embedded parts undergo two types of damages depending on the installation form: pullout damage of parts due to incomplete anchoring and bending fracture damage of shear walls under sufficient anchoring. The difference in bearing capacity is large between the two types.
(2) The load–displacement curves of embedded parts show obvious two-stage variation. In the early loading stage, the curves can almost be fitted to a straight line, the displacements are very small, and the structures are within the linear elastic range. After continuance of loading and cracking, the stiffness degradation occurs, the slopes of curves decrease, and the displacement growth accelerates. The later the stage, the wider the crack distribution, and the severer the stiffness degradation. The embedded parts undergo combined deformation of the tension, flexural, and shear under the action of diagonal tension. The local rotation tendency of embedded parts around the geometric center results in the significantly less displacement and correspondingly smaller strain of the upper anchor slabs than the lower anchor slabs.
(3) Anchoring reinforcement has no obvious effect on the displacement. However, if the anchorage is not reinforced, then the embedded parts will be prone to slip and pullout damage and the anchorage stiffness will be small.
(4) The attached to the wall joints of tower cranes bear the compound stress effect of the tension, flexural, and shear, showing a complex local stress distribution. The possible damage patterns of the embedded parts include the cross-sectional area of the tensile strength of the anchor plate, weld connection of the anchor plate, overall punching damage considering the group anchorage effect, and bending damage of narrow limbs outside the wall, among others.
The following conclusions could be drawn:
1) The embedded parts undergo two types of damages depending on the installation form: pullout damage of parts due to incomplete anchoring and bending fracture damage of shear walls under sufficient anchoring. The difference in bearing capacity is large between the two types.
(2) The load–displacement curves of embedded parts show obvious two-stage variation. In the early loading stage, the curves can almost be fitted to a straight line, the displacements are very small, and the structures are within the linear elastic range. After continuance of loading and cracking, the stiffness degradation occurs, the slopes of curves decrease, and the displacement growth accelerates. The later the stage, the wider the crack distribution, and the severer the stiffness degradation. The embedded parts undergo combined deformation of the tension, flexural, and shear under the action of diagonal tension. The local rotation tendency of embedded parts around the geometric center results in the significantly less displacement and correspondingly smaller strain of the upper anchor slabs than the lower anchor slabs.
(3) Anchoring reinforcement has no obvious effect on the displacement. However, if the anchorage is not reinforced, then the embedded parts will be prone to slip and pullout damage and the anchorage stiffness will be small.
(4) The attached to the wall joints of tower cranes bear the compound stress effect of the tension, flexural, and shear, showing a complex local stress distribution. The possible damage patterns of the embedded parts include the cross-sectional area of the tensile strength of the anchor plate, weld connection of the anchor plate, overall punching damage considering the group anchorage effect, and bending damage of narrow limbs outside the wall, among others.
Authors
- Gang YaoChina
Key Laboratory of New Technology for Construction of Cities in Mountain Area - Chengcheng XuChina
Key Laboratory of New Technology for Construction of Cities in Mountain Area - Yang YangChina
Key Laboratory of New Technology for Construction of Cities in Mountain Area - Mingpu WangChina
Key Laboratory of New Technology for Construction of Cities in Mountain Area - Mao ZhangChina
China Construction Second Engineering Bureau Ltd - Ayad ThabeetSlovak Republic
Technical University of Kosice
About WHCM
Welld Harbin Construction Machinery (also known as WELLD) aims to elevate its position as a premier Chinese сonstruction machinery enterprise by pursuing internationalization and enhancing its industry chain, alongside establishing efficient operational, research and management systems. Through these strategies, the company seeks to deliver high-quality construction equipment to customers globally, reinforcing its commitment to excellence in the engineering machinery sector.
We are a leading tower crane manufacturer provide quality and value to our customers. We like what we do.